\:T 1-\
Tl00 i t 5
/VII ;(1"11
i3~

THE BULL CREEK SITE, 9MEl MUSCOGEE COUNTY, GEORGIA
R. Jerald Ledbetter

Georgia Department of Transportation Office of Environment/Location Atlanta, Georgia
OCCASIONAL PAPERS IN CULTURAL RESOURCE MANAGEMENT #9 1997

Georgia Department of Transportation Occasional Papers in Cultural Resource Management

The Georgia Department of Transportation's (DOT) Occasional Papers in Cultural Resource Management series consists of archaeological research sponsored by the Georgia DOT. These reports have been produced by Georgia DOT in-house archaeological staff and by archaeological consultants under contract with the Georgia DOT. Each report within the series presents information about specific archaeological sites considered eligible for the National Register of Historic Places that would be affected by transportation projects.

Each report addresses research questions included in the Scope of Work for each project and the conclusions and interpretations contained therein reflect the theoretical orientation, background, and assorted biases of the authors. Each manuscript has been prepared as a result of a contract with Georgia DOT. The reports are distributed by the Office of Environment/Location, Georgia Department of Transportation.

For a copy of any or all of the reports, please indicate the specific report; there may be a minimal charge for copying if the report is out of print.

Georgia Department of Transportation Occasional Papers in Cultural Resource Management

No. I - Archaeological Investigations at 9CK(DOT)7, Cherokee County, Georgia; by WilliamR. Bowen (1982).

No. 2 - Cagle Site Report: Archaic and Early Woodland Period Manifestations in the North Georgia Piedmont; by Morgan R. Crook, Jr. (1984).

No.3 - Lowe Site Report: A Contribution to Archaeology of the Georgia Coastal Plain; by Morgan R. Crook, Jr. (1987).

No.4 - Rush: An Early Woodland Period Site in Northwest Georgia; by W. Dean Wood and R. Jerald Ledbetter (1990).

No.5 - A Few Visits in Prehistory: Data Recovery at 9RH18, Randolph County, Georgia; by Christopher T. Espenshade (1993).

No.6 - The Pig Pen Site: Archeological Investigations at 9RI158, Richmond County, Georgia; by R. Jerald Ledbetter (1988).

No.7 -

Data Recovery at Lovers Lane, Phinizy Swamp and the Old Dike Sites, Bobby Jones Expressway Extension Corridor, Augusta, Georgia; by Daniel T. Elliott, R. Jerald Ledbetter, and Elizabeth A. Gordon (1994).

No.8 - The Victory Drive Site, 9ME50, Muscogee County, Georgia; by R. Jerald Ledbetter (1997).

No.9 - The Bull Creek Site, 9MEI, Muscogee County, Georgia, by R. Jerald Ledbetter (1997).

No. 10 - An Archeological Survey ofthe Shoulderbone Tract, Hancock County, Georgia; by Thomas J. Pluckhahn (1997).

Cover illustration: Tobacco Pipe from WPA Excavations of the Mississippian Period Bull Creek Cemetery.

Preface
The passage of the Intermodal Surface Transportation Efficiency Act (ISTEA) in 1991 required states to set aside at least ten percent of Federal funds for transportation enhancement projects. In particular, the Transportation Enhancement Activities (TEA) provision of ISTEA provides the opportunity to fund a range of non-traditional projects, including historic preservation and conservation activities with a direct relationship to the intermodal transportation system. Within the state of Georgia, these projects range from parkways and bikeways to restoration of historic railroad depots, lighthouses, and covered bridges and involve a variety of sponsors, including cities, counties, and preservation groups.
One category of projects eligible for Transportation Enhancement funds is archaeological planning and research. The present study is a synthesis of previous archaeological investigations of the Bull Creek (9ME1) site by Southeastern Archeological Services, Inc. for Columbus/Muscogee County Consolidated Government. A portion of this National Register eligible archaeological resource lies within the area of potential effect of the proposed Columbus Riverwalk project. This report was prepared in compliance with Section 106 of the National Historic Preservation Act of 1966, as amended, as a measure to mitigate the proposed project's adverse effect to the archaeological resource.
The Columbus Riverwalk project is part of Columbus/Muscogee County Consolidated Government's Alternative Transportation System Plan, a portion of which is funded by ISTEA. This project illustrates the investment by state transportation agencies, the Federal government (Federal Highway Administration), and local governments in respecting and preserving the past in building for the future. The Georgia Department of Transportation is pleased to publish this archaeological data recovery report as the ninth in its Occasional Papers in Cultural Resource Management Series.
Elizabeth C. Shirk Staff Archaeologist Georgia Department of Transportation Atlanta, Georgia August, 1997

THIS PAGE LEFT BLANK

Abstract
This report is a compilation and synthesis of previous investigations at the Bull Creek site, 9Me1, located near Columbus in Muscogee County, Georgia. The Bull Creek site is the type site for the Late Mississippian period Bull Creek phase. The site was extensively excavated during the WPA period under the direction of Dr. A.R. Kelly. The WPA excavations of 1936 and 1937 encountered well preserved structural remains and an extensive cemetery area. Subsequent excavations of the site were conducted by Kelly in 1950, David Chase in the 1950s, and Frank Schnell, Jr. in 1959. There have also been other minor excavations on the site. Unfortunately, none of these excavations have been adequately reported.
The Bull Creek site has been severely damaged by land clearing and filling related to commercial development and most of the site has been destroyed. The area formerly containing the Bull Creek site lay within the route of the proposed City of Columbus Riverwalk Project. An intensive archeological survey of the portion of the Riverwalk route which crossed the Bull Creek site was conducted by Southeastern Archeological Services, Inc. in November 1993. As a result of that survey, a determination was made that only a small part of the site remained preserved. A route for the riverwalk was designed to avoid the preserved site area.
The Bull Creek site and a bordering site named the Victory Drive site, 9Me50, were recommended eligible to the National Register of Historic Places. While portions of both sites remain that can provide significant new information, the Bull Creek site was felt to have historical importance because of the extensive WPA-era excavations and its role in defining the Bull Creek phase. Because the continued lack of reporting of previous excavations at the National Register-eligible Bull Creek site constitutes an adverse impact to the site, the logical course of action to mitigate against this adverse impact was to insure that existing notes and records would be assembled into an accessible, final report. This document is the final, comprehensive report.
Efforts to accomplish these tasks met with varying degrees of success. Records and collections of different investigations varied in quality and completeness. However, some information was available for each of the investigations. As a result, we have been successful in assembling a volume that details the history of past investigations. Perhaps more importantly, the volume presents the ideas and site interpretations of past investigators gleaned from field notes and unpublished manuscripts. Reporting of past investigations together with re-analysis of existing artifact collections, has allowed the relevance of the Bull Creek site to be displayed and its role in the Mississippian period of the Southeast to be addressed.
iii

THIS PAGE LEn BLAHK

Acknowledgements
This report is predominantly a compilation of unpublished research relating to a very important archeological site. The Bull Creek site is the location of historically important WPA-era investigations and the type site for the Late Mississippian period Bull Creek phase, yet it has not been well reported. As such, a thorough accounting of all past investigations was necessary. This report should be viewed as recognition of the accomplishments of past researchers Isabel Patterson, A.R. Kelly, Frank Schnell, Jr., and others who salvaged significant information from the site. The Columbus Riverwalk Project provided the opportunity to investigate the site in the field and allowed a re-examination of the documents and collections from previous investigations conducted over much of the twentieth century. Individuals associated with the Columbus Water Works and the Columbus Consolidated Government are to be commended for providing the funds necessary to complete this study. Particular appreciation is directed to Water Works president, Mr. Billy Turner, and Mr. Cliff Arnett, vice-president of Systems Services and Water Resources Operations, who were very supportive throughout the project. David Ashley of Jordan, Jones, and Goulding, Inc. provided valuable coordination and logistical support.
This project was made possible, in great measure, through the diligence of past and present archeologists associated with the Columbus Museum, formerly the Columbus Museum of Arts and Sciences. That institution has been responsible for maintaining manuscripts, records, and collections from most work conducted on the site. The current archeologist for the museum, Mr. Frank Schnell, Jr., requires particular recognition. Frank is the authority on the Bull Creek site and the phases relating to the Mississippian Period of the fall line region of the Chattahoochee River Valley. He freely shared that knowledge as well as the resources of the museum.
A number of archeologists provided comments and assistance in preparing this report. Besides the contributing authors of this report, appreciation is expressed to Jack Walker, Richard Vernon, Dan Elliott, Sammy Smith, Mark Williams and David Hally. This report includes redrafted drawings from old manuscripts and some new graphics, most of which were prepared by Gisela Weis-Gresham. Michelle Walker transcribed the old manuscripts and assisted in editing the report. Tom Gresham assisted in editing the report and Johanna Minich was responsible for much of the report assembly.
v

TIUS PAGE LEn BLANK

Table of Contents

Preface

i

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. iii

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. v

List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ix

List of Tables

xv

Chapter 1 Introduction and Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 1

Setting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 3

Culture History

7

Chapter 2 Early Investigations Background to the WPA Excavations

19

Chapter 3 WPA Excavations of 1936 and 1937 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 33

Lester's Village Excavations .,

" 42

Cemetery Excavations

. . . . . . . . . . . . . . . . . . . . . .. 53

Chapter 4 Kelly's 1950 Investigations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 77

Kelly's Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 84

Results of Stratigraphic Test Pitting and Trenching of the Bull Creek Site. . . . . . . . .. 84

Recommendations for Treatment of Site and Development

87

Chapter 5 Schnell's 1959 Excavations

" 91

Chapter 6 Minor Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 103

1945 Survey by Frank and Barbara Fenenga ..... . . . . . . . . . . . . . . . . . . . . . .. 103

1950s Investigations of David Chase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 105

1960 Assessment by Huscher as Related to Construction of the Walter F. George

Reservoir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 109

1960 and 1969 Investigations at the Go-Kart and Dolly Madison locales.

110

1960 Investigations by Tyler and Jackson . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 117

1970s Developments

. . . . .. 118

1981 Excavations by Jacky Fuller. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 118

Chapter 7 1993-1994 Investigations

125

Chapter 8 Bull Creek Site Material Culture

143

Bull Creek Phase Ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 143

Primary Pottery Attributes

149

Bull Creek Rim Attributes

161

Vessel Form Attributes

169

Summary of Pottery Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 180

Other Ceramic Objects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 191

Shell and Bone Objects

192

Stone Objects

192

Modem Artifacts

194

vii

Chapter 9 Summary and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 195

The Bull Creek Village

198

The Bull Creek Cemetery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 200

The Bull Creek Phase from a Regional Perspective . . . . . . . . . . . . . . . . . . . . . . .. 207

References Cited

211

Appendix A Transcription of Lester's Field Notes, Bull Creek Site, 1936 Appendix B WPA Coding Sheets Appendix C Bull Creek Vessel Form Data

viii

List of Figures

Figure 1.
Figure 2.
Figure 3. Figure 4. Figure 5. Figure 6. Figure 7.
Figure 8. Figure 9.
Figure 10.
Figure 11. Figure 12.
Figure 13. Figure 14. Figure 15. Figure 16. Figure 17. Figure 18.
Figure 19. Figure 20. Figure 21. Figure 22. Figure 23. Figure 24. Figure 25. Figure 26.
Figure 27.
Figure 28.
Figure 29. Figure 30. Figure 31. Figure 32. Figure 33. Figure 34. Figure 35. Figure 36. Figure 37.

Location of Columbus riverwalk-related project areas surveyed by SAS from 1991

and 1994.

1

Location of known archeological sites between the South Commons of Columbus

and Fort Benning

2

Location map of Columbus and the Bull Creek site.

3

Portion of the 1922 Soil Survey Map of Muscogee County

5

Locations of sites referenced in text. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 8

Surveyor's plat of land lot containing the Bull Creek site. . . . . . . . . . . . . . . . .. 14

Portion of Sturges' 1818 map of Georgia showing the settlement of Marshalls on

south side of Bull Creek. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15

Portion of Patterson's (1950) map of the Bull Creek area . . . . . . . . . . . . . . . .. 16

1936 photograph of the Bull Creek site prior to excavation showing the abandoned

corn fields (NPS photograph 2368). . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 17

Portion of a 1938 Soil Conservation Service aerial photograph showing the area of

the Bull Creek site

18

Map of sites found by Mulvihill in 1924. The Bull Creek Site is site 1

19

Photograph of Margaret Ashley and Frank Schnell, Sr., taken during their work in

western Georgia in 1928. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 20

Points of reference described in Patterson letter dated March 9, 1936

24

Ca 1936 photograph of individuals associated with federal relief archeology

25

Close-up of previous photograph showing view of Frank Lester.

26

Drawing of fluted point from Bull Creek (NPS files).

29

1936 photograph of Isabel Patterson at the location of House Site 1 at Bull Creek.. 31

1936 photograph ofIsabel Patterson at House Site 1, view to south toward Bull

Creek Cemetery

30

1936 photograph of Isabel Patterson in borrow pit near House Site 1

30

Redrafted copy of Lester's 1936 excavation plan map of the Bull Creek.

34

Field photograph of excavations in the village area (NPS photograph 2367).

35

Two field photographs of village area excavations

36

Two field photographs of the cemetery area excavations

37

Field photograph of initial trench excavation in the cemetery area at Bull Creek . .. 38

Cemetery map of the Bull Creek site based on Frank Lester's field records. . . . . . 39

Circa 1940 photograph of laboratory workers at Macon cataloging artifacts from

WPA excavations

41

Redrafted sketch map of a portion of the Bull Creek site showing the area of "house

midden"

42

Field photograph of railroad cut in village area also showing road on borrow pit at

right and staked excavations on high ground

43

Field photograph of village excavation and pedestaled grid stakes

44

Photograph of "brickett" material from Structure 1

44

Field photograph showing fired clay covering floor of Structure 1

45

Field photograph of the raised clay "curb" within Structure 1

45

Floor plan of Lester's Structure 1

46

Photographs of two vessel fragments from trash pit adjacent to Structure 1

47

Plan of a portion of Structure 1 showing the location of urn and calcined bone

48

Field photograph of possible burial urn from Structure 1

48

Field photograph of three smashed pots on floor of Structure 2

49

ix

Figure 38. Field photograph showing workers clearing the floor of Structure 2 and the locations

of shattered vessels

50

Figure 39. Modern photograph of restored vessel from Structure 2.

51

Figure 40. Photograph of restored vessel from Structure 2

52

Figure 41. Reconstructed plan of Lester's village excavation based upon field note entries. . .. 53

Figure 42. Field photograph showing excavation of the cemetery, view to the north.

53

Figure 43. Four examples of burial field drawings from Bull Creek cemetery

54

Figure 44. Enlarged view of Lester's Bull Creek base map showing layout of the cemetery

grid.

55

Figure 45. Field photograph of the "more expert N.Y.A. workers" at the cemetery

56

Figure 46. Reconstructed map of the eastern part of the cemetery containing Burials 1-33 . . .. 57

Figure 47. Redrafted profile map of the cemetery area

58

Figure 48. Field photograph of the Burial 3 and first dog effigy vessel

59

Figure 49. Series of 1936 NPS photographs showing the removal of Burial 3 for exhibition. .. 60

Figure 50. Field photograph of Burial 5

61

Figure 51. Field photograph of Burials 6 in foreground and Burial 7 in background upper

right

62

Figure 52. Modern photograph of second dog pot.

62

Figure 53. Field photograph of Burial 8. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 63

Figure 54. Field photograph of Burial 12

. . . . . . . . . . . . . . . . . . . . . . . . .. 64

Figure 55. Field photograph of Burial 14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 65

Figure 56. Field photograph of Burial 16 and third dog pot

66

Figure 57. Photograph of tobacco pipe from Burial 17

66

Figure 58. Field map of Burial 18.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 67

Figure 59. Photograph of small bowl found with Burial 18

67

Figure 60. Field drawing of tobacco pipe from Burial 18

68

Figure 61. Field photograph of Burial 19

. . . . . . . . . . . . . . . . . . . . . . . . . .. 68

Figure 62. Field photograph of Burial 20

. . . . . . . . . . . . . . . . . . . . . . . . .. 69

Figure 63. Photograph of marine shell beads found with Burial 20

69

Figure 64. Field photograph of Burial 22

. . . . . . . . . . . . . . . . . . . . . . . .. 69

Figure 65. Plan map of western portion of Bull Creek cemetery.

71

Figure 66. Photograph of extended burial 38, a multiple burial

72

Figure 67. Field photograph of Burial 44 ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 74

Figure 68. Plan drawing showinginterpretedboundaries of cemetery excavations superimposed

upon Lester's base map.

75

Figure 69. Reconstructed plan of cemeteryshowing area thought by Lester to contain structural

evidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 76

Figure 70. Redrafted copy of a 1950 City Engineers map of the Bull Creek site.

81

Figure 71. Redrafted copy of the map of Kelly's excavations. . . . . . . . . . . . . . . . . . . . .. 82

Figure 72. Enlarged portion of 1950 aerial photograph of the Bull Creek site area

83

Figure 73. Field photograph of Unit 1 showing depth of railroad fill . . . . . . . . . . . . . . . .. 84

Figure 74. Photograph of ExcavationArea 3 of Unit 3, captioned: Dr. A.R. Kelly, Mrs. Wayne

Patterson and Doc discussing plans for work

85

Figure 75. Field photograph with notation: Dr. A.R. Kelly holding pottery vessel to show better

its size and shape. The dirt inside has not been removed

86

Figure 76. Photograph of vessel recovered by Kelly in 1950 . . . . . . . . . . . . . . . . . . . . .. 86

Figure 77. Field photograph captioned: Dr. A.R. Kelly indicating the width ofthe black midden

deposit; charred clay can be seen

87

Figure 78. Photograph dated 1957 showing a Bull Creek dog pot on display at the Smithsonian

Institution

90

x

Figure 79. Copy of a 1957 plat of the J.W. Woodruff property containing the Bull Creek

site

91

Figure 80. Plan map showings areas of Schnell's 1959 investigations

92

Figure 81. Series of photographs showing the 1959 excavations in progress

93

Figure 82. Profile 3 drawing of bank cut near Structure 1 originally excavated in 1936

94

Figure 83. Profile drawing showing the railroad fill at a location nearer the mouth of Bull

Creek and adjacent to XU 2

95

Figure 84. Plan map of pit found in XU-I.

96

Figure 85. Schematic representation of the soil stratigraphy adjacent to XU-2

97

Figure 86. Field photograph showing vessel fragments exposed in excavated Feature 3

98

Figure 87. Photograph of Lamar Complicated Stamped Bowl from Feature 3

98

Figure 88. Illustrations of two Fort Walton-style vessels from XU-2. . . . . . . . . . . . . . . . .. 99

Figure 89. Example of bold incised vessel fragment from XU-2.

100

Figure 90. Plan map showing the location of pits and postmolds in XU-2

101

Figure 91. Copy of a sketch map showing the locations of Fenenga Sites 2-5

104

Figure 92. Examples of ceramics from the Werocoba Creek Site

104

Figure 93. Portion of a 1955 aerial photograph of the Bull Creek site area

106

Figure 94. Copy of sketch map prepared by Chase showing location of excavation units. . . .. 107

Figure 95. Map of the Victory Drive site showing the Feature locations

110

Figure 96.. Facsimile of unattributed newspaper article describing discoveries at the kart racing

track

111

Figure 97. 1960 photograph showing Bettye Broyles excavating a large pit in the kart track

construction area.

112

Figure 98. Plan map of postmold pattern in the Go-Kart locale

113

Figure 99. Two views of pit excavation at the Dolly Madison plant

114

Figure 100. Selected aboriginal artifacts from the Creek pit in the Dolly Madison locale. . . . .. 115

Figure 101. Selected Euro-American artifacts from the Creek pit in the Dolly Madison locale. . 116

Figure 102. Redrafted sketch map prepared by Tyler and Jackson showing landmarks in the

vicinity of Pothole A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 117

Figure 103. Sketch map prepared by Jack Fuller showing the Bull Creek site in 1981.

118

Figure 104. Small complicated stamped vessel from the Bull Creek site.

119

Figure 105. Artist restoration of effigy vessel from trash pit excavated in 1981

. . . . . . .. 119

Figure 106. Examples of Bull Creek phase sherds from a pit excavated by Jack Fuller in 1981. . 120

Figure 107. Examples of incised and complicated stamped sherds from trash pit excavated in

1981.

121

Figure 108. Examples of Bull Creek phase pottery objects and Archaic bifaces found at Bull

Creek in 1981.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 122

Figure 109. Route of riverwalk through the Bull Creek and Go-Kart sites.

126

Figure 110. Photograph of the Bull Creek site as it appeared in 1993, view to the east.

. .. 127

Figure 111. Photograph of landfilling at the Bull Creek site in 1993, view to the southeast. ... 127

Figure 112. Photograph taken in 1993 of the approximately 50 ft thick landfill at the edge of Bull

Creek.

128

Figure 113. Plan map the Bull Creek site in 1993 showing locations of SAS shovel tests and

surface collections.

129

Figure 114. Photograph of 1993 shovel testing.

130

Figure 115. 1993 photograph showing the large erosional feature impacting the Bull Creek

site

131

Figure 116. 1993 photograph of gully wall following profile cleaning.

132

Figure 117. Profile drawing of a portion of the gully wall examined by SAS in 1993. . . . . . .. 133

Figure 118. Photograph of section of profile shown in previous profile drawing

134

xi

Figure 119. Tentative approximation of the boundaries of 9Me1 and 9Me50 based upon survey

results

135

Figure 120. Plan map of data recovery area of the Go-Kart site showing trenches and block

excavations

136

Figure 121. Plan map of block B at the Go-Kart site.

138

Figure 122. Two photographs of the excavation of the large Creek pit in Block B of the Go-Kart

site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 139

Figure 123. Examples of vessels and glass and metal beads from the Creek pit. . . . . . . . . . .. 140

Figure 124. Photograph of Bull Creek phase structure in Block D of the Go-Kart site . . . . . .. 141

Figure 125. Plan map of Block D at the Go-Kart site. . . . . . . . . . . . . . . . . . . . . . . . . . .. 142

Figure 126. Location of Bull Creek showing boundaries of Lamar and Fort Walton ceramic

traditions

143

Figure 127. Geographical extent of the Bull Creek phase . . . . . . . . . . . . . . . . . . . . . . . .. 144

Figure 128. Lamar Complicated Stamped Design motifs from Rood's Landing

145

Figure 129. Graph showing differences in pottery collections from Bull Creek phase and Stewart

phase sites

148

Figure 130. Drawing of typical Stewart phase incised motif

149

Figure 131. Copy of National Park Service summary sheet of 2240 sherds from the WPA

excavations at Bull Creek; various sherd attributes.

150

Figure 132. Copy of National Park Service summary sheet with counts for temper, paste and

texture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 151

Figure 133. Copy of National Park Service summary sheet of 2240 sherds from the 1936-1937

excavations which quantifies attributes of decoration, quality of ware and vessel portidti2

Figure 134. WPA drawings of Bull Creek phase plain and complicated stamped sherds. . . . . .. 153

Figure 135. WPA drawings of incised and punctate sherds from Bull Creek . . . . . . . . . . . .. 154

Figure 136. Enlarged photographs showing tempering agents found at Bull Creek

155

Figure 137. Examples of Lamar Complicated Stamped and Mercier Check Stamped sherds from

Bull Creek

157

Figure 138. Typical examples of incised and punctated sherds from Bull Creek

158

Figure 139. Comparison of pottery counts from four archeological investigations of the Bull

Creek site

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 160

Figure 140. Examples of appliqued rims from the Bull Creek site.

161

Figure 141. Examples of additional rim forms from the Bull Creek site.

162

Figure 142. Copy of National Park Service summary sheet of Bull Creek rim attributes.

163

Figure 143. Graph showing rim varieties identified from a collection of 278 sherds

164

Figure 144. Series of graphs comparing rim type and vessel treatment for sherds examined by

Schnell

166

Figure 145. Photograph of human effigy adorno

168

Figure 146. Animal effigy rim adornos from Bull Creek site collections

169

Figure 147. Examples of vessel forms identified from various Mississippian localities.

170

Figure 148. Partially reconstructed Lamar Complicated Stamped everted rim vessel with

appliqued pinched rim strip

. . . . . . . . . . . . . . . . . . . . . . .. 171

Figure 149. Graph showing the range of rim sizes for everted rim jars from Bull Creek.

.. 172

Figure 150. Examples of Fort Walton carinated bowls from Bull Creek.

173

Figure 151. Graph showing the range of rim sizes for carinated bowls from Bull Creek.

174

Figure 152. Examples of simple rounded bowls with slight to moderate rim constrictions.

175

Figure 153. Graph illustrating range of rim diameters for simple rounded bowls from Bull

Creek. .

176

Figure 154. Example of a simple open bowl from Bull Creek. . . . . . . . . . . . . . . . . . . . . .. 177

xii

Figure 155. Graph illustrating range of rim diameters for simple open bowls from Bull

Creek.

178

Figure 156. Illustrations of effigy bottles from the Bull Creek site with comparative examples

from other western Georgia sites.

179

Figure 157. Examples of motifs for Lamar Complicated Stamped var. Bull Creek and var.

Early. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 181

Figure 158. Example of coarse grit tempered, bold check stamped pottery from the Bull Creek

site

" 182

Figure 159. Comparison of common Lamar Incised designs from three Lamar phases . . . . . .. 184

Figure 160. Examples of Fort Walton incised pottery types and varieties typical of the Bull

Creek phase

186

Figure 161. Incised designs found at Bull Creek. .

" 188

Figure 162. Examples of other pottery types from Bull Creek

190

Figure 163. Examples of ceramic pipes and pottery discs from Bull Creek.

191

Figure 164. Examples of other ground stone objects from Bull Creek. . . . . . . . . . . . . . . . .. 192

Figure 165. Examples of projectile points found on the Bull Creek site which pre-date the Bull

Creek phase occupation.

193

Figure 166. Chert biface associated with the Bull Creek phase occupation. . . . . . . . . . . . . .. 194

Figure 167. Locations of major archeological excavations at Bull Creek.

196

Figure 168. Map showing estimated extent of the Bull Creek Village. . . . . . . . . . . . . . . . .. 199

Figure 169. Plan map of the Bull Creek Cemetery

" 201

Figure 170. Burial orientations in the Bull Creek Cemetery.

. . . . . . . . . . .. 205

Figure 171. Contrasting burial orientations of two burial clusters in the Bull Creek Cemetery. . 206

xiii

THIS PAGE LEFT BLAHK

List of Tables

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

Characteristics of Vessel from Fuller's collection . . . . . . . . . . . . . . . . . . . . .. 123

Tabulation of artifacts recovered from gully midden slump from 9Mel. . . . . . . .. 133

Averett phase radiocarbon dates from the Go-Kart site.

137

Ceramic counts from Schnell's 1959 excavations. . . . . . . . . . . . . . . . . . . . . .. 156

Analysis of sherds from Kelly's 1950 excavation

. . . . . . . . . . . . .. 159

Detailed rim analysis of 278 sherds from Kelly's 1950 excavations at Bull Creek. . 165

Comparison of rim diameter and width of folded pinched rim fold

168

Vessel forms and counts from Bull Creek.

173

Measurements of effigy bottles from Bull Creek.

178

Comparison of temper for zone punctate and incised pottery from Bull Creek

183

Presentation of Scarry's (1985) incised types and varieties of the Bull Creek

Phase. .

185

Incised design categories identified from the Bull Creek site collections.

189

Summary of Bull Creek Cemetery

202

Contrast of Burial Orientations from two burial clusters from Bull Creek

Cemetery

207

xv

TIUS PAGE LEn BLANK

Chapter 1 Introduction and Setting

During the early 1990s, the Columbus/Muscogee County Consolidated Government began designing and constructing several interrelated development projects along its riverfront. The various projects are interwoven geographically, administratively and financially through revenue bond issues, local option sales taxes, Intermodal Surface Transportation Enhancement Act funds and special federal appropriations. Several departments of the City are involved, but the Columbus Water Works is taking the lead on administering the set of projects. The City has been working with the consulting engineering firms of Jordan, Jones & Goulding, Inc. and Hecht, Burdeshaw & Johnson, P.C. in the conceptualization, design and permitting of the projects. The city projects consist of the continuation of sewer line improvements along the river, the continuation of converting the river edge to a landscaped riverfront walkway, a new civic center, an Olympic softball complex, and general landscaping and park/recreation development on a set of contiguous tracts of land that stretch from the south side of downtown Columbus southward to Fort Benning (Figure 1). Some of the planning for the recreation and interpretation aspects of the project, which will focus on tracts near Fort Benning, are still conceptual.

Four segments of the project area have been surveyed, tested and documented as archeological assessment reports (Wood 1991; Wood 1993; Ledbetter 1994a and 1994b).

One facet of the design process has been concern for potential construction impacts to archeological and historical resources in the project area. To allay the City's concerns in this matter, and to be in compliance with Section 106 of the National Historic Preservation Act of 1966, as amended, the City contracted with Southeastern Archeological Services, Inc. (SAS) to perform all necessary archeological surveys, historical research and site evaluations associated with the project.

o

2

5N5--5-~

km

,="",-"""":"""~-~~~~""",,:,,,,,,-~-~---:---,
Figure 1. Location of Columbus riverwalk-related project areas surveyed by SAS from 1991 and 1994.

1

The Bull Creek Site

The project segments consist of sections of Columbus riverfront where substantial improvements to a combined sewer line are to be made, after which the area is to be converted to a riverfront walkway/park. Other areas of proposed development, in large part related to the softball sports complex for the 1996 Olympics, were included in Ledbetter's 1994a survey report.

Cart site (9Me50), which is best known for its historic Creek occupation (ca A.D. 1800). 9Me1 and 9Me50 are actually contiguous sites on the north side of Bull Creek.

Figure 2 shows all archeological sites in the area of the riverwalk route between the South Commons and Fort Benning. All identified archeological sites within and immediately adjacent to the survey areas are shown. Most of these sites are prehistoric and thus relate most directly to the Bull Creek site. The riverwalk route and site locations are depicted on a copy of the 1908 USGS topographic map, which best depicts the landscape prior to modern development.

The archeological survey and assessment work revealed that two sites to be impacted by the proposed project were clearly significant (eligible for listing on the National Register of Historic Places).
The conclusions led to archeological mitigation on the two sites, Bull Creek (9Me1) which contains an important Late Mississippian Bull Creek phase component (ca A.D. 1500) and the Go-

9Me49.

9Me35

N

9Me24
n

9Me14

A
km

Figure 2. Location of known archeological sites between the South Commons of Columbus and Fort Benning (map source USGS Columbus 1908).

2

Introduction and Setting

Both site areas had been severely

included. Realistically, this is the only means

disturbed by past development and only

of examination, simply because there was

portions of each remained preserved. Both

never an adequate determination of the limits

sites have also been subjected to archeological

of the Bull Creek site prior to destruction of

excavation in the past, and the Bull Creek site

most of the site. By piecing together the

has been rather extensively excavated for

results of many surveys and excavations

several decades. The riverwalk was

conducted throughout the present century, we

constructed so as not to disturb intact portions

will attempt to define the extent and status of

of the Bull Creek site. However, state and

the Bull Creek site during aboriginal times.

federal officials agreed that the continued lack

of reporting and publication on various important excavations at Bull Creek was an

Setting

adverse impact that could be remedied by the current project. Therefore, a data recovery plan was developed for compiling the extensive body of hitherto unpublished excavation data for the site. The most significant data relates to the WPA excavations of the 1930s. A synthesis of these excavations and subsequent work on the site is the primary goal of this report.

The Columbus project area is located along the western edge of Muscogee County, Georgia (Figure 3). The most distant points of the overall project area extend a maximum distance of 2.7 km from the river. Bull Creek, which joins the Chattahoochee River approximately 2 km below the South Commons of Columbus, is the most significant tributary stream relating to the riverwalk survey.

Data recovery for the Go-Cart site included extensive field excavation conducted during the summer of 1994. The results of data recovery at that site will be presented as a

Tributaries of Upatoi Creek, another significant tributary of the Chattahoochee River, lie just to the south of the Bull Creek site.

separate report. However, as a result of that

investigation, we have confirmed that

there is substantial overlap between

occupation areas for certain

components, which means the site

boundaries as presently defined are not

entirely meaningful. Perhaps the most

important find of the 1994 excavations at the Go-Cart site was the remains of a

MUSCOGEE COUNTY

Bull Creek phase structure. We also

have evidence that the Go-Cart site

-e.
~

..

occupation areas for the Archaic and

historic Creek components extend onto the Bull Creek site. Because the two site areas are interconnected, the more important archeological discoveries

) Columbus
:@
\
BullCreek Site .... 9Mel

within the defined boundaries of the

Go-Cart site will be included in this report. Pertinent information from other nearby sites, which previous

CHAITAHOOCHEE COUNTY

o

5 Miles

o~~SkIn

researchers have suspected as "part" of

the Bull Creek site will also be

Figure 3. Location map of Columbus and the Bull Creek site.

3

The Bull Creek Site
The Bull Creek site is located in a geographical area known as the Fall Line, which is a zone of varying width that separates the lower Piedmont from the upper Coastal Plain. This zone, also referred to as the Sand Hills and Fall Line Hills, is widest along the Chattahoochee River and narrows to the east as it stretches across Georgia to the Savannah River and into South Carolina (Hodler and Schretter 1986:16-17).
One significant aspect of the Fall Line is that rivers drop dramatically in a short length. At the Fall Line the Chattahoochee River flows across a 4-km long series of bedrock ledges, descending 38 m in elevation before leveling off in Columbus. Chattahoochee Falls is the term given to this series of ledges north of the Dillingham Street Bridge. The backwaters of Walter F. George Reservoir reach this area: which historically has been considered the head of navigation for the Chattahoochee River. From this point south the river widens as it meanders toward the Gulf of Mexico.
The underlying bedrock at the Chattahoochee Falls consists of Precambrian metamorphic and igneous rocks, which form the base of the Piedmont Province. These rocks include granites, gneisses, schists, and granite-gneisses. Below the falls, in the Coastal Plain, are Cretaceous deposits resulting from ancient ocean sediment. Much of the sediment along the river edge below the falls are more recent alluvial deposits from Pleistocene and Holocene times.
Much of the riverwalk survey area follows a nearly level terrace overlooking the Chattahoochee River on the south side of Columbus. Although the Columbus survey area appears nearly level, there is actually about 10 m of elevation difference across the terrace. This relief is most apparent well to the south of the Bull Creek site, where the broad Chattahoochee River terrace abuts the uplands.
4

The main terrace, which includes the Bull Creek site, is about 15 m above average river level. The terrace top varies from about 70 to 73 m (230 to 240 ft) in elevation.
The project area is located upon alluvial soils that form the broad terrace system of the Chattahoochee River. The Bull Creek site vicinity is composed of two principal soil types, both of which provide insights into past land use (Johnson 1983). The area north of Bull Creek is classified as Dothan-Eunola series, 0-3 percent slope. This sandy loam soil of marine origin occurs on stream terraces and can be considered old alluvial deposits, as it is rarely flooded (Johnson 1983:21). The area south of Bull Creek is classified as ChewaclaDogue series. Chewacla-Dogue soils are modem stream deposits consisting of poorly drained to well drained loamy and clayey sands.
Figure 4 shows a portion of the 1922 soil survey map of Muscogee County. The soil associations shown on this early map are important because they represent the terrain immediately preceding the modem period of intensive urban development, clay mining, and landfilling. During the last 100 years tremendous deposits of alluvium have been deposited along the river edge as a result of severe erosion of upland soils in the Piedmont due to poor agricultural practices (Trimble 1974).
Within the Fall Line Sand Hills, upland soils are well-drained and support sparse vegetation represented by scrub oak, longleaf pine and low shrubs (Wharton 1978: 180). Lowland areas along the major streams originally supported a variety of deciduous species similar to that of the Piedmont. Much of the land in the project area was probably cleared for farming early in the settlement of Columbus. In light of the extensive Mississippian and Creek settlement of the area, some tracts of land were probably cleared well before whites settled the area.

Introduction and Setting

ot' Figure 4. PoHion the 1922 Soil SurVeY.Map of Muscogee County showing the project area.

The earliest known mention of Bull Creek itself was by Benjamin Hawkins, who was Principal Temporary Agent for Indian Affairs South of the Ohio. The first entry is dated January 9, 1797.
The banks of the river generally fifty feet high down to the town house (Coweta on the west bank). A creek empties in half a mile above, and it descends with a rapid current, going up this creek a small one {Weracoba} joins it on the left, about 200 yards from the river, continuing up one mile, the lands on both sides to the margin without any swamp; the bank steep enough anywhere for mills.
(Hawkins, 1916:60).

Two years later Hawkins provided another description.
In descending the river path from the falls in three miles you cross a creek running to the right, twenty feet wide; this creek joins the river a quarter of a mile above the Cowetuh town house: the land to this creek is good and level and extends back from the river from half to threequarters of a mile to the pine forest; the growth on the level is oak, hickory, and pine; there are some ponds and slashes back next to the pine forest, bordering on a branch which runs parallel with the river; in the pine forest there is some reedy branches.

5

The Bull Creek Site
The creek has its source nearly twenty miles from the river, and runs nearly parallel with it till within one mile of its junction; there is makes a short bend round north, thense west to the river; at the second bend, about two hundred yards from the river, a fine little spring creek on its right bank
(Hawkins 1916).
A rather poetic description of the setting for the Bull Creek site excavations is provided in an undated manuscript prepared by Isabel Patterson which was eventually edited for brevity and published in Early Georgia (Patterson 1950).
The geographic setting of the Bull Creek Site is significant in understanding the implications of the unusual finds during fourteen months of archaeological investigation. Bull Creek rises in rolling red foothills about twelve miles south of Pine Mountain Valley and approximately the same distance from the Chattahoochee River, which forms the western boundary of this section of the Empire State of the South. Bull Creek flows southward for nearly twenty miles almost parallel to the river until within one mile of their confluence, here the creek makes a short bend around to the north and there turns twice in a westerly direction to join the river three miles below the "Coweta Falls," and a quarter of a mile north of the historic site of the "Coweta Town House" on the west bank of the Chattahoochee. A fine little spring creek, now called "Weracoba," joins Bull Creek on its northern bank six hundred and sixty feet from the river. These streams form the eastern, southern and western boundaries of the Bull Creek Site and Victory Drive - (once the old Federal Road and later Lumpkin Boulevard) forms the northern -Briefly, this ancient home of aboriginal Americans covers three quarters of a square mile in the northwestern corner of section 66 of the Coweta Reserve on the east bank of the Chattahoochee, where in 1828, the "Trading Town of Columbus was established near the Coweta Falls" in Muscogee County, Georgia.
The Chattahoochee appears prominent as an avenue of glorious adventure associated with trails of legend and later history. Its head springs rise in the majestic Blue Ridge Mountains of northeastern Georgia, within a few hundred yards of the Hiwassee, which flows northwest to join the Tennessee and merge with the Mississippi. The course of the Chattahoochee is southwest from the "Valley of Nacoochee" to "Chattahoochee Old Town" in Troup County, where it forms the dividing line between the present State of Georgia
6

and Alabama. The river flows almost due south to the "Coweta Falls" where rough rocks and islands force the stream into two channels and here the river is fordable to the termination of the Falls. A creek once known as "Falls Creek" enters the river at this point just north of the site of the bridge that was built across the lower end of the Falls at Columbus (1832-33) to connect with trails centuries old.
South of the present city of Columbus, the river winds along forming great bends, and Coweta Town on the west bank occupied the first of these. As the river snakes along southward around "Jenny's Island" there are two famous aboriginal sites near the shoals "Woolfolks Mound" on the east and "Coweta Tallahassee" on the west - later known as "Abercrombie Mound," then "Broken Arrow Bend" just north of Upatoi Creek on the east and below this the great "Kasihta Town" of historic times - on the low flat of good land now occupied by Lawson Field - Fort Benning, Georgia.
The most important trail of this section was the branch of the Lower Creek Trading Path from the Savannah, near Augusta, to "Rock Landing" on the Oconee, below Milledgeville, through "Old Ocmulgee Fields" at Macon, and on across the Flint, near Benjamin Hawkins Creek Agency and westerly to Kasihta, where it branched - one going north through the Kasihta Fields to Coweta Falls and the other trail across the Chattahoochee to the west side where Fort Mitchell was later established. This trail was later followed by the Old Federal Road and a line of forts was established.
The "Point Towns" of the Chattahoochee lie in the Oswichee Bend south of Uchee Creek on the west bank and here too was the ancient Apalachicola Town - when the legendary Creeks came to this area. The river below this great bend flows again southward to be joined by the water of the Flint and forms the Apalachicola at the southwestern edge of the present State of Georgia and these continue to the Gulf of Mexico - through
northwestern Florida (Patterson n.d.).
Schnell (1963) provides a description of the Bull Creek site as it appeared around 1960.
A tributary, Weracoba Creek, joins Bull Creek on its north bank about 200 m from the river. These three waterways form the eastern, southern and western boundaries of the site.

The site is roughly bounded on the north by Victory Highway {U.S. 27 & 280}. The site may have extended slightly beyond Victory Highway but the remains, if any, were obliterated by the construction of the Columbus Municipal Airport and a Nehi Bottling Plant. The Bull Creek Site is entirely within the city limits of Columbus, Georgia.
The site itself has been impinged upon by several modern developments. Until a few years ago, most of the back section {away from the river} of the site was covered over by a race track, with stables in the fore section. Previous to that time, a railroad had been built across the site and a railroad bridge spanned Bull Creek at its confluence with the Chattahoochee. The bed for this track covered over or destroyed a large part of the peninsular portion of the site. The back part of the site is currently covered over by a golf driving range.
The soils of this area are derived from the unconsolidated sand and clay beds of the Coastal Plain but may be traced in part to the Piedmont Plateau from which they were transported by the streams and the river. The soils on the site are built up upon a base of fossiliferous marl. The flood waters of the Chattahoochee and its tributary streams such as Bull Creek have periodically covered this area, especially notable floods having occurred in 1919 and 1939. The topography of the land is nearly level above the bluffs of the river, with slight ridges and depressions. The bluffs of the river are about fifty feet high at the site, with numerous springs flowing from its side. Bull Creek joins the Chattahoochee River three miles below the Fall Line which divides the Piedmont region of northern Georgia from the southern Coastal Plain.
Construction of a Go-Kart race track just north of the Bull Creek Site uncovered in 19591960 evidences of an as yet, undocumented historic settlement. Two large pits yielded a large amount of Chattahoochee Brushed pottery, some Kasita Red Filmed, and numerous trade items, including glass, china ware, several iron mugs, and one musket barrel. At various other points on the site isolated finds of trade items have also been
found (Schnell 1963).
It is obvious from these observations that several conditions were highly favorable for the site's location. Perhaps of greatest importance were proximity to several water

Introduction and Setting
sources which included large springs, tributary streams and the river, sufficient elevation above the river to limit the disturbance from frequent flooding, and access to broad expanses of fertile farmland. Upland resources and particularly the rich resources of the falls were in close proximity.
Culture History
As a result of numerous archeological projects and overviews conducted within the region since the 1890s, the culture history of the Columbus/Lake Walter F. George area is relatively well documented. A survey report by Knight and Mistovich (1984) entitled Walter F. George Lake; Archaeological Survey of Fee Owned Lands was produced for the U.S. Army Corps of Engineers and is the most detailed reference for the region. A second detailed culture history of the region has recently been prepared for the Fort Benning area (Elliott et al. 1995). The known culture history of the region will be only briefly summarized here. Emphasis will be placed upon the periods most pertinent to the Bull Creek site, specifically the Mississippian and historic Indian periods. Figure 5 shows locations of important archeological sites referenced in the text.
The first people came to the Chattahoochee Valley during the Paleo indian period, approximately 10,000 to 12,000 years ago (Anderson et al. 1990), when the climate was quite different from today's and a canopy of oaks and other hardwoods covered much of the Southeast. Temperatures were cooler and possibly more equable as a result of the continental ice-sheet still present to the north. As the ice melted, the sea level began to rise and encroach on the Coastal Plain. It is believed that the earliest inhabitants were nomadic hunters and gatherers who occasionally killed mastodons, bison, and other Pleistocene megafauna. Through time these large animals became extinct due to human predation and climate change.
7

The Bull Creek Site PIEDMONT

Roods Landing

Kolomoki .~s~,

COASTAL PLAIN

N

I

o

SOMi

lI ::::====:::::"

Figure 5. Locations of sites referenced in text. 8

As prehistoric cultures adapted to the extinction of the large herbivores and climate changes, white-tailed deer became a prime component of the Indian's diet. The interval between approximately 8000-1000 B.C. is defined as the Archaic cultural period. Early in this period the Fall Line region of southwestern Georgia and southeastern Alabama was regularly exploited by Archaic hunters and gatherers and sites containing stone tools diagnostic of the Early Archaic period are fairly common. By about 6000 B.C. the climate became warmer and drier, and, while the forests still were characterized by oaks and other hardwoods, a more open vegetation pattern is indicated (Delcourt and Delcourt 1985). Relatively little is known of the Middle Archaic occupation of the Chattahoochee Valley. However, both Piedmont (Morrow Mountain) and Coastal Plain tradition (stemmed chert points) stone tools have been reported.
More reliable information is available concerning. the Late Archaic and Gulf Formational periods, which date from ca. 4000-800 B.C. Diagnostic stone tools include stemmed projectile points of various types such as Pickwick, Little Bear Creek, Abbey and Savannah River Stemmed. During this time the climate became moister and may have been similar to current conditions. Pines began to eclipse oaks as the predominant species in Coastal Plain forests. During the latter part of the Late Archaic period pottery was invented, marking the beginning of the Gulf Formational period in the region (Jenkins 1978). The earliest pottery is very distinctive because the clay was tempered with vegetable fiber, probably Spanish Moss. Fiber tempered pottery sites are particularly common along the Chattahoochee River below the Piedmont. It was during the Late Archaic and Gulf Formational periods that squash was introduced as the first domesticated garden vegetable. The addition of pottery and domesticated crops suggests that prehistoric lifeways were becoming more sedentary.

Introduction and Setting
The Early Woodland period dates from approximately 800-300 B'C, in the study area, but is poorly defined due to a genuine scarcity of identified sites of this period. Early Woodland pottery is tempered with sand and surfaces of vessels are often decorated with simple stamped and check stamped designs. During the succeeding Middle Woodland period (300 B.C.-A.D. 500), the Chattahoochee Valley was the scene of distinct changes.
Based on greater numbers of sites, there appear to have been both a population increase and the development of ceremonial activity echoing developments elsewhere in the eastern Woodlands. Corn was added to the other Woodland cultigens, but it did not playa vital role in the diet until the late prehistoric period.
The Mandeville Site (9Cy1), located in Clay County, Georgia is an important mound center during the Middle Woodland period. The site consists of a large village area with two mounds. One mound is an earthen platform which may have supported a ceremonial structure such as a charnel house or dwelling of the village leader, and the other is a conical burial mound. Ceramics from the site include check stamped wares and pottery decorated with new curvilinear designs known as Swift Creek Complicated Stamped.
There are clear indications that during the Middle Woodland period people in the Chattahoochee Valley were interacting with groups in Ohio, Tennessee, Alabama, Mississippi, and Florida. Late in the Middle Woodland period the foreign cultural influence waned and the settlement pattern seems to have become more nucleated, centering on the large sites such as Kolomoki to the south in Early County, Georgia (Sears 1956).
9

The Bull Creek Site
Between A.D. 500-750 the southern portion of the Chattahoochee Valley was influenced by people associated with the Weeden Island culture of Gulf Coast derivation. These groups shared similar pottery-making traits and ceremonial activities. A late variety of Swift Creek Complicated Stamped pottery was made, along with Gulf Tradition incised, punctated, and plain vessels. Kolomoki continued as a dominant settlement during this time. Again, few sites of this period have been identified in the surrounding area, probably because few locales have received more than basic reconnaissance surveys.
Even less is known about developments from ca. A.D. 750-850. Late Weeden Island cultures occupied the Apalachicola River valley in numerous settlements. According to Knight and Mistovich (1984) only a few late Weeden Island components are present in Walter F. George Reservoir and these are restricted to the southern part of the lake. During this interval the area around Columbus appears to have been occupied by cultures producing predominantly plain pottery. This may correspond to Chase's (1959) Upatoi complex, identified from excavations in the Fort Benning area. The Upatoi complex is characterized by a large percentage of plain pottery and small percentages of chevronincised and "ripple tooled" vessels (Chase 1959). If the complex persisted, it may have evolved into the early Mississippi Averett.
The Mississippi period dates from approximately A.D. 900 to 1550. The Averett phase is sandwiched between two emergent chiefdoms to the north and south. The "heartland" of Etowah culture was north of the Fall Line, centered in northwest Georgia and northeast Alabama. To the south the Fort Walton and Rood phase cultures were replacing Weeden Island cultural antecedents in the lower Chattahoochee and upper Apalachicola River systems. Differing from contemporaneous
10

Etowah pottery, characterized by rectilinear complicated stamped jars and bowls, Rood phase ceramics are typically plain, grit-tempered, globular, collared jars with strap handles. Stylistically, Averett vessels are virtually identical to those found in west-central Alabama (West Jefferson phase) and at the Macon Plateau site (Ocmulgee National Monument) on the Ocmulgee River in central Georgia. The sudden appearance of totally new vessel forms made with grit tempered clay suggests a radical cultural change and probably an influx of people from outside the valley. The Averett phase is well dated to between A.D. 900 and 1300 (Schnell and Wright 1993:28; Ledbetter 1995).
South of Columbus, a distinct culture known as the Rood phase developed from ca. A.D. 950 - 1400 (Schnell et al. 1981). In the area encompassed by the Rood phase, a hierarchy of site types is evident. Principal settlements contained multiple platform mounds surrounding a central plaza. Dwellings were scattered up and down the river from the main center. Smaller towns had only one to three mounds. Com agriculture became very important during the Rood phase. According to a recent examination of Mississippian settlement within the region, Rood phase sites do not occur in the Columbus area, although examples of Rood phase pottery do occur on Averett sites (Schnell and Wright 1993:28).
Sometime after A.D. 1300 the Rood phase was replaced by or evolved into the recently defined Singer phase (Schnell and Wright 1993:20). Ceramics of the Singer phase include early Lamar types such as Lamar Complicated Stamped and Lamar Bold Incised, as well as styles similar to Savannah Complicated Stamped. Fort Walton Zone Punctate sherds also occur. At present the territorial range of the Singer phase has not been extended into the immediate Columbus area (Schnell and Wright 1993).

By about A.D. 1400 sites containing Bull Creek phase material commonly appear in the Columbus area. According to Knight and Mistovich (1984) and Schnell and Wright (1993) Bull Creek ceramics can be considered as a blend of generic Lamar with generic Fort Walton of this period. Fort Walton Zone Punctate and Lamar Complicated Stamped pottery characterize the mixture of Gulf and Piedmont ceramic traditions.
Corn agriculture continued to be important in Bull Creek subsistence patterns and the settlement hierarchy was similar to the preceding Rood phase. Large villages appear to have functioned as political and ceremonial centers, while small farmsteads were located in outlying areas. Bull Creek phase sites are more widespread than Rood phase sites and are found throughout the middle Chattahoochee Valley up to the Fall Line at Columbus. Because Bull Creek sites occupy a more northerly territory, there are indications that Averett groups were replaced by, or absorbed into, a true chiefdom level society, one that was contemporaneous with chiefdoms in central Georgia (on the Oconee and Ocmulgee Rivers) and in the tri-state area of northeast Alabama, northwest Georgia, and south central Tennessee (on the Coosa, Etowah, Coosawatee, and Tennessee Rivers).
Recent work by Schnell indicates that the latter part of the Bull Creek period should now be considered a distinct phase (Schnell 1990; Schnell and Wright 1993:21). In this scenario the Bull Creek phase would date from approximately A.D. 1400 to 1475 and the Stewart phase would continue until approximately A.D. 1~50. The Stewart phase is distinguished by a decrease in the proportion of complicated stamped pottery (Schnell 1990:67-68; Schnell and Wright 1993:21).
The Hernando De Soto expedition penetrated the Southeast in 1540 and

Introduction and Setting
represents the first European exploration in the interior. However, it does not appear that De Soto visited the Chattahoochee Valley. Instead, after leaving his winter camp in what is now Tallahassee, Florida, he marched into central Georgia, followed the Flint River for a distance, then turned northeast, crossing the Ocmulgee River just south of Macon. While the expedition did not encounter Bull Creek phase people, there can be little doubt that news of the invaders resulted in disruptions.
It was not until the latter part of the seventeenth century that Europeans made sustained contact with the native societies of the lower Chattahoochee Valley. However, there appears to be a reduction in the number of sites dating to A.D. 1550-1650 as compared to the previous Bull Creek phase. If this corresponds to a population reduction, then European diseases may have had a severe impact. Population reduction must have been proportionately higher in a nucleated settlement where disease could be easily spread from household to household.
Pottery of the Abercrombie phase, the period from A.D. 1550-1650, shows continuity with the preceding Bull Creek/Stewart phases, with Lamar Complicated Stamped still present in low percentages, with bold incised, and Fort Walton zone punctate wares as majority ceramic types. A distinctive attribute of Abercrombie is the occurrence of shelltempering in some vessels. There appear to have been strong cultural affiliations with the Dallas culture of northwestern Georgia and eastern Tennessee (Jenkins 1978) and the Alabama River phase in central Alabama (Knight and Mistovich 1984).
During the first half of the seventeenth century, a chain of Catholic missions was established across north-central Florida and along the Sea Islands of the Georgia coast.
11

The Bull Creek Site
San Luis de Talimali, located in Tallahassee, was the principal mission town. It was protected by a bastioned fort and a Spanish garrison. From this mission in 1679, Father Juan Ocon and two other missionaries were sent to administer to the population in the Chattahoochee Valley, but were ordered to leave by the great chief of Coweta. In 1681, two more Franciscans attempted to establish a mission, reinforced with an escort of soldiers, but again the natives resisted and the foreigners were forced to withdraw.
The settlement of Charles Towne by the English in 1670 proved that the Spanish had only weak control of the Southeast. By 1680 many of the missions along the Georgia coast had been abandoned due to raids by Indians allied with the English, and the friars and their charges retreated southward to the protection of St. Augustine. The English aggressively pursued trade with the Indians, and soon traders were making their way west into the Chattahoochee Valley. From their main supply depot at Charles Towne, the traders exchanged firearms, cloth, rum, and other material goods for deerskins and Indian slaves. The desire for European goods prompted intertribal warfare and slave raiding. The 1685 visit to Coweta by Dr. Henry Woodward, an English trader, brought a quick response from the Spanish garrison at San Luis de Talimali. Commander Antonio Matheos marched into the valley with 250 Apalachee Indians and Spanish soldiers in an unsuccessful attempt to capture Dr. Woodward. They returned in the winter of 1685 and again failed to capture Dr. Woodward, but in the meantime they burned four Indian towns: Coweta, Kasita, Tasquiqui, and Colone. The Spaniards then built a bastioned fort, Fort Apalachicola, at the town of the same name on the west bank of the Chattahoochee River. The Spanish coercion was intended to force the Indians into an alliance, but ultimately it had the opposite effect.
12

Shortly after the completion of Fort Apalachicola, many nearby Indian towns were abandoned, and there was a large-scale migration eastward to the Ocmulgee River, near present-day Macon. There, the Carolinians intensified trade and continued to supply the Indians, referred to by the English as Creeks, with firearms and other goods. In 1704, the Creeks, under the command of Col. James Moore of South Carolina, attacked the mission villages of north Florida and whole villages were massacred or carried off into slavery. The fortified mission of San Luis de Talimali was one of the few to withstand the onslaught. Spanish influence over the Indians withered as time passed, leaving much of the Georgia coast, the Chattahoochee Valley, and north Florida virtually depopulated.
Indian ceramics of this period, the Blackmon phase (A.D. 1650-1715), include the types Chattahoochee Brushed, Ocmulgee Fields Incised, Kasita Red Filmed, and coarse and fine plainwares. Ocmulgee Fields Incised represents the late variety of Lamar Incised. Through time, the incised motifs got more complicated and the width of the incised lines became very narrow.
After years of English exploitation the Indians rebelled in 1715 in what is known as the Yamassee War. A pan-Indian movement, the attacks were well coordinated and focused on white settlements along the east coast; more than 400 colonists were killed. English traders were also systematically executed in Creek and Choctaw towns as far west as Alabama. The revolt was quickly suppressed and the Indians fled to the west, re-occupying the Chattahoochee Valley. At this time many diverse groups settled in the valley including the Oconee, Savannah, Yuchi, Apalachee, Hitchitee, Sawokli, Eufaula and the Apalachicola peoples. It is informative to note that there were language divisions in the Valley after the influx of refugees.

The northern valley towns, including Coweta, Kasita, and Apalachicola spoke Muscogean, while the southern towns, Hitchitee, Oconee, Sawokli and Sawokliutchi, spoke the "Stincard" tongue, or Hitchitee. Sandwiched between these two groups were the Yuchi who spoke a third distinct language. Collectively, these settlements, inhabited by the remnants of once powerful chiefdoms, formed the Lower Creek towns. The Upper Creek towns were located in Alabama on the Coosa and Tallapoosa Rivers.
Aboriginal occupation near Columbus from A.D. 1715-1836 is termed the Lawson Field phase. The ceramics show a continuation from the Blackmon phase and include Chattahoochee Brushed, Kasita Red Filmed, and Ocmulgee Fields Incised pottery types, with Ocmulgee Fields Incised out of production by the mid-eighteenth century. European artifacts are fairly common on Lawson Field phase sites, consisting primarily of gun parts, rum bottle fragments, sheet brass fragments (recycled brass kettles), clay pipe fragments, and glass beads. Apparently, the Creeks did not develop a preference for European ceramics as did many of the Cherokees who lived in north Georgia.
The Indians in the Lower Creek towns became more acculturated than their counterparts in the Upper Creek towns. In part this was due to the efforts of Benjamin Hawkins, the Indian agent appointed by the federal government. He encouraged the Lower Creeks to adopt modern farming techniques, such as intensifying and diversifying their crops, and using the plow. These efforts paid off in 1813 when the Upper Creeks declared war on American settlers in south Alabama and the Lower Creeks joined American forces to defeat the Upper Creeks in the Red Stick War. Tensions were running high at this time. Although Spain claimed Florida until 1821, there was a strong British presence in the town

Introduction and Setting
of Pensacola. British agents were willing to pay Creek Indians a bounty for every American scalp brought in. In addition, the Shawnee prophet, Tecumseh, had visited the Upper Creeks and exhorted them to attack the Americans. There was a skirmish between the two forces at Burnt Corn Creek, but the major event occurred in 1814 when the warriors from the Upper Creek towns, known as the Red Sticks, attacked and massacred nearly 500 men, women, and children at Fort Mims, located north of Mobile. The Red Sticks were defeated by American and Lower Creek forces under the command of General Andrew Jackson at the Battle of Horseshoe Bend in central Alabama. Jackson then marched into Spanish West Florida, where he captured Pensacola after a brief siege.
In 1825, Creek territory in Georgia and Alabama was ceded to the United States in exchange for an equal amount of land further west. The spokesman for the Creeks was William McIntosh, a mixed-blood chief. Acting without consulting other leaders, he signed the fraudulent Treaty of Indian Springs, which led to his assassination. By 1827, Georgia claimed control of the former Indian property and a state lottery for Muscogee County was held. Land lots were 202.5 ac in size (Figure 6). In 1830 the federal Indian Removal Bill was passed. Fort Mitchell, located on the west bank of the Chattahoochee River, became the main collection point for Indians living in the valley. Between 1830 and 1836 the fort served as the point of embarkation for the long journey to Oklahoma. Due to the fraudulent treaty, a new Creek "war" flared up along the Chattahoochee in 1836. The small settlement of Roanoke, located a few miles north of present day Georgetown, was attacked and burned and twelve of the inhabitants were killed by Indians. Similar raids up and down the Chattahoochee River resulted in regular patrols by the Georgia militia.
13

The Bull Creek Site

The area

around the Bull Creek

site has received

relatively little atten-

tion with respect to

detailed mapping or

historical accounting

over the past two

centuries. Following

the establishment of

Columbus in the latter

1820s, the area con-

tinued to be used pri-

marily for farming.

Much of the area lay

within extensive land

holdings of large

plantations. Wool-

folk's plantation,

which lies to the south

of the Bull Creek site,

is perhaps the best

documented (Ashley

and Schnell 1928).

Later, much of the

land was gradually

Figure 6. Surveyor's plat of land lot containing the Bull Creek site.

mined away for the manufacture of brick.

The area was grad-

The location at the head of navigation

ually developed with the expanded population

on the Chattahoochee River and the presence

growth of Columbus and following the est-

of tremendous water power potential were

ablishment of Camp Benning in 1918. Within

important economic factors for the establis-

the project area south of Bull Creek, dev-

hment of Columbus in 1828. The town was

elopment has not been as extensive.

laid out in a grid system with a long axis

paralleling the river to the west. Commons

From both archeological and historical

were established to surround the town on all

perspectives, an important period of time

four sides. The history of Columbus has been

associated with the project area relates to the

detailed in a number of reports and papers.

frontier community of the late eighteenth and

From an archeological perspective, this history

early nineteenth centuries. During this period

is detailed by Lupold and Schnell 1991; Wood

the area was the point of direct contact be-

1993; Ledbetter 1994. The reader is directed

tween Creeks and white traders and settlers.

to these documents for historical accounts of

During this time the Creek town of Coweta lay

the city north of the Bull Creek site

directly across the river in present-day Russell

boundaries.

County, Alabama.

14

Introduction and Setting

On the Georgia side, settlements

mouth of Bull Creek (Sturges 1818). Maps

composed of Creeks, whites and half-breeds

prepared by DeJarnette (1975:9) and Swanton

developed around Kennard's Ferry located in

(1922: plate 9) show the location of Marshalls

the South Commons of Columbus and

slightly below the confluence. A map

Marshall's Trading Post, which lay south of

prepared by Isabel Patterson of the Bull Creek

Bull Creek. The settlement at Kennard's

area shows the location of Marshall's Trading

Ferry was located at the point where an old

Post as just south of Bull Creek near the

trail crossed the Chattahoochee River in the

confluence with the Chattahoochee River

South Commons area (Lupold and Schnell

(Patterson 1950:34).

1991:13). The settlement known as Wewoka,

which developed around the ferry, contained

Information concerning the trading

several log houses, an inn, and four stores.

post is presented in an early archeological

Possibly, as many as three hundred people

report of the Bull Creek area (Schnell 1979).

may have lived in Wewoka prior to the

In that report, Schnell notes that archeological

founding of Columbus in 1828 (Martin

site 9Me30 is the apparent location of

1874:10). Kennard's Ferry and the old trail

Marshall's trading post. He also makes

appear on the earliest survey maps of

reference to the July 7, 1797 records of

Columbus. The location of Kennard's Ferry

Richard Thomas, secretary to Benjamin

lies just south of present-day Oglethorpe

Hawkins, concerning an Indian trader by the

Bridge. Following the formal creation of

name of Thomas Marshall, "a trader in the

Columbus in 1828 the Wewoka settlement dis-

Coweta Town, very well off in negroes, cattle,

persed (LupoId and Schnell 1991:13).

etc." (Hawkins 1916:467). This reference

infers that Thomas Marshall was the proprietor

The history, as well as the location, of

of Marshall's trading post.

Marshall's Trading Post

is less well defined. A

map published in the

early nineteenth century

(Sturges 1818) shows the

settlement of Marshalls

on the south side of a

large creek opposite the town of Coweta (Figure 7). This stream has been interpreted by many as

--- ---.._- --...

Bull Creek (Brannon

1930; DeJarnette

1975:19). The larger

stream to the south called

Hatchee-thluc-co on the

1818 map is present-day

Upatoi Creek (Brannon

1930). The Sturges map

,

r-'
(

shows Marshalls as a

\)

small group of structures Figure 7. Portion of Sturges' 1818 map of Georgia (published by Eleazer

clustered around the

Early) showing the settlement of Marshalls on south side of Bull Creek.

15

The Bull Creek Site

An unpublished manuscript by Margaret Ashley and Frank Schnell, Sr., provides a somewhat different interpretation of the Sturges map. These researchers spent several days in 1928 digging on the Kyle's Landing site (9Me2), located about 4 krn south of Bull Creek, and recovered European trade material that postdated the mound constructed on the site (Ashley and Schnell 1928:13). Ashley and Schnell concluded that the village site at Kyle's landing was Benjamin Marshall's trading post (Ashley and Schnell 1928:14).
The authors note: In looking over various
maps in an endeavor to identify this site with some historically known village, I noticed that there is a place marked "Marshall's" on plate 9 of Swanton's "Early History of the Creek Indians and Their Neighbors." If the Kyle's Bend site is the remains of this trade village it would be dated at about 1818. This would not mean that the mound was built at that time, but rather that an old village to which the mound belonged had been made into a trading post by Benjamin Marshall. This Indian, a halfbreed, was a contemporary of William McGilivary and Opotheoholo. His name is annexed to treaties of 1832 and from the history of Columbus one can well judge that he played an important part in the early days of the city, as he was one time granted land in Alabama opposite Columbus, on the land
upon which Phoenix now stands (Ashley and Schnell 1928:14).

Patterson's published map does not show the ferry at the end of the old path (Patterson 1950:34), notations on Patterson's 1907 field map do identify the ferry as such.
Aside from the initial date of operation, little is known of the ferry. The owners of the ferry were well-known and influential people. Seaborn Jones was a highly respected businessman and politician who initiated manufacturing at the falls of Columbus in 1828 with construction of a grist mill. Jones also served as a member of Congress (Martin 1874:43).

Another site of some historical importance is the ferry established in 1831 by Seaborn Jones and Stephen M. Ingersoll (Martin 1874:29). According to Martin, a new ferry, located about one mile below the town of Columbus, was established in June of that year (Martin 1874:29). The approximate location of the ferry is shown in Figure 8 on the map of the Bull Creek excavations of 1936. Although

MOliN'
MoVNC
Figure 8. Portion of Patterson's (1950) map of the Bull Creek area. The location of the 1831 ferry is shown at the end of the old path (label is our notation).

16

Introduction and Setting

Dr. Stephen M. Ingersoll was an able businessman who moved to Russell County, Alabama soon after the founding of Columbus. Ingersoll maintained friendly relationships with the Creeks and served as a mediator between the Indians and white settlers (Martin 1873:42). Apparently, Ingersoll died soon after the ferry was established (Martin 1873:42).

The Bull Creek site was a large corn field at the time of the first excavations (Figure 9). However, the Bull Creek site had already been affected by a number of landscape modifications associated with the mining and transportation of clay used for the production of bricks. The remains of a 1920s railroad bed and bridge are still evident across Bull Creek.

The elevated railroad

bed and trestle were built

across the mouth of Bull Creek

in 1926 by the Columbus Brick

and Tile Company to transport

clay from pits south of the

creek to their plant. The

bridge washed away during a

flood in the following year and was rebuilt at a new location at the confluence of Werocoba

Figure9.1936 photograph of the Bull Creek site prior to excavation
showing the abandoned com fields (NPS photograph 2368).

Creek and Bull Creek

(Patterson 1950:37). Borrowing of fill dirt for

survived modern destruction. That area lies

construction of the first bridge apparently

beneath the fill dirt of that narrow railroad

destroyed substantial portions of the prehistoric

spur embankment. Figure 10 shows the site

site (9Me1) at Bull Creek and the Confluence

area as it appeared in the late 1930s.

site (9Me32). There is an ironic twist to this

damaging period of construction. Today, only

one small part of the Bull Creek site has

17

The Bull Creek Site
Figure 10. Portion of a 1938 Soil Conservation SerVice aerial photograph showing the area of the Bull Creek site (area near the mouth of Bull Creek is distorted by photo overlap). 18

Chapter 2 Early Investigations Background to the WPA Excavations

The Bull Creek Site first appeared in the literature in 1925 (Mulvihill 1925:36-41). In the previous year, Frank J. Mulvihill of Cambridge, Massachusetts made a survey of planned recreational areas for the city of Columbus. While studying on the ground for the possible creation of parkways and parks in the Bull Creek drainage, Mulvihill noted and made collections from several Indian sites. Figure 11 shows the locations of sites found by Mulvihill. The Bull Creek site was the first located and was designated Site 1. Mulvihill's short article on his findings gives the following description of the site.
Site 1. At the south end of Tenth Avenue in Columbus is the city cemetery. Alongside the cemetery between it and the river, is a road or wagon path which follows the top of the river bank in a southeasterly direction. The road passes a rifle range and continues on to a pasture where, upon opening and closing a gate, a much frequented path to the right leads to springs at the bottom of the river bluff. But, continuing on the road, a house and farm buildings are reached. To the north of the house the surface of the ground was turned up. In such places were found fragments of pottery. Of the one hundred and twenty-five pieces selected no two seemed to match or come from the same pot. Some few are reddish, most of them are gray, and some are bluish black on one side. All are unglazed, and the degrees of hardness varies. The mixture used is sand and clay. Sharp quartz is evident in some fragments, while a few had shell particles for a binder. There are plain surfaces and ornamented surfaces. Various traces of basket weaving or grass mat patterns distinctly show on some, while others had bands of incised lines and dots, or both, five pointed stars in relief and vertical imprints giving an effect of beading. No fragment is large enough to reconstruct or work out the original design or convey the size and shape of
the original vessel. (Mulvihill 1925:37).

found a possible Indian Mound near the Bull Creek site. While the mound was later determined to be a modern feature (see following), Mulvihill's published account of this mound (Site 5), so close to the Bull Creek site, has been a matter of much confusion through the years.
In 1928 Margaret Ashley, a student of W.K. Moorhead of the Peabody Museum, conducted survey and excavations in the Chattahoochee/Flint river valleys. She was engaged by Mr. Tom Houston of Columbus to make a survey of the Columbus area to find potentially productive sites for excavation. During this survey Ashley and her assistant Frank T. Schnell, Sf. (Figure 12) revisited the sites described by Mulvihill. The photograph showing the two archeologists was taken that year at the Neisler Mound site on the Flint River.

Key TO Lo eAT/oNS
Sihs t wAien /~diOJl
fOlJ."d o~ the $u.rPo.ee
CoLCJ!'1l3CJ.5, GoR.GtA
SPT. 6-q If:;",!

Relies

were

err;
C,..oL.1JMf>\JS
a e.R

Mulvihill's ceramic descriptions match well the types found by later researchers at Bull Creek. Mulvihill also

Figure 11. Map of sites found by Mulvihill in 1924. The Bull Creek Site is site 1 (redrafted map by Mulvihill published in 1925).

19

The Bull Creek Site

specimens were found, though Mulvihill (p. 37), states that pottery fragments were plentiful. He lists this place as Site # 1. Across the creek from this site is still another. It covers a rather large area and offers much pottery and flint material. The dirt that has been thrown up on either side of the railroad bed contains fragments as well. I do not doubt but that this was part of site # 2, though I have
given it the number of site # 6." (Ashley and Schnell 1928: 15).

Ashley's Site 6, later named the Confluence Site (Fenenga and Fenenga 1945), was described in further detail.

It covers a rather large area and offers much pottery and flint material. The site is just at the junction of Bufl Creek and the Chattahoochee River and it was found that the village was heaviest on the creek side rather than the river. This was
also true of Site # 2. (Ashley and Schnell 1928: 17).

Ashley and Schnell also examined Mulvihill's Indian Mound located north of the Bull Creek site. Mulvihill had described the mound in the following manner:

A conical mound used at the

present time as the butt of a rifle range.

Figure 12. Photograph of Margaret Ashley and FraDkSchnell,
Sr., taken during their work in western Georgia in 1928 (Columbus Ledger Magazine, October 21, 1928).

was unable to get an explanation of its origin or why it is so situated. A shape not natural on the plain caught my eye. If it were constructed for use as a rifle range

butt, the handling of the earth to make

Ashley expended relatively little time in examining Mulvihill's Site 1. Apparently

the mound is a better piece of work than I can credit the average worker of today. The soil or marl was scooped from an area of uniform distance

the land had lain fallow since the time of

from the base of the cone. Where successive

Mulvihill's visit and there was little surface exposure. Ashley's report describes the Bull Creek site, which she numbered as Site 2, as follows:

peckings of shot have exposed the earth of the mound back of and above the target pits, an examination shows the marl in laminations. This condition might be expected if the soil were basketed and deposited by dumping successive

baskets during its construction.

A camp site was located in a field above

the right bank of Bull Creek a short distance from

its confluence with the Chattahoochee River.

Because of the dense growth of grass few

20

In relation to the river the mound occupies a site on axis lines of the river so that a view can be had from it south southeast about two miles, and west northwest for about the same distance. It is possible that the Indian was never concerned with it, but the location with relation to the river and the unnatural shape of the plain has not been
explained to me (Mulvihill 1925:38).
The results of the investigations by Ashley and Schnell follow.
There is a small mound on the right bank of the Chattahoochee south of Columbus on the rifle range. This is listed by Mulvihill as Site 5, p. 38, and is described as a "conical mound" possibly of Indian origin. We made two trips, August 24th and 27th, out to the range and on the second visit tested the mound. This was to verify the report that it had been constructed as a rifle butt during the Spanish-American War, the earth coming from the target pit that is at its base. A test pit was dug on the summit and several test holes made at different points on the side. The dirt was uniform red clay with some broken stones. No pottery or flints were found or any other signs to indicate Indian occupation. Because of the sites on Bull Creek being so near, it is quite natural that one would take the mound to be of Indian origin. I am
satisfied that it is not (Ashley and Schnell 1928:15).
During the same period, Ashley and Schnell examined Kyle's Mound, located approximately 4 Ian south of the Bull Creek site. This large site complex, which includes Kyles's Mound and the Cooper site (9Me2 and 9Me3) dates to the Mississippian and historic Creek periods and has produced substantial amounts of pottery contemporaneous with the Bull Creek site occupation (UGA and Columbus Museum site collections). Ashley found the mound entirely washed away by the river and the village site was being seriously eroded. Several days were spent digging, resulting in the conclusion that the village area contained European trade material and postdated the mound (Ashley and Schnell 1928:13). The researchers concluded that the village site at Kyle's landing was Benjamin Marshall's trading post (Ashley and Schnell 1928:14).

Early Investigations
It wasn't until 1936 that the site was again brought to the attention of archaeologists, through the efforts of Isabel Patterson as part of a survey of the Chattahoochee valley. As a setting to the following chapter, the events leading to the site's rediscovery will be examined.
Archeology became intimately associated with the federal relief programs of the early 1930s as a means of employment for large numbers of individuals. Dr. A.R. Kelly, the federally-funded archeologist stationed in Macon, Georgia, coordinated these programs throughout the state through various sponsors. Isabel Garrard Patterson became the sponsor for the Columbus area. Patterson's involvement is detailed in a series of letters written by Kelly and Patterson, which have been compiled as a separate volume (Ledbetter 1995b). Passages that relate directly to Bull Creek are excerpted here. The first identified correspondence from Kelly to Patterson, dated August 26th, 1934, discusses the means by which an archeological project might be instituted in Columbus:
Finally, I am aware that Columbus is the locus of an area in which a number of important documented historic sites belonging to the historic removal of the lower Creeks are found. Particularly do I wish to see the site of the very large Coweta town located near the present Columbus on many of the early maps. On the Chattahoochee, somewhat further removed from Columbus, should be identified the site of a Kashita village, possibly two of them.
Within the week I shall present to Miss Shepparson in Atlanta a rather expansive program of archaeological and historic research. It does not seem wise to deluge her with too many separate projects. What is granted by way of specific local research will largely depend upon the interest taken by the communities involved, their willingness to contribute a small sum of money to defray additional expenses, the labor supply and work relief needs of the area, and the changing set up and purposes to be achieved under the new program of work relief planned for this fall and winter.
21

The Bull Creek Site
At Milledgeville, for instance, the local people are much interested in having a scientific field party explore mounds and historic village sites {Old Oconee Town} near that city. When they have organized local sentiment, provided a sinking fund of two or three hundred dollars to take care of incidental expenses not provided for by government disbursements, and have presented their requests vigorously to Atlanta through their divisional engineer under the usual GERA procedure, it is probable that Miss Shepparson will ask me to investigate, determine the scientific value of the projected work and make recommendations to her office. If they have something of real interest and give substantial indications that they want archaeological work done, it is very probable that a project will be initiated.
That was the way things were handled at Dublin. The Dublin people are much interested and have been for some months. The local D.A.R. and business men are combining to set up a sinking fund to take care of expenses not provided for under the present GERA set up.
I would suggest that you appraise the situation at Columbus, do some discreet "politicing", work with your district and divisional engineers for a specific project to explore the historic Coweta town near Columbus. This seems to me to be the procedure most calculated to get more immediate results. Also it is a plan which will fit in with a program already initiated for the whole state.
I have read most of your articles in the Constitution. They are admirable. I am hoping that another year of expanded research and dignified publicity may result in some permanent organization to carryon such work in this area
(Ledbetter 1995b:1).
A month later, a second letter dated September 28, 1934 was sent to Mrs. Patterson.
The Columbus situation will probably rest until you get back from your trip. Dean Smith of Wesleyan told me yesterday that he has an older brother living on the Chattahoochee who had been interested in a scholarly way in historic Creek towns on that river. He wants to do a bit of reconnaissance. Something may come out of this
(Ledbetter 1995b:4).
22

The steps toward implementation of a Chattahoochee valley survey were spelled out in a letter from Kelly to Patterson dated Christmas day, 1935.
The Chattahoochee Survey is practicable. I would suggest a short range project involving the whole Georgia span of the river and its immediate tributary streams, concentrating on the towns mentioned in history for special reference and exploration. Time, four months; from February to June---possibly to July, depending on the life of the WPA. Supervision: James A. Ford. The project should be written for labor only, to be supplied locally as shovel men are needed on various sites in the different counties. Expenses to be funded locally interested parties need not amount to more than $1250, as follows: 1) Ford's salary underwritten for four months, $150 a month---$600; an assistant---he need not be an archeologist---$250 for the period, about $62.40 per month, oil and gas allowance per month, $75, making $300 altogether; miscellaneous items of expense, including photograph and necessary materials, paper, etc. $100. Grand total, $1250.
If it proves desirable, and the time limit of WPA projects allow, the survey might be carried to July for an additional month of survey for $250, making a total for 5 month survey of $1500. All collections should remain in Georgia, with the proviso that they be properly housed; preferably at the University of Georgia; with a possible proviso that representative samples of materials be sent as a duplicate collection to the National Museum. Since you will be drawing heavily upon Swanton and Setzler, I would suggest this proviso as a matter of policy.
Mr. Woodruff and Mr. Peabody are both interested in the University of Georgia and in the Chattahoochee. They might be willing to underwrite the project for the amount named if you put the matter to them along the lines suggested.
Ford would employ the survey methods used to advantage by him in the lower Mississippi. He would sink small test pits in the village sites; might do some exploratory work in sinking pits on sites where village sites were known or expected to be on basis of historical descriptions, but where river silting had buried the remains two feet or more. Pottery and flint, thus obtained, would be of incalculable value, in working out a tentative archaeological chronology for the Chattahoochee. Later, as time, funds, and proper institutional

guidance permitted, large scale work might be done on key sites as that at Kolomoki. Ford's work would give the archaeologist of the immediate future something to shoot at, a peg on which to set out details gleaned from more intensive work on single sites.
No large number of workmen would be required, perhaps 15-20, in some cases as many as 50; depending on the size of the site, its assumed importance, and the amount of overburden to be removed. Ford will be moving rapidly from site to site, going downstream. He would have to have some flexible arrangement by which the required number of men could be supplied him from WPA rolls in the different counties. I would say, however, that you are not entirely dependent upon WPA for the success of your program. For the same amount of money, $1250, Ford, with one assistant, could carry out a four months survey which would yield valuable results. His energy and drive are tremendous when his enthusiasm is aroused.
Please advise me what you think of this prospectus. Ford is young, his enthusiasm is infectious; he is likeable. I think he would do a great deal to develop interest in archaeology in your section if the plan outlined did materialize. And I think you realize, as I do, that Georgia archaeology will flourish only when interest becomes more general and is not confined to a single locality. That is meant in no way to belittle what has been accomplished at Macon by the inspiration of a few individuals. Put that way, I believe the General and the Doctor and Lint would
agree with me (Ledbetter 1995b:9).
As the time approached for the Chattahoochee work, Kelly was forced to scramble to find a field director as Ford was transferred elsewhere. The problem is mentioned in a letter to Patterson dated February 19, 1936.
I am not forgetting that the primary person to satisfy in the Chattahoochee is you as sponsor. You were willing to have either Ford or Hanna. For the same amount of money we can probably get an experienced field man in archaeology. Possibly Frank Setzler can help us out of our difficulties with a suggestion.
I have only two other men in my organization whom I would trust with the responsibility of helping you in your Chattahoochee

Early Investigations
work, both engineers whom I have trained in field methods during the past two years, Frank Lester and Joe Tamplin. They are thoroughly acquainted with all the techniques of excavation, profile reading and recording, draughting, mapping, cataloguing required in the projected work at or near Columbus. Moreover, they know the ropes in handling NYA and other relief workers, and in making contact with WPA officers. I would sorely miss either of these men but am ready to send you either one you might want if you would rather have a Georgia man or someone local than to bring in someone from the outside (which latter procedure might mean more delay).
It would not be necessary to pay them, either of them, as much as $1 50 a month. If acceptable as assistants to you, and I am ready to recommend them without any doubts, $30 a week would be adequate. Transportation was an item of extra expense to be considered in the field budget whether Ford, Hanna, or anyone else was assigned to the job. I rather hoped that you would be able to find support for your plans in handling that factor.
Frank Lester went to the University of Tennessee. Joe Tamplin is a graduate of Georgia Tech P.S. You probably remember Lester, engineer in charge of Mound D a nice clean-cut
youngster (Ledbetter 1995b: 12).
For nearly two years, Patterson communicated with Kelly and others in an attempt to bring a major archeological project to Columbus. During that time, Patterson backed some level of local site survey, thus providing both Kelly and John Swanton of the Smithsonian Institution with sherd collections. Because several large historic Indian villages existed in the Columbus area, it logically followed that one of these towns be selected for excavation. This all changed in early March 1936, with Patterson's discoveries at the Bull Creek site. Patterson's letter describing the discovery is dated March 9, 1936. Figure 13 shows a portion of a contemporary map with points of reference described in the following letter. A map originally prepared by Patterson has not survived.
23

The Bull Creek Site
My Dear Dr. Kelly: Yesterday afternoon I found the road by the Columbus Air Port was dry enough to get over in my car. It was too muddy and really impassable when Frank was here, so I could not show him one of my mainest treasure finds -
A house in a mound! There is about a foot of top soil then a layer of about 2 inches of black dirt. Just below this are wood timbers, which seem to slant a bit; then at intervals of some several feet are upright wood pieces. In absolute conjunction with this wood are fragments of pottery sticking out of the bank. It has been dug into some and I found a bit of wood loose and about to fall, also a piece of pottery right by it just hanging in loose dirt, so I am sending it for you to see. Please then send it to Frank Setzler with this letter.
This mound seems to have been a house site of some kind - and there is a good portion of it undisturbed - but I fear people are slipping down there and doing some private digging. It is on a corporations property - on the bank of the Chattahoochee - 411 feet north of the mouth of Bull Creek - I stepped it off - (There were 274 of my paces, which measure 18 inches). Height of mound about 10 or 12 feet. I drove to the edge of Weracoba Creek which joins Bull Creek just before it reaches the river.

Weracoba is exactly one mile south of the Air Port Hanger (car mileage 817.5 creek - 818.5 hanger).
There is a large field - full of broken pottery - where plows have thrown it up - which extends from a flat just north of Weracoba, along to its confluence with Bull Creek and on to the river. There are several large bumps of dirt, full of pottery, that may be mounds in this field; - the mound by the river with timbers exposed - is just opposite this field and about 1/1 0 of a mile from the creek - Weracoba. (817.5 creek Weracoba along the new road 817.6 point opposite mounds).
Just opposite the Hangers, where a city golf course is being constructed a mound was leveled. It looked high to me in past years, now it is about even with the Roadway by the Hanger. Some smart city officials we have. Yesterday, I collected some more sherds along here on a site covering 2 tenths of a mile - mileage as measured from Weracoba CreekWeracoba Creek . . . . . . . . . . . . . . . . .. 81 7-5 Point opposite mounds and field of pottery 817-6 Rough farm road to mounds on side toward Chattahoochee River . . . . . . . . . . . . . .. 81 8-0 Old mound (now leveled) but field of pottery on a high flat - opposite the Columbus Air Port Hangers . . . . . . . . . . . . . . . . . . 81 8-3 to 81 8-5

STREET MAP
OF
CCOJLUMIB3U
GEORGIA
o"",-==-::-===,;;;3000 Feet
OFFICE OF CITY ENGINEER ADOPTED BY THE CITY COMMISSION. JAN.l0.1928

1, WeracobaCreek 2. Point OpoositeMoundsand Pottery Field 3. Road To Mounds
4. Pottery on High Flat

Figure 13. Points of reference described in Patterson letter dated March 9, 1936. The reference points are shown on a redrafted portion of a city map of the era.
24

Jim Woodruff got hurt yesterday sprained his knee. He had expected to go to see Dr. Peabody today, but that has to be delayed - so I am arranging to get the money myself. How soon can Mr. Frank Lester come over? I am all set now. Can't you and Mrs. Kelly, Dr. and Mrs. Griffin come over this Saturday? I would like Lester even before that if possible. I have just got permission to go into this Mound that I am sending the wood and the pottery from - and that winds up all the land on each side of the River from Columbus through Fort Benning and from Girard Ala. to the Shagee Creek below Oswickee Blvd. in Alabama. May I hear from you as soon as possible
(Ledbetter 1995b: 14).

Early Investigations
According to Patterson's letter, Frank Lester (Figures 14 and 15) had been in Columbus just prior to the Bull Creek site discoveries. Lester returned to Columbus within a few days and began excavations. Lester's report, which will be covered in the following chapter, describes the excavations at Bull Creek, but little else, particularly with respect to the ongoing "Chattahoochee Survey." Two newspaper articles published in April 1936, provide some details.

Figure 14. Ca 1936 photograph of individuals associated with federal relief';;~cheology. Photograph at Macon
earth lodge shows Kelly, center, and Lester, far right (photograph courtesy Columbus Museum).
25

The Bull Creek Site
Close-up otprevious photograpllSb.6wing view of Frank Lester.
26

Early Investigations

Valuable Finds Spur Interest In Archaeological Explorations In Famed Chattahoochee Valley

Indian Village Site Being Explored and Remains of House Located As Noted Anthropologist Points to Wonderful Possibilities in Area by Isabel Garrard Patterson

The recent visit of Dr. Frank M. Setzler to Columbus has aroused a keen interest in the wonderful possibilities for
scientific excavation in this section of the state.
Dr. Setzler is the acting head curator of the Department of Anthropology of the National Museum at Washington. His opinion is most valuable for he has had years of experience in "field work" in several sections of the United
States and has contributed much information to his chosen profession. He has been promoted
rapidly to his present position on account of his ability and his splendid work which has national recognition.
Dr. Setzler inspected a number of sites in the Chattahoochee Valley during the latter part of February. He has been greatly interested in this locality
for it is significant that any attempt to develop a
systematic procedure for archaeology in the southeastern United States must begin with the knowledge of the locations and movements of the Indian tribes in historic times.
"The present and future archaeological investigations in Georgia must depend to a large extent upon the results obtained from the excavation of selected sites in the Chattahoochee valley. The historically docu-
mented sites in and around Columbus provided they contain sufficient archaeological evidence will enable the anthropologist to reconstruct the aboriginal history of Geor-
gia." This quotation is
made from a letter received from Dr. Setzler, dated March 7, 1936.

Very Favorable Report A very favorable
report coupled with the recommendation started immediately has resulted in the actual beginning of this very important work.
The first site selected for study is just north of Bull Creek on the banks of the Chattahoochee River.
The middle of March, Mr. Frank E. Lester came to Columbus from Macon to conduct scientific exploration and make a survey according to approved methods.
Mr. Lester has been at work at Macon since the beginnings of the project which was inaugurated in December 1933. He is a native Georgian and feels unquestionable pride that all have in this particular section of the United States.
He has been in charge of the very important excavation of the now famous "Council House" of "Old Ocmulgee Fields" which is also recorded as "Mound D" of the Macon group.
Although there are some 800 known mounds in the state of Georgia work has been concentrated at Macon because of paramount interest of persons in this locality.
A National park has been secured for the region termed Old Ocmulgee fields. It is the largest archaeological expedition ever conducted in America.
Equipped for Survey Mr. Lester is
thoroughly acquainted with all the techniques of excavation, profile reading, recording, draughting, mapping and cataloguing required for an accredited survey and proper exploration of ancient village sites and

mounds. Mr. Lester has
been the engineer in charge of the work in Macon and Columbus is extremely fortunate in having been able to secure his services. He is highly recommended by Dr. Setzler and by Dr. Arthur R. Kelly who is the head of the archaeological excavation around Macon.
Last Saturday Dr. Kelly, Mr. Linton Solomon, Mr. Hugh Hanna and Mr. Antonio Waring motored over from Macon to see the work that is being done in the Chattahoochee valley; concentrating on a house site that has been discovered in a village site on Bull Creek.
These men were surprised and greatly impressed by the wealth of material which has been located.
Value of Survey The approved
methods of archaeological survey might seem tediously slow to those who are unaware of the necessity for comprehensive reports of preliminary work. It is necessary however that such information be obtained in order to facilitate the expediency of more intensive exploration on spots which prove to be the key sites by information gathered on the survey.
The site selected is laid off in squares of convenient size so that definite measurements can be made of all materials found in such squares and properly recorded according to their depth and placement with exact relationship to other such material definitely associated with them.
As work progresses, photographs and exact cataloguing of materials and engineering

records are taken along with archaeological interpretations which are placed on such material. Therefore such work properly done can be of untold value for comparative studies by scientists in other sections of the country whether they have visited the particular site in question or not.
Exploration in Valley Exploration of
this nature has been started in the Chattahoochee valley. A village site is being explored and the remains of a house has been located. Excavated portions of this house reveal a hard-packed clay floor with clay plastered timbers for walls and roof. The briquettes show use of grass or other vegetable fibers mixed with the clay to give better binding qualities. Some of these briquettes have been found mounded around sections of charred timber supports and others show lines caused by the lashings which held these timbers in place. Fragments of pottery remain on the floor showing definite signs of human occupancy at the time that the house was destroyed. Beyond the immediate confines of the house there is a very rich midden deposit indicating comparative extensive occupation.
Such materials as bits of bone flint, pottery and those botanical specimens as have been able to withstand the ravages of time all have their proper place in reconstructing the story of this aboriginal occupation.
(Columbus Ledger Enquirer April 5, 1936.

27

The Bull Creek Site

survey included the territory adjacent to

Indian Relics to be Sought in this Area

the Chattahoochee from the famous Coweta falls to the mouth of the Flint River

at the southwestern corner of the State of

Jaycees Will Sponsor Archaeological Survey in Bull Creek Section

Georgia. Each scene of aboriginal encampment was mapped, collections of potsherds and flints obtained and

The board of directors of the Junior chamber of commerce, preceding a meet-

Many think relics are beneath the ground in this vicinity where a burial ground

catalogued for study. It immediately became apparent that a number of the sites visited could not be identified as a

ing of the membership of the and an Indian village are

location of the tribes who occupied this

body last night, voted to thought to have been located.

territory within historic times, so the sites

sponsor an archaeological Officials of the Jaycees hope

of unknown age have been designated by

survey for indian relics along Bull Creek between Columbus

to uncover many valuable relics and at the same time

their present geographical boundaries.

and Fort Benning.
Mr. Lester, archaeological engineer, will conduct the survey for the Jaycees. If the survey shows evidence of man relics in the section, an effort will be made to have the Smithsonian Institution conduct excavations.

learn a great deal about the early life of the people.
Excavation Planned Mr. Lester will make
a number of test excavations. He has already done some preliminary work along the banks of the creek.
(Columbus Ledger April 24, 1936)

The most remarkable of these incidental sites were found in section sixtysix of the historic Coweta Reserve, on a narrow bluff immediately north of the confluence of Bull Creek and the Chattahoochee River, and three miles South of the Court House in Columbus, Georgia. This particular area, so rich in tradition and history is comparatively unexplored and unknown to students of American prehistory, to supplement, therefore, other observations these previously unrecorded

sites are now described as "The Bull Creek

Patterson's column provided the

Village" and the "Bull Creek Cemetery."

citizens of Columbus a "text book" explanation of how scientific archeology was to be conducted at the Bull Creek site. The second

Careful examination of the many local collections obtained from the field adjoining this narrow point of land provided additional infor-

article shows that public support was generated, although the first order of importance was the recovery of valuable relics.

mation for the survey record. The surface material of this field area, now included in "The Bull Creek Village," covers approximately three quarters of a square mile. The northern boundary of this area is

Two Patterson manuscripts provide

the present city limits of Columbus, and the eastern boundary is formed by Weracoba Creek

important background information for the

{see Figure 13 city map}.

WPA excavations. The first is entitled Archeological Survey of the Chattahoochee Valley (patterson 1936), the second Bull Creek Village (Patterson 1937). Excerpts from Patterson's manuscripts endow a quixotic

The collection made by the survey party suggested such a wealth of material that further investigation was warranted and the owners of the land, the Muscogee Trading Company of Columbus, gave permission for this work.

prologue to offset, to some degree, Frank Lester's rather succinct presentation of excavation data which will follow.

A spur railroad was constructed across the western part of the field and extended along the narrow bluff between Bull Creek and the river

An exceptionally interesting Indian site in Muscogee County, Georgia was discovered recently in the course of an archaeological survey of the Chattahoochee Valley. Attention had been increasingly drawn to the significance of the historically documented habitations of the native tribes found in this region and an investigation was made in an attempt to determine the present

by means of a cut through the northern portion and a fill toward the southern. The trestle crossed over the mouth of the creek. The remains of a railroad cut had left exposed an occupational level and evidences of a house site. Pieces of pottery and timbers were found protruding from a dark layer of earth about 12 inches below the present ground level, on March 9, 1936.

surface indications of these sites. The preliminary

28

Early Investigations

The former inhabitants had a fine view of the Chattahoochee and the low terraces along its western bank where the historically famous Coweta Town was located.
Mr. Frank E. Lester, who had assisted Dr. A.R. Kelly in the exploration of Old Ocmulgee Fields near Macon, came to Columbus on the eleventh of March to direct the excavations at Bull Creek. Work started the following day with a small crew.
As the work progressed, local people became more interested in the site, and, fortunately, a number reported finds in this area. Probably the most amazing specimen produced is a spear head, found by Bentley Chappell, Jr., that has been identified by Dr. Frank H.H. Roberts, Jr. {Figure 16}. This generalized southeastern type of Folsomoid material was picked up in the field about four hundred feet northeast of House Site #1 of the Bull Creek Village
(Patterson 1936).

The approximately 11,000

year old Paleoindian point shown in

Figure 16 was the subject of several

letters between Patterson and Kelly in 1935 (Ledbetter 1995b:5) although there is no indication that Bull Creek

Figure 16. Drawing of fluted point from Bull Creek (by Bettye J. Broyles).

was ever viewed as a potentially important Paleoindian site. Patterson viewed the find as highly significant and indicated that a

Seaborn Jones and S.M. Ingersoll established "a new ferry" in June 1831, one mile below the town beyond the South Common. The distant scene

search for similar material led to her discovery of the burned structure at Bull Creek (patterson 1950:37). Patterson's Bull CreekVillage manuscript, which is actually a photographic journal, details the discovery.

beyond the roadway on the right shows "Shepherd's Plantation Gardens" now partly occupied by the landing field of the Municipal Air Port, and Guyner's hill against the horizon, where a residential section began in the late thirties. About 1890, the family of Edward Thompson Shepherd

decided to erect a brick plant in their gardens lying

The first evidences of an Early American house, approximately three miles below the "Falls of the Chattahoochee" and four hundred and fifty feet north of the mouth of Bull Creek were found by Isabel Garrard Patterson, March 8, 1936 (Figure 17). She is indicating these signs of habitation in the accompanying photograph to Myra Bullard Hart and calling it the "Bull Creek Village." Trees in the left background of this photograph mark the remains of an old trail which is still discernable, and where according to written documents

along Weracoba and Bull Creeks, and in 1920 they sold this property to the Columbus Brick and Tile Company. In 1925, this company employed engineers to survey the western part of this area with the purpose of constructing a spur railroad and a trestle across the mouth of Bull Creek to transport materials from the clay deposits south of this creek to their plant. Robert Wiley Smith was in charge of these operations and, in 1926, the rails occupied the site of the present road shown in the photographic record.

29

The Bull Creek Site
':;.,
Figure 17. 1936 photograph of Isabel Patterson at the location of House Site 1 at Bull Creek, view to the NW (Columbus Museum files).
Figure 18. 1936 photograph of IsabelPatterson at House Site 1, view to south towardBull' Creek Cemetery (Columbus Museum files). 30

The pine forest in the left background of this {second} picture {Figure 18} is on the site of Ben Marshall's Trading Post on the south side of Bull Creek and exactly opposite the principal structures of the historic "Coweta Towne" of 1739 on the right bank of the Chattahoochee. The Bull Creek Cemetery was found May 3, 1936 by Alfred Owen Blackman II about four hundred feet south of House Site No. 1 {in the right foreground of the photograph}. The site of the cemetery is not clearly visible, but can be located by the sycamore tree without its leaves left of the roadway, outlining the edge of this picture. The spur railroad constructed in 1926 was abandoned in 1929 as it became unsafe due to flood waters and soil erosion on this high peninsular between the last bend of Bull Creek and the Chattahoochee. The cut through the high bluff in the foreground is approximately fifty feet above the normal water line of the river, and here the road runs straight but across the recently filled land the path extends in undulating curves for over two hundred feet, then proceeds in a straight line to the supports of the former trestle across the creek. The earth borrowed from the high bluff on right and some immediately left of the roadway was used to make the fill between this point and a lower bluff at the mouth of Bull Creek. The two bluffs were called "Indian Mounds," but the Brick and Tile Company

Early Investigations
of the Twentieth Century were not impressed by the fragments of ceramic art which they disturbed. Scattered bits of pottery almost pave the roadway today, but pieces of other vessels found protruding from a dark layer of earth in the embankment twelve inches below the present level of the bluff were the first indication that a part of this house remained unharmed for study. Charred timber and hardened clay were seen during a closer inspection and these offered substantial evidence of Indian occupation.
This {final} scene shows the site of the great "Coweta Town" across the Chattahoochee from the prehistoric sites at Bull Creek {Figure 19}. The trees in the background are on the Alabama side of the river and those with bare branches closer to high bluff as well as the one at the left of this picture are on Georgia's banks. The elevation of bluff above existing roadway is approximately seven feet, and a section of the "borrow pit" composes the foreground.
With Patterson's manuscripts providing a historical setting, Lester's report will now be presented in the following chapter.

Figure 19. 1936 photograph of Isabel Patterson in borrow pit near House Site 1, view to the west toward Alabama (Columbus Museum files).
31

THIS PAGE LEFT BLANK

Chapter 3 WPA Excavations of 1936 and 1937

As a 1930s public works project, a substantial amount of time and manpower was expended on the field work at Bull Creek and subsequent laboratory procedures, such as cataloging of the artifacts and transcribing field notes and records. As a result of this painstaking work, a substantial body of information has been preserved from these six-decade-old excavations. Of course, some of the paperwork has deteriorated over the years and there are problems involved in interpretation of the field and laboratory techniques of the period, but overall, the surviving documentation provides an excellent record of the excavations at Bull Creek. Fortunately Frank Lester, prepared a plan map that shows the areas of site excavation. This map provides the basis for interpretation of the WPA-era excavations at Bull Creek {Figure 20}.
Isabel Patterson's letter of March 9, 1936 to A.R. Kelly, which described her discovery of an aboriginal structure within a presumed mound at the Bull Creek site, brought almost immediate results. In less than a week Frank Lester had arrived and field work commenced. Lester's excavations began in mid-March 1936 and were to continue until early April 1937. During that year Lester supervised groups of children and a number of Muscogee County men in extensive excavations.
Following field work, Lester prepared a manuscript entitled a Preliminary Report of Excavation at Bull Creek Village Site, Muscogee County, Georgia (Lester 1938). The report is one part of a compilation of documents transcribed by National Park Service (NPS) personnel. The documents are sequentially numbered from page 27 to 93. Accord-

ing to a handwritten note attached to the manuscript, the first 26 pages have been misplaced and the original contents are unknown. The main sections of the manuscript include the preliminary report (pages 27-44), transcribed field notes (pages 45-74) and various lists and compilations of artifact and burial data (pages 74-93). The exact date of Lester's manuscript is undetermined but it probably was written in the late 1930s. Copies of the manuscript, curated today in the archeological files of the Columbus Museum, have been made available for this report. The manuscript will be referenced as dating to the year 1938, which is our best approximation of the date of completion. A copy of Lester's field notes is present in Appendix A.
Lester's plan map (see Figure 20) shows two main areas of excavation at Bull Creek. The northern area, which lies just west of a large borrow pit, represents the village excavations where two Mississippian period structures were excavated. Examples of photographs taken in 1936 of general excavation in the village area are illustrated in Figures 21 and 22. Relatively few photographs of the village area excavations exist. Other photographs, which detail the structure excavations, will accompany the text of Lester's report.
The second area of excavation was designated on Lester's map as the burial ground. This is the famous Bull Creek cemetery from which three dog effigy vessels were excavated. Figures 23 and 24 present photographs of the cemetery area excavations. A reconstructed cemetery map showing the distribution of burials and excavation boundaries is shown in Figure 25. The cemetery map has been reconstructed from Lester's field records.

33

The Bull Creek Site
\,
\

\
\
,\
Ii \
\\

1)

SKETCH MAP

Ii
II I !

BULL CREEK EXCAVATION

I

Figure 20. Redrafted copy of Lester's 1936 excavation plan map of the Bull Creek.

34

WPA Excavations
Figure 21. Field photograph of excavations in the village area (NPS photograph 2367). 35

The Bull Creek Site Figure 22. Two field photographs of village area excavations (NPS photographs 2381 and 2370).

WPA Collections Figure 23. Two field photographs of the cemetery area excavations (NPS photograph 2491 and 2414).

The Bull Creek Site
Figure 24. Field photograph of initial trench exca.vation in the
. cemetery area at Bull Creek (NPS photograph 2488). 38

fl .F .

,./.
~
IJ iii

WPA Collections
4-27 4-28
. . 4-25 4-26
I( ., ~"
"0
. . 4-21 4-22

(?)
"$:r '(j~..:::.1 ~
4-1 I 4-12

/

BULL CREEK CEMETERY
,.

ON
A0 0 / /

2m 5 Feel
/

L33 LJI L29 L27 L25 L23 L21 LI9 LI7 LI5 LI3 LII

L9

L7

L5

L3

LI

4-1 4-2 RI

R3

Figure 25. Cemetery map of the Bull Creek site based on Frank Lester's field records.

The Bull Creek Site
Lester excavated Bull Creek using 5 x 5 ft squares, which was a simple and straightforward means of site excavation. However, the stake and grid designations utilized by Lester differ substantially from the standard bicoordinate system of today and requires some explanation. Lester did not utilize a comprehensive site-wide grid. A series of 5 ft-wide trenches, designated Controls 1 through 4, were excavated and subsequent expansions were tied to these trenches. At Bull Creek, Control Trenches 1 and 2 were excavated in the village area and Control Trench 4 was excavated in the cemetery. The location of Control 3, if it ever existed, has not been determined.
Trenches were placed in a generally northerly orientation (grid north) but were not aligned with magnetic north. According to Lester's field notes, the control trenches initially excavated in the northern area were parallel and lay on a line 15 degrees west of magnetic north. In the cemetery area, grid north appears to be approximately 45 degrees west of magnetic north. The extreme divergence from magnetic north in the cemetery area causes problems when trying to interpret Lester's text with respect to burial positions and orientations.
Each control trench was staked at 5-ft intervals, forming a series of individual 5-ft squares. Stake locations were designated first by a Control Trench prefix number (1,2,3 or 4), followed by an individual stake number. Number 1 was assigned to the stake in the extreme southwest corner of each trench. Odd numbers were assigned to stakes along the western line and even numbers were assigned to stakes along the eastern line. As an example, the first square in Control Trench 1 (southwest corner stake) would contain stake number 1-1. The designations for the remaining stakes would be 1-2 in the southeast corner, 1-3 in the northwest corner and 1-4 in the northeast corner.
40

Lester's system became more cumbersome as the trenches were expanded to the east or west. Moving to the east, each 5 ft stake interval was designated by the letter R followed by an odd number (Rl for 5 ft east, R3 for 10 ft east, R5 for 15 ft east, etc.). Moving west, stakes were designated Ll for 5 ft west, L3 for 10 ft west, L5 for 15 ft west, etc.). A fully designated stake number to the east of a control trench would encompass the coordinates of the eastern line of that trench plus the R number (for example 1-6R3). To the west, the control trench stake from the western line would be used with an L number (for example 1-5L3). In this example, the actual distance between the two points represents the distance of R3 to the trench (15 ft), the width of the trench (5 ft), and the distance to L3 (15 ft), for a total of 35 ft. In actual practice, the locations of objects or features found during excavation were identified by measuring from the nearest stake. For example, an object might be assigned a location of 20 inches south and 15 inches west of stake 1-5L3.
Essentially, each 5 x 5 ft square was an individual excavation unit and individual Control Trenches existed independently from the rest of the site. This would raise serious problems for interpretation unless a field map was prepared which shows the exact locations. In the case of Bull Creek, the less than complete map shown in Figure 1 does exist. Most of the squares which were actually excavated can be determined through examination of field notes. Certain relationships, such as the distance between the village excavations and cemetery excavations must be taken at face value as depicted on the field map. One point is clear, Lester's map {see Figure 20} is not a layout of the completed excavations. The map appears to be a plan for excavation which shows the grid orientations of the two excavation areas. By following Lester's field notes most of the actual area of excavation can be recreated. This is one goal to be addressed in this chapter.

Artifacts and notes from Bull Creek were treated in the same manner as those from other sites excavated during the WPA period. Material was transported to the regional laboratory in Macon where a staff of technicians carefully washed and individually numbered each specimen (Figure 26). Artifacts were cataloged on individual provenience cards. All notes were transcribed in triplicate and all photographs individually numbered. Artifacts, such as aboriginal pottery, were subjected to detailed attribute analysis (Fairbanks 1940:27).
Lester's preliminary report, which concentrates on the cemetery excavations and the two structures excavated in the village area, is a condensed version of his field notes. Greater detail is found in his field notes, which have been transcribed and included in the appendices volume of this report. Further insight into Lester's excavations is gained through examination of a series of manuscripts

WPA Excavations
prepared by Isabel Patterson. Patterson's record of the excavations include a report in Early Georgia, the publication of the Society for Georgia Archaeology (Patterson 1950), and several scrapbooks, photographic journals and manuscripts which detail various aspects of the WPA-era excavations. In addition to the published report, sections of three Patterson manuscripts will be cited. These manuscripts are titled Archeological Survey of the Chattahoochee Valley (Patterson 1936), Bull Creek Village (Patterson 1937), and Notes on the Exploration of the Bull Creek Site, Columbus, Georgia (Patterson n.d.). The last manuscript is an earlier version of the 1950 paper.
Lester's record of the 1936 excavations at Bull Creek will be presented under two headings, a report of the village area and the cemetery excavations. The bulk of the information is derived from Lester's preliminary report, supplemented by references to field notes and observations by Patterson.

Figure 26. Circa 1940 photograph of laboratory workers at Macon cataloging artifacts from WPA excavations (from Fairbanks 1940:30).
41

The Bull Creek Site
Lester's report will be presented as originally written with minor editorial changes. These changes include standardization of burial designations and deletion of a few specific references to photographs, maps, or drawings that have been removed from the original manuscript. In certain instances, coordinates have been omitted. Photographs and maps that illustrate various aspects of the project have been added to the original text.
Lester's Village Excavations
WPA excavations in the village portion of Bull Creek commenced in mid-March with Mrs. Patterson accompanying Lester and a small crew of workers. The beginnings of the

excavation are detailed in a letter from Lester to Dr. A.R. Kelly dated March 15, 1936. Figure 28 is a sketch map prepared by Lester to illustrate his initial findings.
Dear Dr. Kelly:
Arrived safely but was unable to reach Mrs. Patterson until Thursday morning. We visited the site on Bull Creek first, made a simple sketch map and that afternoon went over in Alabama to a rather large village site just across from Bull Creek.
Mrs. Patterson was mistaken about it {the Bull Creek site} being a mound, as it is just a rather heavy midden deposit about two hundred feet long, covered with a yellow sandy clay silt deposit.

culflVQfed field

~fJO$ed cuf shows occuootion from (f) fo(l). /-louse debris from@ to@.
Bull Cree/( - Muscogee Co., Cia. FE Lesfer - Mar 19.36
Figure 27. Redrafted sketch map of a portion of the Bull Creek site showing the area of "house midden" (copy of original 1936 map by Frank Lester).
42

WPA Excavations

The house debris is almost in the center of the deposit. All this is exposed in the side of the cut made for a railroad a few years ago. External appearances, very similar to Mossy Oak.
Started excavations Friday (the 13th) with a crew of three negro boys and myself. So far we have made excellent headway, finding quite a bit of material in the very rich midden deposit. The majority of the sherds are extremely small fragments, plain, and very hard tempered. Yesterday, I started hitting baked clay and quite a number of briquettes showing reed imprints. Also the top of what is possibly a fire bowl, however, future troweling will prove more enlightening. It is doubtful if enough of the house remains to get a very complete floor plan for the river has made rather deep inroads from the west and the railroad cut on the east so that only a small portion from eight to twelve or fourteen feet wide is left. However, Mrs. Patterson was interested in getting it out and also it would not seem such a bad idea to start this group off with something that they can actually see.
Everything is progressing very nicely at present, of course, it seems slow after the large crews at Macon but we are making headway. I hope that it will be possible for you to come over
with Hanna when he comes (Lester 1938).

cultivation for quite a number of years and the occupational level is so close to the surface that all signs have been destroyed except in the section marked on the map {see Figure 27}. The underlying soil in the immediate region is a yellow clay silt with a very high percentage of mica in it (Yellow Congaree Silt). Top soil is ordinary grayish tan sand.
The remains of an old railroad cut has left exposed an occupational level and evidences of a house site, briquettes and charred timbers showing in the side of the cut. This area was staked off in five foot squares and overburden removed by
horizontal method {Figures 28 and 29}.

The first portion of Lester's preliminary report dealing further with the village area is presented in the following paragraphs. The area of initial excavation and the "house midden" deposits exposed along the railroad cut are shown in Figures 20 and 27. The cemetery area had not been discovered when this map was produced.

In March 1936 work was begun on an archaeological survey of the Columbus area of the Chattahoochee Valley through the efforts of Mrs. Wayne Patterson and other locally interested people with the co-operation of the National Youth Administration.

The first site selected for investigation was located on the banks of the Chattahoochee River at the point of confluence of Bull Creek. The site is situated on a high bluff, approximately fifty feet above the normal water line of the river, and the surface material of the village covers approximately three-quarters of a square mile. However, the entire section has been under

Figure 28. Field photograph of iiilroad cut in village area also showing road on borrow pit at right and staked excavations on high ground (NPS photograph 2369).

43

The Bull Creek Site

The briquettes themselves are rather massive in size and show the use of grass, straw and other vegetable fibers as a binding agent. It appears that the interior posts of the house were also stuccoed with briquette materials, a large section of one of the interior posts was removed intact from the floor, briquette materials still in place around it.
Working through briquette material {we} uncovered a floor of hard-packed clay with baked areas, presumably patches in spots. Post moulds are from sixteen to twenty-two inches apart and indicate a square or rectangular house with slightly rounded corners. Interior posts suggest cross beams for holding roof supports. Mould impressions on briquette materials show the use of small poles, one to two inches in diameter, bound together with vines or thongs as cross members {attached drawings are referenced in text but have not been located}.
Lester's field notes reference a prepared raised clay area (curb) which was first thought to be a prepared clay hearth but later thought to define the edge of the structure. Figure 32 shows the feature. An excerpt from Lester's field notes for March 14, 1936 describes the feature.

Figure 29. Field photograph of vil1l:l.g~ eXdlvation and pedestaled grid stakes (NPS photograph 2378).

The upper soil level was of the yellow subsoil thrown up by the excavation of the cut, next was plowed soil from the late historic cultivation with a slight trace of sod development just under it. Very little material was found in any of those levels but immediately below this at a depth of approximately eleven inches below surface is a reddish sandy clay level just above the roof debris.
Subsequent troweling uncovered a mass of moulded briquette material belonging to the wall and roof of the house {Figures 30 and 31}. This material takes a decided rectangular form and indicates the remains of a house site twenty to twenty-five feet across and seven to ten feet wide, other portions having been destroyed by railroad c o n s t r u c t i on.
44

Figure 30. Photograph of "brickett" material (impressed daub) from Structure 1 (Columbus Museum files),

WPA Excavations

Figure 31. Field photograph showing fired clay covering floor of Structure 1 (NPS photograph 2372).

Figure 32. Field photograph of the raised clay "curb" within Structure 1 (NPS photograph 2374).

The Bull Creek Site

A baked clay ridge was exposed, 1" deep in red level, this ridge appears to be floor material, and is shaped something on the order of a fire bowl, except that it has a very small degree of curvature .... Immediately adjacent to the baked ridge mentioned above, are other baked areas appearing at irregular intervals and slightly different levels. It is probable that all of these belong to the
wall and roof structure (Lester 1938:46).

of large sherds (Lester 1938:51, Figure 34). There are also additional references in both Lester's field notes and Patterson's manuscripts relating to the burial urn that raise questions concerning the accuracy of Lester's interpretation presented in the preliminary report.

The text of Lester's preliminary report continues with reference to an unusual vessel.

Field note entries for March 19 and May 22, 1936 state:

A small portion of a burial urn was found on the floor of the house covered with fragments of calcined human bones. This and a few miscellaneous sherds were all that remained on the floor but immediately beyond the confines of the house proper is found a comparatively rich midden accumulation about ten inches thick, also several refuse pits were found close by containing quite a number of sherds, charred animal bones, and broken bits of stone. At the present time the floor proper has not been cut through as there are no indications of floor pits
(Lester 1938:2728).

Started troweling floor section and uncovered a piece of broken pottery, laying flat on the floor, with a few fragments of human bones in it. This is located just on the edge of the bank {road side} 19": S - 16" E of sta. 1-50 (Lester 1938:47).... Also found some small calcined bone on floor of house site between Sta. 1-47 and 1-48 to 1-49 and 1-50. Bones are very tiny and have been burned completely through (Lester 1938:51).
Chattahoochee River
Edgeof BOA/( 50'titrp

This concludes Lester's report on Structure 1. Figure 32 shows Lester's map of the postmolds, fired clay features and one large pit associated with the structure. An examination of Lester's field notes provides further insight into the structure. The large pit adjacent to the structure produced a number

~5dia.
cr:"...

+r-ss

+'-51

+'-ss

,LfOUSE SITENO. t

BULL CREEKEXCAVATION

'*- S'Q'~ 2.5'

July I!JJ6

Borrow Pit

Figure 33. Floor plan of Lester's Structure 1 (redrafted copy of Lester's original drawing).

46

WPA Excavations

~
,.:/.;.
~;:~:// -> ": .

. '.7-:'
. ~: .. :::~
.. ~.:
... -.~
. . ' "

'<;,..<., .".:.

o

3

e-.

i

em

Figure 34. Photographs of two vessel fragments (complicated stamped and plain bowls) from trash pit adjacent to Structure 1 (NPS files).

47

The Bull Creek Site
Interestingly, Lester's reference to the urn does not indicate the bones were calcined while the second entry concerning burned bones does not indicate that they were human. Apparently the bone was not saved. Figure 35 represents a composite drawing of the two areas of bone based upon a sketch from Lester's field notes.
Patterson referenced the burial urn in one manuscript and mistakenly inferred the presence of a cremation (Patterson 1937). The photograph is reproduced in Figure 36 with her caption.
According to field notes, additional work took place around Structure 1 during May 25 and May 26, 1936: Continued work on
#1 house site; more evidence of post mould impressions. Removed moulded briquette from floor of house ... Very little pottery showing up on house floor. Floor still shows hard-packed clay variety, with layer of charcoal just over floor. What posts remain in holes are partially charred, but what fiber remains uncharred is totally rotted.

,/-4r-7-------------------/-i50 I

I

I

I

I

I

I

I

I

I

I

I I I

Calcined Bone

I

IL

~

/-47

/-48

Figure 35. Plan of a portion of Structure 1 showing the location of urn and calcined bone (adapted from field sketch maps).

It appears that the posts charred, all around the outside and the portion still remaining in natural state is the center of the charred portions (Lester 1938:51 ).

Along the eastern edge of the embankment ... a small portion of a burial urn was found on the floor of House Site No. of Bull Creek Village. The broken urn of hand molded clay pottery with pinched rim was covered with fragments of calcined human bones. The larger piece of pottery on the right is the identical one found protruding most prominently from the side of the high bluff above the roadway in the preceding photographs ...The knife at left in this picture shows the approximate size of the broken pieces of the urn (Patterson n.d._:photograph 12)
Figure 36. Field photograph of possible burial urn from Structure 1 (NPS photograph B.C. 12).
48

WPA Excavations

The remains of a second structure were also found at the northern end of Lester's excavation. Lester's preliminary report deals with this second structure only sparingly. Significantly, several smashed vessels were discovered on the floor, providing important data not recovered from the first structure. One factor probably limited the amount of time allotted to the excavation of the second structure. The Bull Creek cemetery was discovered at nearly the same time and apparently diverted attention away from structure excavation. It is unclear just how much of the structure was actually excavated. Lester's conclusion that postmolds were absent probably reduced the interpretive importance of the find and may have been one reason for the limitation of effort expended. Excavations conducted at the site more recently indicated that postmold patterns survive on the site even when other structural remains have been plowed away (see Chapter 7). Fortunately Mrs. Patterson provides additional details. Lester's account of the excavation of Structure 2 follows.
About one hundred and sixty feet north of house site number one the ridge widens out considerably and test trenches were dug in this area. The soil levels remain essentially the same except that this section has been under cultivation for the past few years. The plowed soil contains quantities of sherds and there is a trace of an occupation level immediately below the plowed soil; occupation level runs out between stations 1115L5 and 1-119L5.

evidences of post mould impressions are gone {Figure 37-38}. From the remaining briquette material it appears that wall construction was very similar to the first house but lack of post moulds make any architectural floor plan impossible.
Further exploration of the immediate area failed to yield any information and Control #2 was excavated just across the railroad cut from the above houses. Intensive plowing had destroyed all evidences of an occupation level. Pottery fragments are found mixed with late historic material and one refuse pit was found containing quite a bit of material but otherwise there is no definite sign
of occupation (Lester 1938:29).
Based upon a list of the special finds recorded in the field notes, portions of several restorable vessels were found on the floor of Structure 2. The list included the following vessels:
Find 37. Broken pieces of a pot, found on floor level of Structure 2. Bowl and rim found 42" east and 7" north of station 195Ll. Bottom found 32" east and 14" south of 1-95Ll. Pot approximately 9" in diameter, plain design, with fluted rim.

In this area was found the remains of another house but constant plowing had damaged it so much that it is impossible to get any comprehensible floor pian. Briquette material, several large pots, and fragments of charcoal, resting on what was evidently the floor and small baked areas give some clue as to approximate size but all

~;i~~f~~ili!~[~~~i~~~!~~;\!:j:l\~-:::".=:..:,::::-'.'
Figure 37. Field photograph of three smashed pots on floor of Structure 2 (NPS photograph B.C. 22).

49

The Bull Creek Site
Figure 38. Field photograph showing'Work:erSClearinglhe floor of Structure 2 andfhelocatioDS
of shattered vessels (NPS photograph 2382). 50

Find 38. Pieces of a large pot, notched rim, incised and punctated design on body of pot. Pieces found grouped together on floor level of Structure 2 between stations 1-97L1/L3 and 1-99L1/L3.
Find 39. Pieces of a small pot (approximately 9" diameter) resting on floor level of Structure 2, 7" west and 31" south of station 1-99L3. Pot has notched rim and plain body. Body notched due to firing.
Find 40. Several pieces of a large pot found on floor of Structure 2. Found 24" south and 18" west of station 1-99L3, 13" below surface.
Find 41. Portion of broken rim of large pot (13" dia.). Found resting on floor level of Structure 2, 13" below surface, 8" west of station 1-99L3. Pot shows painted or beaded rim and circular stamped design.

WPA Excavations
Find 42. Portion of rim of large pot (13 " dia.). Found resting on floor level of Structure 2, 13" below surface, 13" south and 18" west of station 1-99LE.
According to National Park Service records, only one of the vessels (Find 38) was restored, although a second restored vessel was found in the collections. The second smaller vessel (Park Service No. 39-22151) apparently was not assigned a Find Number, but is catalogued as found on the floor of Structure 2. Figure 39 shows the smaller restored vessel. Figure 40 shows the partially rereconstructed Find 38, a large bowl which had been restored but has since deteriorated.
A search of collections to relocate other restored Structure 2 vessels was unsuccessful. However, the descriptions suggest these vessels are large bowls similar to Find 38.

cm Figure 39. Modern photograph of restored vessel (ace. no. 39-22181/22151) from Structure 2.
51

The Bull Creek Site

o3
~
em

Figure 40. Photograph of restored vessel (Find 38) from Structure,2 (photograph from files of Columbus Museum).

Lester's field notes provide some additional insight into Structure 2. The entries for May 7-18 include:
Continued work on House Number 2 with very small group of workers. Removed pottery from floor level and continued on down. Ground so hard that necessary to keep wet all the time. Floor appears to be very uneven, sloping toward the southwest. Found baked area in S.W. corner at least 3" below level of that in N.E. corner. Still finding some pottery in floor level and traces of briquette material. Very few post moulds have turned up at the present time. Floor area is very
52

uneven, showing patches of baked material in spots. All indications point to use of hard packed clay floor, with patched places baked in place (Lester 1938:50).
Relatively little information may be gained from examination of Lester's field notes with respect to the results of excavation elsewhere in the village area. Figure 41 shows the total area of Lester's excavations based upon field note entries. These entries are not complete and do not, for instance, define the extent of excavation of Control Trench 2 or the location of Trench Control 3.

WPA Excavations

--
Figure 41. Reconstructed plan of Lester's village excavation based upon field note entries.

Cemetery Excavations
The discovery of an aboriginal cemetery at Bull Creek in 1936, rich in exotic burial offerings rapidly stimulated the interest of Mrs. Patterson, the citizens of Columbus, and, for a brief while, the archeological community. The existence of the cemetery was unknown when Lester's excavation began. The cemetery was buried beneath the railroad fill near the mouth of Bull Creek. Ironically, the construction of the railroad levee which elsewhere destroyed portions of the village had actually helped preserve the cemetery area by capping the deposit and reducing the impacts of erosion and looting. Figure 42 shows the depth of excavated soil in the cemetery area.

Figure 42. Field photograph showing excavation of the cemetery, view to the north. Note the location of road in upper left of photograph in relationship to the excavation line (NPS photograph 2485).
53

The Bull Creek Site Lester's preliminary report provides
some information on each of the burials excavated at Bull Creek. Unfortunately, most burial descriptions are minimal. The field notes contain more information, especially in the form of field sketches of many, but not all, of the burials. Figure 43 illustrates examples of field drawings. Photographs exist for a
1-/3

large percentage of the burials but many photographs are of poor quality. As one facet of the present study, the burial remains, curated at Columbus Museum, were examined by Chad Braley but an exhaustive osteological analysis was not conducted.

I, .. -.
Loca.f;~n aur;'~( :# S i

~, 1fJ ~\

._.~ litt,

J

~

b ~\'
" ............&r

~

"C"

~

~

~~_I,!II~

\Q
~

,,
!Y
t

Figure 43. Four examples of burial field drawings from Bull Creek cemetery (redrafted from field notes). 54

WPA Excavations

The composite map of the Bull Creek cemetery (see Figure 25) is not based upon an original field drawing. The cemetery map represents our redrafting of a map in the Park Service files which appears to have been produced using the individual burial sketch maps in Lester's field notes. There is no indication that a complete cemetery map was ever produced by Lester. As previously noted, there are problems concerning the burial orientations furnished in Lester's report. Based upon the Bull Creek base map, supposedly produced by Lester (see Figure 20), grid north for the cemetery actually lies nearly 45 degrees west of magnetic north. This alignment appears to be confirmed by the photograph shown in Figure 42. In that photograph, the line forming the boundary of the cemetery excavations intersects the roadway of the old railroad corridor at a noticeable angle. If this point is correct, all references to orientation should be adjusted by approximately 45 degrees (Figure 44). The text of Lester's report uses grid north bearings for burial descriptions.

bones sticking out of the bank and in great excitement reported their find to Mr. Lester. Naturally the search for other evidence of skeletal remains commenced on May 6, 1936. Pits were carefully excavated by the more expert of the N.Y.A. group of workers in this area (Figure 45) although some were left to complete the work on the House Sites of "Bull Creek Village" ... The displaced skull was listed as Burial #1 even though it was removed from position by the person who
found it (Patterson n.d.).
Lester's preliminary report provides fairly detailed descriptions of many of the burials. Contemporary photographs and sketches, sections of the cemetery plan map, artifact illustrations, and some details from Patterson's manuscripts are inserted into the text of Lester's report.

The events leading to the discovery of the cemetery are not mentioned in Lester's report but are recounted in detail in Patterson's writings. The following excerpts are from Patterson's original version of her 1950 paper.

About four hundred feet south of House Site #1 on the bank of Bull Creek two burials were found about 100 yards from the creek's confluence with the Chattahoochee. Mr. Alfred O. Blackmar III and some friends were having target practice on what they thought was a gourd, but when they crossed the creek to examine their marksmanship, they discovered they had been shooting at a human skull. This had been washed out of the bank by the floodwaters during the winter of 1936 and only the skull remained in the yellow clay muck. They examined the bank above and saw a portion of the lower leg bones and thigh

/
Figure 44. Enlarged view of Lester's Bull Creek base map showing layout of the cemetery grid. Note difference in grid north and magnetic north.

55

The Bull Creek Site

Lester to compass directions are referenced to grid north. The quotation includes the complete text of Lester's cemetery excavations.

About four hundred feet south of house site number one a burial was discovered by a visitor. The body had been washed from position by the high waters of Bull Creek and only the skull remained. Further investigation yielded the tibia and femur of another burial protruding from the bank. Control Trench 4 was staked and excavation was begun in this section.

The profile at this point shows about three inches of water-laid sand, four to five inches of yellowish clay fill from the railroad bed, historic sod development seven to ten inches, ten to twelve inches of accumulated tan sand, Indian occupation level, four to six inches thick, tan sand about twenty to twenty-six inches and underlying yellow clay loam with heavy silt deposit. The burial pit first showed about six inches above the clay loam and the burial was located at a depth of sixty-seven inches or about two inches in yellow clay loam.

Figure 47 shows a profile drawing of the cemetery originally prepared by Lester. The drawing may represent the western edge of the original trench excavation.

Figure 45. Field photograph of the "more expert N.Y.A. workers"

Burial 2 is of the contracted

at the cemetery (NPS photograph 2395).

flexed type, the body is laying on the

right side, oriented east and west with

As previously noted, the locations of all burials are not precisely known. Figure 25

the head to the west. The bones are in a rather poor state of preservation, all of the left side of the skull being gone. None of the cervical vertebrae

represents a revised version of a NPS file composite map of the cemetery upon which we have added approximate locations of missing burials using field notes. Locations of a few burials could not be approximated. For the following descriptions, the cemetery map has

and only the lower five joints of the thoracic vertebrae in place. The knees are drawn up almost to the thorax and the arms are folded with the left hand going between the legs about midway between the knees and foot. The skull is rather long in proportion to the width but the fact that it is partially crushed may cause this to look more

been cut into two halves to show greater

pronounced than it really is.

detail. Figure 46 shows the eastern half.

Again as previously noted, any reference by

56

WPA Excavations

4-25 4-26

Bl8
u~ .~ 0 <J

~ B19 !)

B5 11l\t'~""

4-21 4-22











~ ,= 1) iY~'=B37

B35



.

B29!

B8~~

Q/ B7
~ ~'i,.





t ~


B27
1,?

j
~
Bll

~



~-~

7J~o
O'\~)B6

Bl5 &f1. e

~.~'@ Bl4

B13 ff.~ B22

;;:. .@~
B28
.". ~
B26 (?)

@)
BlO

Bl6

~
""'"

:..


~:~.I

~



e

.B3
~
~k

..




.

.

B25

.

.

4-11


4-..12

c;::::.:~ ..

.z B30"\'tl',~0ri ~~

.;~::'

::.:~~:> B17 ll'~~

"t
ei Bl2


<i"~B2
.

B31







~o.~;:'

.

If e1



> /'.

/

Q

1
~q

B32

e

B24

,/

./



4-5
:,/

.

/

/

'4-. 6

.

B34
It, "'"

./

e

. ~ B33

~

.

..

/'

/' ./ /'

4.-3

4-. 4

.

/'

/'

./

/'

L13 Lll L9 L7 L5 L3 LI 4-1 4-2 RI R3

Figure 46. Reconstructed plan map of the eastern part of the cemetery containing Burials 1-33 (dotted outlines are approximations).

57

The Bull Creek Site

4-7L5

" -9LS

'1-15L5

q -I7LS

Q-/9/S

q'21L5

.-23L.5

q-2SLS

Q'27LS

@ water -Iaia sand
yellow clay @ histone sad
@ tan sand
@ occopation level
(j) tan sand @ yellow loam

... ,.... _ -.:.\:
Typical Profile Bull Creek Cemetery

Figure 47. Redrafted profile map of the cemetery area (based upon original 1936 drawing by Lester).

Extension of the trench brings forth another burial about six feet north of Burial 2 (see Figure 46). The body is in an excellent state of preservation, laying on the left side, oriented north and south with the head to the south. It is the same type of contracted flexed burial as number two with the knees drawn up to the face. Both hands go between the legs, right hand about four inches below the knee and left about five inches above the ankle.

There are some indications of post mortem movement of the body, one of the lumbar vertebrae being about six inches below the pelvis, and some of the bones of the upper body being slightly displaced. Burial furniture consists of a few beads of pierced seed {shell} around the neck and a painted dog-effigy type of water bottle {Figure 48}.

58

WPA Excavations

Mrs. Patterson considered the Burial 3 find so remarkable that the entire burial was removed intact (Figure 49). The burial was carefully removed to the lobby of the Merchants and Mechanics Bank of Columbus for exhibition during June and July. Following the exhibition the burial was moved to the U.S. Post Office until it could be placed in Columbus' new museum (Patterson 1950:39).

Lester's report continues:

In extending the excava-

tion northward to remove Burial 3

another burial (Burial 4) was

discovered about four feet to the

north at a depth of forty-four

inches below the surface and only

twenty inches below the bottom

of the occupation level. A portion

of the skull and pelvis were de-

stroyed by the laborer, but such

bones as remain are in a very

good state of preservation. The

body is laying on the left side

oriented northwest and southeast,

of with the head to the southeast.
Figure 48. Field photogra.ph the Burial 3 and first dog effigy vessel The position is a slightly different

(NPS photograph 2396).

type of flexed burial in that the arms are straight at the sides,

The bottle is in a perfect state of preservation and is placed just between the skull and the knees. It is in the shape of a small dog with a circular body, short legs, stubby tail, and the facial characteristics are accented by the use of the paint. The design on the body of the vessel is a

vertebrae straight, legs folded under the body, pelvis to knee at right angles with the vertebrae and feet only a few inches below the pelvis. To all appearances this is the burial of a female. No burial furniture associated with this burial at all.

double scroll or spiral with annular bands around the dog's tail, the neck of the vessel is decorated with spiral vertical lines coming down onto the body of the vase.

Excavated a 15ft wide trench north 25 ft from Burial 3 to a depth of seventy-eight inches at stations 4-23L1. Pottery occurs in quantities in all levels down to occupation level and a few sherds

This burial was removed in situ by pouring

have worked into tan sand below.

a concrete moulding around the block of soil on

which it was resting and by slipping a steel plate

under the form. The approximate weight of the

exhibit is between seven hundred and fifty and one

thousand pounds (Lester 1938:31).

59

The Bull Creek Site

SKELETON OF INDIAN IS ON DISPLAY HERE
The skeleton of an Indian, who apparently lived in a settlement in this vicinity some 500 years ago, recently discovered on the banks of the Chattahoochee east of Columbus, is now on display, just as he was buried, in the lobby of the Merchants and Mechanics bank.
Excavation work has been carried on in this section for several weeks under the direction of Frank E. Lester who has had the cooperation of the National Youth Administration and the Smithsonian Institute, as well as that of several interested Columbus citizens. The present find constitutes an important link in the reconstruction of burial and cultural customs of the people who inhabited this part of the Southeast hundreds
of years ago (The Columbus Ledger 13, 1936).

upper left: burial supported by concrete slab upper right: burial is boxed and hoisted away lower left: burial is transported to Columbus lower right: display is complete
Figure 49. Series of 1936 NPS photographs (accompanied by newspaper article) showing the removal of Burial 3 for exhibition.
60

Excavated Burial 5 which is of the contracted flexed type, laying on the left side, oriented south southeast, north northwest, head to south southeast {Figure 50}. The remains are in a very poor state of preservation, all of the skull gone except upper and lower teeth - no trace of cervical vertebrae or scapular. Right arm folded across the body, left arm straight, going between the femora with hands at heels. Burial furniture consists of a small greenstone celt about seven inches long. This was stolen by a visitor on the job. Burial 5 was found directly under station 4-21 L3, thirty-six inches in tan sand, seventy-two inches below surface.
Four burials (6-9) were initially found in this section. Burial 6 {Figure 51} was found directly under station 4-17L 1 seventy-eight inches below surface and forty-two inches in tan sand, no trace of a burial pit. This is the flexed type of burial, body laying on right side, oriented northwest and southeast with head to the northwest. The femora are at right angles to the vertebrae and the feet are folded back almost touching the pelvis. On the whole the bones are in a rather poor state of preservation and were considerably damaged after discovery by heavy rains. Portions of the pelvis, both femora, patellas, and tibiae; both hu-

WPA Excavations
rneri, one of the ulnar, both clavicles and scapulars, and four points of the cervical vertebrae remain. The skull has been rather badly crushed and only a few fragments were salvaged. Burial furniture consists of a small greenstone celt about four inches long, two small greenstone chisels, a black flint knife of very excellent workmanship, several pieces of worked flint of no definite shape, and a mass of red pigment, apparently some ferrous oxide used as paint. From the position of the pigment it appears that the paint was probably in a small pouch of some kind.
Burial 7 was found forty-two inches in tan sand, seventy-eight inches below surface {see Figure 51}. The remains are in a very poor state of preservation, the body is laying on the left side, oriented northwest and southeast with the head to the southeast. The remaining bones consists of portions of the skull, right fibula, both femora and tibiae. It appears that at least one of the arms went between the legs as fragments of the right ulnar and radius were found in such position. Another of the dog-effigy type of painted vessels and a small greenstone discoidal stone are found associated with the burial {Figure 52}.

Figure 50. Field photograph of Burial 5 (NPS photograph 2406). 61

The Bull Creek Site

Figure 51.

photograph of Burials 6 in foreground and Burial 7 inbackground upper right (NPS

photograph B.C. 49).

The vessel is identical with the one previously described, except that it was rather badly broken by ground pressure. All of the parts were in place and the vessel has since been
restored (Lester 1938:33).
The second dog pot also received some public attention. A short newspaper article in the Columbus paper described the find and following restoration the vessel was put on display alongside Burial 3 and the first dog pot.

Figure 52. Modern photograph of second dog pot.
62

Another Piece of Indian Pottery Found Along Banks of Chattahoochee

A strange looking effigy vase in the form of a dog with a painted design encircling the vase, apparently representing the horizon, was found recently near the Bull Creek site on the Chattahoochee river.
It was discovered about twenty feet from the skeleton now on display at the Merchants and Mechanics bank, and was also with an Indian skeleton.
Judging from the design and shape of the vase, or water jug, it was buried about 500 years ago, and was perhaps the favorite water bottle of the Indian. It is also the opinion of the workers that the design on the vase must have been first put on with some chemical known to them and then burned in the clay. At any rate, the design

is still plain, and the paint is black.
This is the second vase that has been found near the Bull Creek site on the banks of the Chattahoochee river, and is one of the four now known to be in existence. The other two vases were discovered in North Georgia and near the Flint river.
The first vase unearthed here was also buried with an Indian, the body having been buried in a flexed position, with the jug between his knees and his head. The design on this bottle is also very distinct and is the same as that on the newly discovered jug.
Both water bottles are on display at the Merchants and Mechanics bank, together with the skeleton.

Columbus Ledger July 21, 1936

WPA Excavations
lower body consists of parts of the pelvis, both femora and tibiae, and small fragments of the fibulae. Traces of the metacarpal bones between the femora. Apparently this is of the same contracted type of flexed burial as number eight. The body is laying on the left side oriented north and south with the head to the south. And is located sixty-six inches below surface, thirty-two inches in tan sand. Burial furniture consists of a small greenstone chisel three and a half inches long.
Burial lOis located south of Burial 9, fifty-four inches below surface, twentyseven inches below the occupation level. The burial consists of a few crushed fragments of a skull, no other traces of the burial being evident. There is no sign of a burial pit or outline of any of the decayed bones.
In working the section thus far investigated down to natural formation a few more burials have been located {Figure 54}. These were made from the same occupation level as those above, but at slightly different depths. On the whole, there is nothing to indicate any difference in the time interval or in the methods of burial.

Burial 8 is of the contracted flexed type, the body laying on the left side, oriented east and west, with the head to the east. The body is directly under station 4-19L3 at a depth of sixtyeight inches below surface, thirty six inches in tan sand. The face is exceptionally long in proportion to the size of the skull and the skull appears rather long for its width. The cervical vertebrae, portions of the scapulars and both clavicles remaining intact. The left arm is folded across the body with the hand resting at the pelvis, the right arm is straight, going between the femora with the right hand below the right tibia. Pelvis, tibiae, femora, fibulae, three sections of the lumbar vertebrae, and the metatarsus bones in excellent state of preservation. No trace of burial furniture associated with the burial.
Burial 9 is in a very poor state of preservation, nothing remaining of the upper body except the teeth, the cervical vertebrae, and a small section of the right humerus about four inches long,

Figure 53. Field photograph of Burial 8 (NP8 photograph 2415) .
63

The Bull Creek Site

;J,~"

.....................

/

:~:( ..

,;~

1:::::.<, ~

FigUre 54. FieldphOtograp

1)f:f)1~ph B.C.Ci2), band-tintoo for Patterson.

Contrary to Lester's description, the burial lies on its right side.

64

WPA Excavations

In working the section west of Burial 6 another burial (number eleven) was found at a depth of seventy-three inches below the surface, thirty-eight inches below the occupation level in tan sand. This is the flexed type of burial, oriented south southeast and north-northwest. The arms are straight at the sides extending almost halfway down the femora. The femora point straight downward with the feet drawn up almost touching the pelvis. The bones are in a fairly good state of preservation, a few of the ribs and the metatarsal bones remaining in place. The skull is rather small and the individual was of comparatively small stature. This is a female burial.
Burial 12 {see Figure 54} consists of another of the contracted type of flexed burials, the body is laying on the left side oriented northwest and southeast with the head to the northwest. It was found south of Burial 4 at a depth of sixty-nine inches below the surface, thirty-five inches below the occupation level. The remains consist of fragments of the skull and portions of the femora and tibiae. The body is in a very poor state of preservation, but enough remains to indicate that the legs were drawn up almost to the thorax, and represents the typical type of contract flexed burial.

an interesting pathological enlargement. A small depression is present at the lambdoid intersection. It is the burial of an old adult, male individual {Figure 55}.
Burial 15 found just a few feet west of Burial 6, was found seventy-eight inches below the surface, fifty-three inches in tan sand. The body is in a very poor state of preservation, oriented north northwest and south southeast with the head to the south southeast. It is the contracted flexed type of burial, arms appearing to go between the legs, but it is impossible to determine any exact relation, body laying on the left side. The remaining bones consist of the skull which is rather badly crushed, short sections of the humeri, traces of ulna and radius of right arm, fragments of the pelvis, and only traces of the femora and tibiae. No trace of the fibulae or any of the appendages
(Lester 1938:36).

Burial 13 was found immediately below a large tree between stations 4-17-418 and 4-15-4-16. Nothing is left except portions of the skull, a few of the teeth, and a few sections of the cervical vertebrae. From the remaining fragments it appears that the body was laying on the left side, oriented north and south with the head to the south. There is no trace of a burial pit or burial furniture. This burial was found forty inches below the surface, twenty-six inches in tan sand.

Another burial (Number 14) was located about ten feet to the north in the same section as number thirteen but at a slightly different level, being fifty-one inches below the surface, thirty-four inches in tan sand. It is of the contracted type, laying on the left side, oriented east and west with the head to the east. The knees are drawn up toward the body with the arms passing between the femora and the hands at the tibiae. On the whole, the body is in a good state of preservation and the skull is in exceptionally good shape. The skull is slightly misshapen, the right parietal shows

Figure 55. Field photograph of Burial 14 (NPS photograph 2420).

65

The Bull Creek Site

Lester's preliminary report mentions Burials 16 and 17 but omits descriptions of these two important burials. The report does mention burial artifacts, a third dog pot and an effigy pipe following the discussion of Burial 20 (page 37) but the artifacts are not referenced to burial number. For the sake of convenience those paragraphs are inserted at this point along with the field note descriptions of the two burials.

Burial 16 consists of a few fragments of tibia and femora, bones are scattered over a large area and this is really the remains of two bodies; general orientation is south southwest and north northeast. Burial furniture consists of the 3rd effigy dog vessel {Figure 56}. Found 36" below
occupation level (Lester 1938:60).

The vessel is slightly different from the

other two in that the neck is higher and smaller,

the designs being painted black on a red back-

ground, design slightly different, and the legs are a

little longer and at a slightly dif-

ferent angle (Lester 1938: .37).

Figure 56. Field photograph of Burial 16 and

third dog pot (NPS photograph 2431).

Burial 17 is of the con-

tracted flexed type, oriented

NW-SE head to NW. Body

laying on right side with the

knees at right angles to the

vertebrae and both arms going

between the legs. Skeletal

remains in rather poor state of

preservation, one side of skull

crushed and all of metacarpal

and metatarsal bones gone.

Only a few joints of the lumbar

vertebrae remain but 5 of the

thoracic ribs. Portion of skull

was all that could be removed.

Found 70" below surface, 40"

below occupation level in tan

sand. Burial furniture includes a

very remarkable human effigy

pipe {Figure 57}. This pipe is a

remarkable piece of work, facial

characteristics are typically

Indian, and the whole expression

has been captured by the artist.

A tobacco "cake" almost a

sixteenth of an inch thick still

remains inside the pipe.

Figure 57. Photograph of tobacco pipe from Burial 17 (Columbus

Museum files).

66

WPA Excavations

Burial 18 is located directly under station 4-24, sixty-one inches below surface, thirty-nine inches below the occupation level, and twelve inches in yellow clay loam. The body is laying on the left side oriented southeast and northwest with head to the southeast. The knees are drawn up. We have here a typical Bull Creek type of contracted burial, the remains consisting of a few fragments of the skull and sections of the femora {Figure 58}. Burial furniture associated with the burial is a small bowl, approximately three and a half inches in diameter with a design of incised scrolls combined with punctate markings, several large beads of shell, a small piece of conch shell core, and an unbroken clay pipe of the human effigy type {Figures 59 and 60}.

Bones

Pipe.
BOW/~
+
Sta 4 -21f

c:Sk.t.///

Figure 58. Field map of Burial 18.

3
i
em
Figure 59. Photograph of small bowl found with Burial 18 (Columbus Museum file photograph).
67

The Bull Creek Site

Although the Bull Creek cemetery is by no means completed we may briefly summarize the facts known about the village from what has been accomplished thus far. Flexed burials of two slightly different types represent the burial custom of the inhabitants. However, there is no consistent orientation. It appears that the contracted burials must have been lashed into position, but there are no traces of coverings of any sort whether of hides, bark, or blankets.

------ ........

From the special finds, pottery finds, and lack of any European culture, it is sufficient to say that this site represents a prehistoric village very similar to the Lamar village below Macon. The principal difference is a slight influx of Gulf Coast material and the use of the painted effigy ware as burial offerings.

Figure 60. Field drawing of tobacco pipe from Burial 18 (Lester 1938).

Burial 19 is only a few feet to the east of Burial 18 at a depth of sixty-seven inches below the surface, forty-five inches below the occupation level, fourteen inches in yellow clay loam. It is a female burial, laying on the right side, oriented northwest and southeast with the head to the southeast. This body is contracted more than any found to date, the knees resting against the chest. The skull is rather badly crushed and none of the cervical vertebrae remain, but the rest of the body is in a very good state of preservation {Figure 61}.

Burial 20 is an example of the other type of flexed burial found at Bull Creek {Figure 62}. The body is laying on the left side, oriented east southeast and west northwest, with the vertebrae straight and femora at right angles to the body, feet drawn up almost touching the pelvis. Male skeleton in a fair state of preservation, remaining bones consisting of skull, cervical vertebrae, both femora, portions of both tibiae, clavicles, scapulae, and small sections of the pelvis. A string of seed beads was found in place around the neck but nothing else with the burial {Figure 63}.

Figure 61. Field photograph of Burial 19 (NPS photograph 2427).

68

Figure 62. Field photograph of Burial 20 (NPS photograph 2428).

WPA Excavations
Burial 22 was lying on the right side, oriented NW and SE with the head to the NW {Figure 64}. The skull has been crushed but otherwise the body was in an excellent state of preservation. The pelvis was slightly higher than the head and had been slightly misplaced by root growth but was in an excellent state of preservation. The body was found 54" below surface, 40" in tan sand. Remaining bones consisted of skull, clavicles, scapular, both humeri, ulna, radii, a few of the metacarpal bones, practically all of the vertebrae, tibiae, femora, fibulae, metatarsal bones, and phalanges.
Burial 23 consisted of a few small fragments of scattered bones that had reached such a state of decay that it was impossible to determine anything about them other than that they were human remains, apparently they were part of the long bones of the body but were beyond all definite identification. Found 57" below the surface, 33" in tan sand.
Burial 24 consists of broken fragments of the skull and portions of the humeri, femora, and tibiae. The body is lying on the left side, oriented northwest and southeast with the head to the southeast. It appears that this is a burial of the contracted flexed type. Found 28" in tan sand 59" below surface.

Figure 63. Photograph of marine shell beads found with Burial 20 (Columbus Museum files).
Burial 21 was located 36" below surface, 21" in tan sand. The remains were in a very poor state of preservation, consisting of portions of the skull, tibiae, femora, ulna, and radii. The body was lying on the right side, oriented NNW and SSE with the head to the NNW. The legs were flexed back under the body with the arms straight at the sides.

.;:::.;
Figure 64. Field photograph of Burial 22 (NPS photograph 2433).
69

The Bull Creek Site
Burial 25 is in a very poor state of preservation. The body is lying on the right side oriented NW and SE with the head to the SE. Remaining bones consist of portions of the skull and fragments of the femora, tibiae, and fibulae. From the position of the femora and tibiae it appears that the legs were doubled back under the body with the knees in front, flexed burial of the 2nd Bull Creek type. 52" below surface, 21" in tan sand.
Burial 26 is an extended flexed burial, lying on the left side oriented north and south with the head to the south. Skeletal remains in a very poor state of preservation consisting of fragments of the skull and portions of the tibiae and femora, 79" below surface, 41" in tan sand.
Burial 27 was found 75" below the surface, 41" below the occupation level in tan sand. The remains are in a very poor state of preservation but it appears that the burial is the typical type of Bull Creek flexed burial with the knees drawn up toward the face. Remaining bones consist of fragments of the skull and tibiae and femora. Oriented NNW and SSE with the head to the SSE. No burial furniture.
Burial 28 is located only a foot away from burial #27,80" below the surface, 35" below the occupation level, 5" in reddish brown loam. The body was lying on the back flexed to the left. Oriented ESE and WNW with the head to ESE. Remaining bones consist of portions of the left tibiae and femur and the skull with all of the facial bones missing. Burial furniture associated with the burial included a small amount of red pigment, a piece of mussel shell, a fragment of worked bone presumably an awl, several other pieces of animal bone, and a small stone of the same material as the pigment.
Burial 29 was found 88" below surface, 47" below the occupation level 6" in reddish brown loam. Nothing is left except sections of the tibiae, femora and a short section of the right humerus. Body oriented NW and SE with the head to the SE. There is no trace of the skull at all. Appears to be typical contracted flexed type. Burial furniture consists of a small ball of the red "paint" pigment.
Burial 30 is located 49" below the surface, 11 " below the occupation level. Oriented NNW and SSE with the head to the NNW. Lying on the right side. Remaining bones consist of portions of the skull, tibiae, femora, and a small section of the pelvis. No burial furniture.

Burial 31 was a triple burial the first of this sort found at Bull Creek. The remains were in a very poor state of preservation and it was impossible to determine the exact relationship of the individual bones. It appears however that at least two of the bodies were flexed in the typical Bull Creek manner. The other burial consisted of the skull only. (A) consisted of the skull only, was slightly higher than the other two and all of the skull was gone except a portion of the back occipital and part of the lower jaw. All that remains of (B) was the skull, frontal bones and facial bones missing, and a short section of the right humerus. (C) consisted of portions of the upper and lower jaw, traces of both humeri and radii, and sections of both tibiae, fibulae, and femora. Found 61" below surface, 23" below occupation level, oriented NW and SE, heads NW.
Burial 32 was the contracted type of flexed burial, lying on the right side, oriented Nand S with head to the N. The remains were in a rather poor state of preservation, remaining bones portions of the skull, tibiae, femora, humeri, and ulna. Burial furniture consists of several rocks, flint chips, and fragments of animal bones - impossible to determine whether or not they had been worked, all of these articles were grouped around the neck. 52" below the surface, 34" in tan sand.
Burial 33 was located 36" below the surface, 7" in yellow clay loam, body lying on the right side, oriented Nand S with the head to the S. Nothing remains except portions of the skull and teeth, tibiae, femora, fibulae. From the position of the bones it appears that it was the typical type of contracted flexed burial.
Burial 34 was found to the west of Burials 32 and 33 at a depth of 37" below the surface, 16" in tan sand. The body was lying on the left side, oriented east and west with the head to the west. The remains were in a very poor state of preservation, consisting of skull, which was almost entirely decayed, and sections of both tibiae, which were slightly higher than the skull. This burial was evidently of the contracted, flexed type with the arms between the legs. No burial furniture {Figure 65}.
Burial 35 was located to the west of Burials 27 and 28, 67" below the surface, 27" below occupation level in tan sand. The body was lying on the left side, oriented east and west, with the head to the east.

70

WPA Excavations

/

/

/

/

/
/

/

/

/ / /.

Find 72


/

/

/

./

/
/

B44

. /
/~/o.~



B46

;'1

,~



if


B45

D B42
()

















. t~
B43

. B36
.. ~ ~



,J


B40

L33 L31 L29 L27 L25 L23 L21 L19 LI7 LI5 L13 Figure 65. Plan map of western portion of Bull Creek cemetery.
71

The Bull Creek Site
Burial 35 was of the contracted, flexed type, with the knees drawn up toward the chest and the right arm crooked at the elbow - not going between the legs as is the usual case. The skeletal remains were in a rather poor state of preservation. The skull was very badly decayed, nothing left of the upper body except a short section of the humerus and ulna of the right arm. The lower body consisted of a few fragments of the lumbar vertebrae, practically all of the pelvis, and portions of both tibiae, femora and fibulae. No burial furniture is associated with the burial.
Burial 36 was found to the southwest of Burial 35 at a depth of 60" below the surface, 35" below occupation level in tan sand. The body appeared to have been placed on the back with the head turned to the left, oriented east and west with the head to the east. The knees were evidently against the chest and were found about 6" inches higher than the skull. Only the skull, both tibiae, and traces of the right humerus remained.
Burial 37 was found immediately east of Burial 35, 86" below the surface, 36" below occupation level. The body is oriented east and west, head to the west. Though of the contracted, flexed type, it differs from previous burials in that the body was face downward. It appeared that

the legs were lashed up with the knees touching the chin, with the left arm flexed downward on top of the right thigh, the right arm flexed up with the hand at the face. Remains were in a rather exceptional state of preservation. No burial furniture.
Burial 38 {Figure 66} was found approximately 15ft northwest of Burial 35 at a depth of 36" below the occupation level. This is a double, extended burial. Both bodies were on the right side, oriented northwest-southeast, with the head to the northwest.
A. These remains were in rather poor state of preservation. The skull was crushed and broken parts have been scattered. Remaining fragments consist of portions of skull, humeri, ulna, radii, portions of tibiae, femora and fibula.
B. These remains were in much better state of preservation. Remains consisted of skull; one side of which is rather badly decomposed, sections of ulna, humeri, radii, clavicles, scapulas, all of the vertebrae, with part of the ribs attached, part of the pelvis tibiae, femora, fibula and most of the bones of the feet.

..,-,-.

;::"::;:::::::'"

Photograph of extended burial 38, a multiple burial (NPS photograph 2453).

72

From the comparative size of the bodies it appeared that A was female; B was male. No burial furniture was associated with this interment.
Burial 39 was located to the southwest of Burial 38 at a depth of 42" below the occupation level. The body was lying on the left side oriented southwest-northeast with the head to the southwest. This burial was of the contracted flexed type with the legs drawn up toward the body and the head tilted downward toward the knees. It was interesting to note that instead of the hands going between the legs as is the usual case in this type of Bull Creek burial, the arms were flexed at the elbow and both hands were at the chin. Skeletal remains were in excellent state of preservation, consisting of: skull in excellent condition, all of the vertebrae, humeri, ulna, radii, all the metacarpal bones, portions of the pelvis, tibiae, femora, fibula, metatarsal bones.
Burial 40 consisted of a few fragments of the skull found to the south of Burial 36, at a depth of 61" below surface, 42" below occupation level in tan sand. Careful troweling failed to reveal any other traces of the body.
Burial 41 was found approximately 10ft west of Burial 40 at a depth of 39" below the occupation level. The body was lying on the right side oriented east and west with the head to the west. It was of the contracted flexed type but it was impossible to tell the exact position of the arms as only a small portion of the left humerus remained. Remaining bones consisted of fragments of the skull, small parts of both clavicles, pelvis, tibiae and femora. There was a small piece of animal bone close to the upper part of the humerus, probably accidental placement as part of pit fill.
Burial 42 was found south of Burial 39, 26" below occupation level in tan sand. Burial was of contracted flexed type oriented north and south with the head to the south. The body was lying on the left side. The feet were gone but there must have been a contact with the pelvis with the knees drawn up to the chest. The left hand went between the thighs about 2" from the pelvis and the right arm was across the chest making the ulna and radius vertical. The head was turned to the left thus making it face downward. On the whole the body is in fairly good state of preservation except that the back of the skull is rather badly crushed.
Burial 43 was of the extended flexed type. The burial was located west of Burial 41 and was

WPA Excavations
found 50" below occupation level. Body was lying on the left side oriented east and west with the head to the east. The feet were flexed back almost touching the pelvis and the knees were straight down away from the body. It was impossible to tell the exact position of the arms as there was no trace of them. Only the skull has been rather badly crushed, sections of the clavicles, vertebrae, both femora, tibiae and parts of the pelvis remained.
Burial 44 {Figure 67} was found to the northwest of Burial 42 at a depth of 26" below the occupation level in tan sand, oriented east and west with the head to the west. This was a rather unusual type of burial in that it was placed in the pit face downward. The arms were straight at the side and the legs were slightly flexed to the left. It appears that the body must have been lashed in this position before interment. On the whole this is the most complete burial yet to come from the Bull Creek village. All of the body remaining with the possible exception of a few of the ribs.
Burial 45 was found south of Burial 44 at a depth of 33" below the occupation level. The remains were in such a poor state of preservation that it was impossible to get any exact information about this burial but it appeared that the body was lying on the left side oriented east and west with the head to the east and from the short sections of the tibiae, femora, it seems that it was of the contracted flexed type.
Lester's field notes describe at this time the recovery of a small vessel at a depth that should only be associated with a burial.
Found a small unbroken vessel, Find #72, at a depth of 27" below occupation level; 33" north of Station 4-17L21 on line of stakes. There is no evidence of burial pit in this location and no trace of any bones, but this vessel must have been an article of burial furniture associated with a burial
(Lester 1938:73).
Lester's observation is of critical importance because it demonstrates the difficulty of defining some of the burial pits. Coupled with other field note entries that comment on this difficulty, it is quite probable that some burials were missed entirely. Apparently, only burial pits that intruded the richest occupation midden zones were well defined. However, it seems that most burial pits did intrude into dark Bull Creek middens of earlier occupations.
73

The Bull Creek Site

Figure 67. Field photograph of Burial 44 (NPS photograph 2460).

Burial 46 was found to the west of Burial 45, 28" below occupation level. It was impossible to get any accurate picture of the burial as nothing remained except the crown of a few teeth and short sections of both femora and tibiae. From the position of these it seemed to be of the contracted flexed type lying on the left side oriented north and
south with the head to the south (Lester 1938:44).
Lester's preliminary report abruptly ends at this point without further discussion. Perhaps a portion of the report has been lost, but, just as likely, the report ends with the entry for the last burial {Number 46}. Several field notes concerning the excavation and observations are not presented in the preliminary report. Lester's notes include a running account of excavated squares. While this record may not be absolutely correct due to some omissions and transcription errors from the original field notes, it does allow for an approximate recreation of the boundaries of

the cemetery excavations. Importantly, these notes also provide a few specific location references to important landmarks, such as the western edge of the railroad levee and the upper edge of the river bluff. Lester notes that the western edge of the railroad fill lies 20 feet west of station 4-27L15 (Lester 1938:70). He also notes that the western edge of the excavation was extended to the bluff edge on a line from station 4-13L33 to 4-21L27 (Lester 1938:74). These locations correlate very well with Lester's burial ground grid plan shown on the site base map (see Figure 20). Figure 68 illustrates our approximation of the cemetery limits as interpreted from Lester's field notes. Importantly, this reconstruction seems to support the conclusion that all of Lester's burial orientations are based upon a grid north and not a magnetic north.

74

WPA Excavations

II

II

II

II

II

I
I

II . .'31~?~3e~"6'3~ Bu(.f'a/ Ground

I ..... ' . '3..\\/':;>~~9..A

/. . . I I ..' .:.:~iy ~}.6i"\

if!'li. \ I I ~.;.:,. .

I I ,..<-. . . I.... ,., ~.,., .

...)/' . -~'9-1-";0. \
. "\?.p...~..\.\1.\c\;'~06\.t...\

I
I

I
I

. .....
....

.,.\..')"03\',."/\1\~~.~

\

I

",~

.:../

../>
/"""

"

""
,,

I I I ./ \"''''''',.,''

I " I I _ ...."- ..,/ 10"""

. /..... ./.\.,v. . . / "

>

I
I
I

I
I 1./

/

// /

/

/

{/

", >

I

/

J

I

I

/

II

I

I

II

I
I I

:o
t\:l
~

II I

/,'.,,"./"/ ,.:/I/ /I

/ / /

/

/

(

I~ I i

I ::!'1 I I

I ::::: I \

I

\

I II

I I\

II

Figure 68. Plan drawing showing interpreted boundaries of cemetery excavations superimposed upon Lester's

base map.

Lester's field notes also contain observations concerning the possible existence of structures in the cemetery area. By omitting these observations from the text of his preliminary report, questions concerning the nature and origin of the burials remain unanswered. Based upon the information presented in Lester's report we cannot determine the relationship of the cemetery area to the rest of the site. Among the possibilities are 1) the burials lay within a cemetery segregated from the rest of the village; 2) the burials lay beneath

mound fill which might once have existed but had been leveled by railroad construction; 3) the burials lay beneath and surrounding the floors of domestic structures; or 4) the burials lay beneath the floor of a large ceremonial structure such as a council house.
Lester's field notes indicate that evidence of structures was found during the cemetery excavation but that this evidence was dismissed as too insubstantial. Two examples of Lester's tentative evidence follow.
75

The Bull Creek Site
September 22. The profile at this point shows a band of clay from 2" to 6" thick between station 4-17L9 and 4-23L9, suggesting a house floor. There is also some trace of a sod development below the general occupation level, though it shows no definite form at this particular point (Lester 1938:65).
December 29 " .. began work on cut from 4-11 L13 -L 15 to 4-27L 13 -L15. Working down to occupational level to examine what appears to be a house site.
Lowered section between 4-11 L13 -L15 and 4-27L13 -L15, to the Indian occupation level cleared whole area 4-11 L9 -L15 to 4-27L9-L15. While the area contains a very rich midden accumulation and several buried and charred areas and a few miscellaneous post mould impressions, there is no definite sign of a house site. There is quite a profusion of pottery and a few briquettes (Lester 1938:70).
The area of the cemetery which concerned Lester is shown in Figure 69. The area is intriguing in that it represents somewhat of a void of bodies.

An important consideration is that the cemetery lay within sandy deposits while the two structures excavated in the village area lay within a well-consolidated silty sand. These differences in soil would have meant different preservation states for any structures constructed in the two areas. The soils containing the village area structures would have provided nearly ideal conditions for preservation whereas the sandy soils in the cemetery area would have been greatly affected by leaching and various forms of bioturbation. Also, the village structures had been burned, resulting in brick hard surfaces which would have been less affected by various forms of erosion.
Lester's work left a number of unanswered questions concerning the nature of the cemetery. Determining the answers to these questions eventually led Dr. A.R. Kelly to return to the site to conduct further excavations. These and other investigations that were conducted over the next four decades will be the subject of following chapters.

4-27 4-28

Clay Floor in Profile ~(:-.

4-25 4-26

t(

~.

.",

4-21 4-22

I'

v
4-11 4-12

AGN

~2m

o

s Feet

/

/

w3 wI L29 L27

L9 L7

L5 ts

LI 4-1 4-2 RI

R3

Figure 69. Reconstructed plan of cemetery showing area thought by Lester to contain structural evidence.

76

Chapter 4 Kelly's 1950 Investigations

A.R. Kelly, by 1950 head of the Department of Archaeology and Anthropology of the University of Georgia, returned to the Bull Creek site to conduct further excavation in the Spring of 1950. The intent of this work was to expand upon the work of Lester and Patterson in 1936 and 1937. As previously noted, Lester's preliminary report was simply an abbreviated rewrite of his field notes. The manuscript chronicled the field project but there was little interpretation and virtually no artifact examination. The inability to produce a detailed site report clearly frustrated both Kelly and Patterson during the following years. Unfortunately, Bull Creek was like most WPA excavation projects in that adequate funding was available for excavation but not for write-up. The closest Kelly came to producing a meaningful presentation was a paper prepared immediately following Lester's field work which examined Bull Creek within the context of other important Lamar sites.
In 1937, Kelly prepared a paper for the annual meeting of the Society of American Archaeology in New Haven, Connecticut entitled Lamar and Related Sites Excavations (Kelly 1937). In that paper, the Bull Creek site was accorded its highest level of archeological significance, being included with such sites as Lamar, Mossy Oak, and Irene. This brief moment of fame departed quickly and the Bull Creek site slipped into obscurity.
Kelly's New Haven paper was never published and subsequent publications by Kelly included only brief references to the site. For example, in his Bureau of American Ethnology report on Macon Plateau, Kelly simply noted the contributions of Patterson and Lester and stated the importance of Bull Creek "lies in the fmding of a specialized funerary ware of painted effigy dog pots found with burials in the village midden. The domestic ware from the midden

represented typical examples found in the Lamar pottery complex. The same conjunction of Lamar-like pottery traits with painted dog pots had previously been noted for the Neisler site, on the Flint" (Kelly 1938:51). On the same page, Kelly mentioned that a separate report for Bull Creek was to be prepared. Thereafter, Bull Creek was relegated to one example of the Lamar phenomena.
The New Haven paper presents a rare glimpse into Kelly's basic concepts of interpretation of the Bull Creek site. Selections from correspondence as well as pertinent portions of the paper are presented here to explore these concepts. The basic ideas for the New Haven paper were discussed in several letters between Kelly and Patterson. Besides detailing Kelly's views, these letters provide insight into the academic isolation of the south at the time and the fundamental fmancial and political realities of Depression-era archeology.
In a letter dated October 22, 1937, Kelly conveyed the major concepts of the paper to Mrs. Patterson.
You will appreciate the importance of Irene in connection with previous research carried out at Lamar Mounds in Macon and at Bull Creek, at Columbus. These three sites together give an idea of the cultural characteristics of tribes who inhabited Georgia at about the same time over wide areas, as indicated by the Chattahoochee, middle Georgia, and coastal sites mentioned. These sites show types of pottery, mound characters, burial traits, game pieces, ornaments, pipes, and other cultural features implying a mixture of cultural habits between two groups. One of these, represented by the Lamar complicated stamped pottery indices, native to the southeast; the other indicated by the incised ware, almost certainly were immigrants from the lower Mississippi east, crossing many streams, and finally settling on the Georgian rivers.

77

The Bull Creek Site
The Kasita and Kawita were in the vanguard of these early tribal movements. Bull Creek, Lamar (Ocmulgee Old Town), and Irene represent stations on which settlements and remains of these first contacts of the ancestral Creeks with "strange Indians" already inhabiting the area were made.
Time indicated for these settlements would be pre-De Soto according to my conceptions. And I believe that Dr. Swanton would concur in this judgment as he feels that the restless, unsettled condition of the country at the time of De Soto was an aftermath of tribal dislocation incident to these original migrations of the Creeks.
The above summary gives about the gist of the remarks which I would like to make at New Haven. I hope that this will serve to cover the information you advised me you would like to have.
Please let me have your reactions (Ledbetter 1995b:26).
The New Haven paper is mentioned in a letter from Kelly to Patterson dated November 4, 1937.
I am sending on a copy of manuscript "Lamar and Related Site Exploration", to be offered for the meeting at New Haven. It strikes me that this topic would be of more general interest to Georgians in general as well as to my scientific colleagues. It touches on crucial exploration in three areas of Georgia, the Chattahoochee, the Ocmulgee, and Savannah. It brings into comparison other site exploration familiar to students in the Southeast who know
existing archaeological literature (Ledbetter 1995b:28).
Mrs. Patterson replied to Kelly in a letter dated November 16, 1937.
Thank you a million for your letter of November 4th containing a copy of your manuscript "Lamar and Related Site Exploration," to be offered for the meeting at New Haven .... It would be difficult to describe my appreciation of your going and telling about Georgia. I am very grateful to you for this information for I had only fragmentary inklings of much that you now make plain ....
I am sure the trip would do you a lot of good and do much for Georgia. I am glad that you had such a nice talk with Mrs. Woodward and I am delighted over the new light in regard to emphasis between state universities, Park Service, and the
78

Smithsonian. I hope Georgia will not be a laggard in
taking advantage of this service (Ledbetter 1995b:30).
In a letter from Kelly to Mrs. Patterson dated December 18, 1937, the realities of WPA era archeology are evident. Kelly's letter stresses the benefits of involving the larger academic community in the work in Georgia but the main topic of the letter is simply pulling together enough money to make the trip.
I have less than a week in which to make preparation for the New Haven jaunt. That is my fault alone because I have not pressed this matter on the one or two occasions when we had occasion to discuss it. The situation has developed in an unanticipated manner and I think I should be frank in dealing with the embarrassing features of the case.
My impression over the telephone when the New Haven matter was presented by you was that the Colonial Dames and others felt strongly that Georgia should be represented at the meetings and that papers dealing with Georgia archaeology, which has attracted such widespread attention, should be given. That there was a feeling that I should have been present at the last meeting and must certainly attend this one . . ..
Now that my name and that of G.Willey are down for papers on the morning of Dec. 30th at New Haven, I must go. And I will manage in some fashion. But still I am going to be stubborn about the finances unless it turns out that you have got the support you expected.
I have no objection to letting your organization help me to make the trip because I feel that the subject of my paper has great significance for work in which they have been particularly interested for some time; i.e., the Lamar sites description pretty well outlines the archaeological picture of what Georgia was like at or just before the time of De Soto's journey ....
I don't know what I'll do about clothes--and in writing that I'm not talking in the same sense as the debutante worrying about a coming out party. Gosh, it may be alright to go without a hat in the amenable Georgia Climate; that can be set down as eccentricity, maybe I'm just collegiate---but a 1200 mile trip to the less clement North is another matter. However, that's a purely personal element so we'll
skip it (Ledbetter 1995b:35).

A letter to Pattersondated January 12, 1938 gave Kelly's impressions of the meetings.
The New Haven Conference was from most points of view successful and propitious .... As to the reading of the papers, I should advise you that I found the fifteen minutes allotted as expected altogether a handicap to permit reading the entire paper. It was necessary to talk more or less extemporaneously .... I am writing H.C. Collins of the National Museum to send my papers so that I can add notes bibliographically with an eye to publication in American Antiquity. This will give fuller publicity to the interesting connections between Lamar and other related field work in the Georgia area, notably the
work at Bull Creek and at Savannah (Ledbetter 1995b:37).
The actual details allotted to the Bull Creek site in Kelly's paper are minimal. Because analysis had not been conducted on the excavated material, Kelly focused upon the high points of the site and specifically the cemetery. Kelly was obviously faced with the problem of being required to include some information on all of the sites sponsored by the individuals and communities of the region. Excerpts of Kelly's paper which deal specifically with the Bull Creek site follow.
The Bull Creek site on the Chattahoochee, within the outskirts of Columbus, Georgia, was explored under the field direction of Frank Lester, engineering student trained in archaeological field methods at Macon. Recognition of the importance of the Bull Creek site grew out of historical documentary studies made by Mrs. H. Wayne Patterson of Columbus, interested in checking the location of historic Creek settlements on the Chattahoochee, following leads provided in publications of Dr. J.R. Swanton. The investiqations employed National Youth Administration workers during a period extending from March, 1936 to April, 1937.
The work at Bull Creek uncovered remains of a village site exposed in the river bank and in an old road cut; also, a burial ground in conjunction with the village revealed features of special interest. Concisely, Bull Creek shows a typical Lamar-like site, minus pyramidal mounds, with Lamar complicated stamped and broad-lined incised ware contrasting with a very ornate painted effigy ware found with burials in the cemetery belonging to the village. No less than three dog effigy vessels were found with

Kelly's 1950 Investigations
burials. In the same midden in which the burials occurred and in association with other burials in the same cemetery were catalogued numerous sherds and some whole pots characteristic of typical Lamar complex. There was no evidence of later intrusion of the burials with the specialized effigy funerary ware nor could the profile features be interpreted to mean that the burials with effigy pottery were preexisting.
At the Neisler mound and village site on the Flint river, near Roberta, Georgia, thirty miles south of Macon, river wash has exposed burials associated with painted dog effigy pottery vessels identical with those found at Bull Creek. So far as the study of surface collections from the Neisler site can indicate, it is evident that the same combination of characteristic complicated stamped pottery and
incised ware occur here as at Bull Creek (Kelly 1937:5-7).
Kelly's paper continues to summarize other sites and examine Lamar traits. Burial traits were singled out for Bull Creek. Kelly noted that while many Lamar traits showed great consistency across a broad geographic area, burial customs observed at several key sites showed wide divergence.
The general condition is to find primary flexed, semi-flexed or extended burials in large graves or pits; group burials are frequent; burial furniture tends to be abundant, consisting of pottery, polished stone artifacts, and pipes; burials occur in house floors, outside the walls of the houses, on the ramps of house mounds, or in the midden layers.
At Bull Creek on the Chattahoochee and at the Neisler mound on the Flint, we have noted the presence of a specialized burial ware of painted dog effigy pots found with flexed primary burials. Ordinary domestic ware of the Lamar type occurs with other burials on the same sites .... Other Lamar-like sites, as judged from surface features and pottery in
Alabama, show urn burial also (Kelly 1937:12-13).
Kelly's work with the National Park Service took him away from Georgia for several years but he continued to conununicate with Mrs. Patterson about Bull Creek. Upon his return to Macon, Mrs. Patterson attempted to jog his memory. This is evident in a letter from Kelly to Patterson dated May 17, 1945.
79

The Bull Creek Site
You want to know about Bull Creek - I don't know and will have to check. Better write me in detail on what you have in mind. Remember, I've been away a long time and its a slow torturous business trying to pick up all the loose ends. And there are plenty of them - things have been going to rock and ruin these years - deterioration is accelerating now and I'm hard pressed to get funds
out of NPS, or to get adequate help (Ledbetter 1995b:43).
Archeological interest in Bull Creek revitalized with threatened commercial development. TIlls threat is mentioned in a letter from Kelly to Patterson dated November 8, 1949.
Dear Isabel, I haven't heard from you since the flurry last summer in the matter of Bull Creek site and successful efforts to stay off threats of construction there. I have since had further correspondence with the Secretary of the Chamber of Commerce who seems to be a good friend of your museum project and in helping to do something with
archaeology around Columbus Falls (Ledbetter 1995b:44).
In a letter dated November 29, 1949 Kelly discussed with Mrs. Patterson his plan to write a Bull Creek report.
Yes, I'll write it up for publication if we can assemble sufficient notes and data. And we'll plan to get it out in our new series on Anthropology and Archaeology at the University Press, if I can rake up about $300 to get it out. It will probably cost about that much on the basis of what it takes to get out Bill Sears' Kolomoki Season I, now in press. Citizens of
Blakely gave us the money for that (Ledbetter 1995b:45).
Ironically, Kelly's plans for writing up Bull Creek were postponed by the opportunity for further excavation on the site. Apparently, the project was supported in large measure by the Columbus Chamber of Commerce. Virtually nothing is known about the excavations except the material included in a Chamber of Commerce newsletter article and a report written, at least in part, by Kelly.
The field project was obviously meant to provide data for additional large scale excavation
80

of the site. The city supplied the labor and a photographer. The photographer was responsible for some of the best, albeit excessively posed, documentation of any of the early excavations at Bull Creek. The photographs have been preserved in an annotated album compiled by Mrs. Patterson (Columbus Museum files).
The city also prepared a detailed map of the Bull Creek site (Figure 70). A second version of the map, prepared by Frank Schnell, Jr., is shown in Figure 71. Figure 72 shows site area as it appeared in a 1950 aerial photograph.
The text of the newsletter is presented as an introduction to Kelly's report. The text of Kelly's report that deals specifically with the field investigations follows. The introductory culture history portion of Kelly's report has been omitted.
The text of the chamber of commerce
article titled Local Archaeological Possibilities Studied May 1950 follows.
Definite steps toward developing archaeological possibilities in Columbus were taken this week, when the Board of Directors requested President J.W. Woodruff, Jr., to appoint a Committee to study recommendations made to the Board by Dr. A.R. Kelly, Head of the Department of Archaeology and Anthropology at the University of Georgia.
Dr. Kelly met with the Board of Directors on Monday, May 22, at the request of Chamber officials. Mr. Mahan, also of the University, and Mrs. Wayne Patterson, local archaeological enthusiast, were also guests of the Board.
Chamber officials had requested Dr. Kelly to visit Columbus several months ago to investigate local possibilities and to submit his recommendations regarding the archaeological possibilities existing in and around Columbus. Dr. Kelly's report and recommendations were based on his subsequent visits and studies. Dr. Kelly pointed out that the Chattahoochee River area, near Columbus falls, has long been recognized in the annals of American archaeology and ethno-history as one of the most populous areas in aboriginal times. He stated that more evidences of this occupation were exhibited in the form of sites.

Dr. Kelly went on to state that "Dr. John R. Swanton of the Smithsonian Institute probably showed more Indian village site locations on the Chattahoochee, within a 30-mile radius of the Columbus falls, than at any other point on the route to the Gulf." Dr. Swanton's showing of the historic and proto-historic tribes was based on studies of 17th and 18th century maps of the region, bolstered by researches on the ground. According to Dr. Kelly's report, archaeological surveys undertaken thus far have tended to substantiate the indications of the ethno-history.
Dr. Kelly pointed out that the Bull Creek site is suffering from erosion and that commercial development in that area, which will follow the

Kelly's 1950 Investigations
completion of the Chattahoochee-Apalachicola-Flint project, will destroy valuable relics existing in that area.
Dr. Kelly recommended to the Board that the area be excavated immediately, and that relics salvaged be removed from the site and stored in a place for safe-keeping until such time as a local museum is available to house the materials. He recommended against preserving the relics "on the site," but that they be removed before they are destroyed by commercial development and erosion. President Woodruff announced that he would
appoint a Committee in the near future (Progress, Volume III, No. 25, May 1950:3).

.. --r ~,)":i~;,))
~ 1936 FINDINGS
c:::2Zl 1950 FINDINGS

LAY~UT
SHOWj~~G
CITY OWNED PROPERTY ALONG CHATTAHOOCHEE RIVER AND BULL CREEK WHERE FINDINGS OF INDIAN INHABITATION WERE MADE
DRAWNBY ~e ;;If-/
OFFICE OFCITY ENGINEER CITY OF COLUMBUS, GA.
1950
SCALE 1;100' APFlOXIMATELY

Figure 70. Redrafted copy of a 1950 City Engineers map of the Bull Creek site. 81

The Bull Creek Site

<>;
.... :.'J":::.

. .'.'

\

.....

N
!

9MEI
Schematic Map of
Excavation Units Kelly: 1950

o

30m

o

-100ft

. ," , . : ...:: '".: . '.

Figure 71. Redrafted copy of the map of Kelly's excavations (adapted from Schnell 1963)

82

Kelly's 1950 Investigations
Figure 72. Enlarged portion>of1950aenalphotograph of the Bull Creek site area. USDA, Soil Conservation Service photograph taken approximately two months prior to Kelly's field work.
83

The Bull Creek Site

Kelly's Report

importance of the site for commercial development poses a threat to the preservation of any

The first three pages of Kelly's 1950

prehistoric cultural remains still remaining in the ground. The area between the river and the creek

report, entitled Report on the Archaeological

has been cut back by both streams and is

Possibilities of the Bull Creek Site, Columbus, Georgia, with Recommendations for Development presents a background to the prehistory and early history of the region and an introductory paragraph for Kelly's work in

undergoing rapid erosion. It has also been severely hacked by the construction of the industrial railway embankment constructed in 1926, so that, practically, one needs to know just how much of the deepest and richest deposits, which yielded 46 Indian burials, several house sites, and much in the

early May 1950. The background pages have been omitted. The quoted portion of the report begins with the introductory paragraph on page 3 and continues through to the end of

way of relics possessing high artistic qualities, still remains to be salvaged. A review of findings in five units of stratigraphic trenching and test pitting of the area follows.

the narrative account of the report (page 7). Figure 72 is an enlarged portion of an aerial photograph taken of the Bull Creek area in

Results of Stratigraphic Test Pitting and Trenching of the Bull Creek Site

March 1950, two months prior to Kelly's excavations. Several of the landmarks referenced by Kelly, such as the fisherman's

Unit #1 {see Figure 71}. Beginning on the summit or parapet of the embankment fill directly overlooking the confluence of Bull Creek and the

path, are visible in the photograph.

Chattahoochee river, a 5 foot trench was put down

through the fill and extended down slope toward

This brings us to the present survey reconnaissance by A.R. Kelly of the Department of Anthropology and Archaeology at the University. Investigation was undertaken with a view to determining what changes had taken place in the condition of the Bull Creek site in the last 12 years,

Bull Creek for a distance of twenty five feet or more {Figure 73}. The purpose was to determine depth of fill at this point, and to ascertain whether slumping and erosion sinks in the embankment summit could possibly indicate subterranean pits or Indian burial areas.

how much of the site

remained in relatively

undisturbed condition,

and how profitable

further exploration on

the site might be

calculated to be.

Most of the Bull Creek

site proper, located

between Bull Creek

and the Chat-

tahoochee, at the

junction, is on pro-

perty of the City of

Columbus. (see map,

Figure 70). This pro-

perty fronts on the

river and has strategic

importance in future

industrial develop-

ment, perhaps as a

site for a municipal

dock when Columbus

eventually has a nine-

foot channel to the Gulf. The critical

Figure 73. Field photograph of Unit 1 showing depth of railroad nil. Dr. Kelly is
shown standing at upper end of trench (Columbus Museum files),

84

Kelly's 1950 Investigations

The excavation revealed clearly that the

might not have been found or excavated, and the

embankment at this point had been filled to a

indicated job was to complete the exploration of

depth of twenty feet or more and that no aboriginal

this mound. If house structures, floors, wattle and

occupation level could intervene to account for the

daub walls, refuse pits, and other features were

observed slumping of fill dirt. Further exploration

abundant, with no lenses of basket-filled dirt in

at this point is therefore not recommended to

place indicating a mound, then the 1936-1937

uncover archaeological features in the original

burial discoveries would probably relate to burials

occupation level.

made in house-floors. In this case, any large thick

black occupation level still remaining, with

Unit #2. {See Figure 71 map for reference

indications of house structures, would suggest that

to unit location}. This unit was a profile clearing of

further burials and pottery and other objects

a portion of the eroded terrace overlooking Bull

associated would continue to be found in further

Creek at a point where considerable broken pot-

exploration. Excavation #1 of unit #3 revealed

tery, animal bones, and scrap flint indicated mid-

scattered pottery under about 4 feet of recent

den spill from an occupation level, exposed by the

water-laid sands and gravels deposited in flood

erosion of the area in recent floods. No house

stage by Bull Creek or the back swirl of the Chat-

sites, burials, or other archaeological features were

tahoochee. No house floors, pits, burials or spe-

uncovered, but several bags of pottery and other

cific structures were uncovered in place. Ex-

remains were catalogued, showing that the original

cavation #2, made farther upslope from Excavation

village level at this point had a midden thickness of

#1 , revealed the buried occupation level extending

about 12 inches. The profile cleared is forty feet

upward, with some pottery in place but no features

or more from the embankment fill.

uncovered, and no indications of mound structure.

Excavation #3 was put down on the north side,

Unit #3 {See Figure 70 for approximate

facing Bull Creek, of the backfill to the 1936

locations}. This unit consists of three separate

excavations. Here a deep, crescent-like, eroded

excavations made in the marginal slopes around

bank of Bull Creek, was cut back and profiled

the site of the 1936-1937 excavations made by

along two converging shelves {Figure 74}.

Engineer Frank Lester and Mrs. Wayne Patterson.

That work {1936} uncovered the remains of 46

prehistoric burials,

accompanied with rich

grave furniture,

including no less than

3 effigy dog pots of

unusual artistic merit.

The problem here

was: had the 1936

exploration unwittingly

cut back from the

original exposed Indian

burials in the bank of

Bull Creek into what

was really an Indian

Burial mound, partially

mantled and con-

cealed by the 1926

railway embankment

fill, or were these

burials nested in deep

village midden, the in-

terments having been

made in the floors or

immediately outside

the walls of houses?

If a burial mound was

indicated, then other burials and structures

Figure 74. Photograph of Excavation Area 3 of Unit 3, captioned: Dr. A.R. Kelly, Mrs. Wayne Patterson and Doc discussing plans for work (Columbus Museum files).

85

The Bull Creek Site

In both pro-

files, rich black dirt,

full of pottery, animal

bones, flint indicated a

thick occupation level

14 inches or more in

thickness, under about

3 feet of recent soil

mantle, most of which

was detritus from the

railway embankment

some thirty feet away.

What was significant

here was that in one

place we found baked

clay daub, showing

the impressions of

wattles, extending for

3 feet in the profile.

This almost certainly

indicates a house

structure with wattle

"

and daub walls, whose floor section will be found to extend back into the unexcavated ground for

Figure 75. Field photograph with notation: Dr. A.R. Xiiiy holding;~n;;" vessel to
show better its size and shape. The dirt inside has not been removed (Columbus
Museum files).

twenty feet or more.

Close by in Excavation #3, Unit #3, was a

aeological materials. Also, that house sites, and

complete pottery vessel, of the plain ware asso-

very likely additional burials, will be encountered in

ciated with Bull Creek people, exposed about half-

further exploration at this point.

way down in the block

soil occupation zone

{Figures 75 and 76}.

No definite human

burials were un-

covered in this initial

cleaning of the face of

the cut. But the con-

ditions imply that

burials originally found

in 1936 in this area

did occur in house

floors or in the village

midden around house

walls. It is also evi-

dent that 14 inches or

more of thick rich mid-

den still remain in the

area between Bull

Creek and the bluff

overlooking the Chat-

tahoochee, at varying depths beneath the clay embankment, and that this occupation is very rich in arch-

o
t-

em

f,'

Figure 76. Photograph of vessel recovered by Kelly in 1950 (Columbus Museum files).

86

Kelly's 1950 Investigations

Unit #4 {Figure 71}. On the eroded terrace overlooking Buff Creek, at a point down slope from the bare road area frequently used for car parking, and just behind the site of a former shanty (one outbuilding in ruins remaining), this profile cut was made and cleared horizontally for a space of ten feet or more, to check black topsoil and subsoil in which considerable potsherds and other midden was showing on surface, and in the eroded bank. Ten or twelve bags of pottery all broken, and other artifacts were recovered here. The situation suggests from the amount of occupational debris, that there is a prehistoric Indian house structure to be found within a few feet of this point.

can be no doubt but that it extends back underneath the embankment to join up with the rich black Indian village midden revealed in the excavation to Unit 4.
In addition to the above, profiles were cut to determine if all of a house structure, uncovered by Mr. Lester 1936, had been removed. This portion of the site is on a path entering the embankment path, alongside the scar of the old borrow pit from which soil was removed in 1936 to make the embankment fill. Apparently Mr. Lester did remove enough of the overburden here to uncover as much of his structures as then remained.

Exploration was not continued to bring this out as we did not wish to expose valuable structures that might excite visitors to the area to

Recommendations for Treatment of Site and Development

dig into the site during the absence of the archaeologists. Enough was found to indicate the probable location of a house structure at this point. Also, that the original village occupation zone was still strongly indicated here, supporting conclusions drawn from Excavations #2 and #3.

Analysis of the results of trenching and test pitting, from the beginning of the old railway embankment and the borrow pit to the junction of Bull Creek with the river, shows that there is a long narrow, fingering section of rich, deep midden, covered with varying depths of recent fill, 6-8

Unit #5 {see Figure 71}. Almost directly opposite the railway embankment from Unit #4, on the exposed face of the high bluff overlooking the Chattahoochee, and where a path had been worn down to the river by fishermen, we found strong outcrops of black midden in the path and in

feet nearer to the borrow pit, 12-20 feet closer to the junction. Down slope, toward Bull Creek side, varying patches of this village midden are relatively unmantled by detritus from the embankment fill, and can be uncovered under 3-4 feet of overburden.

slumped dirt. Cutting

a new, clean profile

here revealed again

14-18 inches of black

midden dirt, original

village midden, pre-

served under 3 Y2 to 4

feet of the em-

bankment fill. Pottery,

flint, animal bones

were abundant. Some

charred clay running

about half way

through the midden

suggests burning was

extensive at one stage

of the occupation;

possibly some of

these represent dis-

integrated "bri-

quettes," clay-daubs

from Indian houses.

The occupation here

has an imposing depth

and is very rich {Figure 77}. There

Figure 77. Field photograph captioned: Dr. A.R. Kelly indicating the width of the black midden deposit; charred clay can be seen (Columbus Museum files).

87

The Bull Creek Site
Unit #3 excavations suggest a cluster of house sites, in which complex of structures, apparently in 1936, some 46 burials were exposed. A considerable remaining portion of this village area is still preserved, insulated underneath the heavy railway fill. Definite signs of Indian house floors were uncovered in Unit #3, and elsewhere the occupation is found to have a depth of from 12 inches to nearly 18 inches in thickness.
The narrow, fingering strip of land, precariously wedged between Bull Creek confluence with the Chattahoochee river, has been rapidly cut away by both streams; at least ten feet on the Chattahoochee side, and probably more on Bull Creek side has been eroded during the last twelve years since the site was partially explored in 19361937. Action of the streams is to undercut the banks, and subsequently extensive slumping takes place, huge blocks of embankment and underlying village midden sliding down slope. Unless some salvage procedure is undertaken, in a few years the whole point of land between the two streams will have been cut down to form a rough eroded mass with all original cultural context lost.
The immediate, specific recommendation therefore is that the portion of Bull Creek site indicated, between the creek and the river on city property, be excavated as soon as possible to recover archaeological materials and information that will be irretrievably destroyed by natural forces or by industrial development of the site. This section alone would probably occupy a trained archaeological supervisor and 5 or 6 workmen a whole summer to complete. This, even though not all of the strip indicated might not be completely excavated.
I also recommend that the collections and notes be preserved at the archaeological laboratory of the University of Georgia, to be studied, with the preparation of a report on the archaeological findings. This project of study and publication can be considered apart from the immediate recommendation for the salvage operations to recover the material that will be lost. The material and notes will, of course, be the property of the sponsor, the University offering facilities for storage, preservation, and laboratory analysis.
No treatment of the site with a view to preservation of exhibits in place of a museum is recommended. This hardly seems practicable because of the severe erosion of the area by both Bull Creek and the river. There would be not only the expense of protective covers, museum installations, but a heavy expenditure to erect pilings and insulation from flood damage. The area
88

is too constricted and fingering for effective maneuvering in any permanent planning. The only conclusion possible is to plan on exploration, as complete as possible with entire removal of all specimens to a place of safety, later to be exhibited in a museum building elsewhere.
A specific recommendation is made that the exploration be undertaken for the purpose not only to salvage the relics and data, but to preserve these with a view to their ultimate utilization in appropriate exhibits exemplifying the early aboriginal history of the Columbus area. In short, that the project be undertaken both to save the archaeological values, and in planned contemplation of a museum project that would develop. It is my sincere judgement that the publicity attendant upon the city's undertaking this work would help very materially in interesting your citizens in a museum as a development. There are many other sites, and many other existing collections, valuable materials that loyal people will doubtless make available to you as soon as a definite plan for a museum is matured.
That portion of the Bull Creek site between Bull Creek and Victory Drive, and Weracoba Creek, is also village site; how rich and complicated by house sites and archaeological features remains to be determined. This part of the area is not in as immediate danger from total destruction and might be held in abeyance, pending the results of exploration on the critical area herein described. Possibly, the archaeologist employed in exploring the critical area, could find time during his tenure to make adequate test pits in the adjoining portions. I recommend that this be done if time and nature of the summer's work will permit.
I am appending a suggested budget that would cover 10 weeks of exploration on the Bull Creek Site. This includes provision for an archaeological supervisor, the employment of 5 workers, supplies, photography, engineering recordation. The cost of preparing the materials, studying them, getting out a suitable publication, can only be estimated when the exploration is completed. For that reason I recommend that consideration of this
item be postponed (Kelly 1950:3-7).
Kelly's budget provided for 2000 man hours over a 10 week period. Pay for a supervising archaeologist was to be $75 per week and sixty-five cents per hour for five archaeological laborers. With supplies, the entire project would have cost only $2400. The project was not implemented.

Obviously, the primary purpose of Kelly's 1950 address to the Chamber of Commerce and the subsequent brief report was to convince the city of the need for additional archeology before the site washed into the river or was leveled for construction. The project produced a substantial number of artifacts. The artifacts, presently curated by the University of Georgia, in Athens, have been examined by several researchers. Artifacts have been accessioned by excavation unit but aside from a few handwritten field notes on small fragments of the paper bags, no record of the excavations exists today. It is fortunate that a copy of Kelly's report to the Chamber of Commerce and Patterson's scrapbook have been preserved at the Columbus Museum.
This record of Kelly's brief investigations in 1950 provides valuable information concerning the layout of the village area but is most critically important to interpretation of the "Bull Creek Cemetery." Kelly was able to confidently demonstrate that the cemetery area contained ample evidence of structures. Kelly proposed that the burials excavated by Lester lay beneath the floors and surrounding the walls of a complex of house structures (Kelly 1950:5-6). The possible presence of structures in the cemetery had been considered but dismissed by Lester (Lester 1938:70).
Kelly implies, with specific reference to associated "village middens," that these cemetery area structures were domestic houses. Typically Lamar houses contain burials beneath the floor or just outside the structure walls. Kelly does not address the possibility that some form of ceremonial structure existed within this relatively small portion of the site which appears to contain the vast majority, if not all, of the interred burials.
The supposition that a mound might have once existed above the cemetery appears to have been discarded without further comment as the concept of habitation structures was embraced. Apparently, Kelly found no evidence of basket-loading but conceivably, the

Kelly's 1950 Investigations
scale of Kelly's excavations was not adequate to sufficiently explore the possibility.
Kelly stated that the village midden extended along Weracoba Creek all the way to Victory drive and at the time potentially contained well-preserved deposits. This area would encompass the western half of the Weracoba Creek site which had been judged distinct by the Fenengas during their 1944 survey. Kelly's observation provides important evidence for reconstructing the extent of the Bull Creek village. By incorporating the Weracoba Creek local (presently Georgia State Site Files Number 9Me379) into the Bull Creek site, the village area would extend from the river for a straight-line distance of approximately 500 m along that tributary. While impressive, this boundary estimate does not reach the extent required for a village area covering threequarters of a square mile proposed by both Lester and Patterson.
Documentation accounting for the lack of implementation of Kelly's proposal for additional work at Bull Creek has not been found. The most reasonable likelihood was simply that the cost was too great. However, Mrs. Patterson still continued to press Kelly to prepare a report of Lester's excavations. Perhaps Kelly's final endeavor to undertake the manuscript is referenced in a March 7, 1951 letter to Mrs. Patterson. In that letter Kelly discusses the realization that Bull Creek with its Fort Walton influence actually differs substantially from basic Lamar.
I know you have always been concerned to get Bull Creek into print. I have talked this over with Bill Sears. We are prepared to write up Bull Creek, or to start on it at least, this Spring; working with the materials I brought back from last year's test excavations, and then going into the old notes and materials if we can get access to the stuff at Ocmulgee. I think we can produce a valid report, and study of the Bull Creek stuff I brought back in the laboratory now is indicating a much wider spread of archaeological materials, with regard to source, than we had expected.

89

The Bull Creek Site

For one thing, while the rim treatment, paste, and general morphology of the bulk of the stuff at Bull Creek is definitely in the time level and "feel" of Lamar, the specific designs have a Savannah stamped tradition; also there is a lot of pottery that must refer to influences coming up the Chattahoochee from Florida, Fort Walton influences. This means a very interesting variant in Bull Creek of the widespread Lamar manifestation.
I probably won't have the money to publish the report right away but at least we intend to get the manuscript ready for publication if at all possible in the near future. It has occurred to me that the Columbus people, if they knew an important manuscript had been prepared, with beautiful illustrations and drawings, ready in the press for publication, that some help financially might be forthcoming. I think as much as $500 would insure action by the Press. The expense of preparing the publication can be borne by our Department here, utilizing existing personnel and laboratory facilities. At any rate, I knew you would be interested, and I wanted to keep you informed
(Ledbetter 1995b:49).

Donalsonville, from a soldier at Fort Benning, who is a trained archaeologist, former Carnegie Museum man, and who apparently is permanently stationed at Benning. I have written him to contact you, as I know you would want to meet him, and he might be the boy to help you with any outstanding problems in connection with your museum project. I know you are probably exasperated with me for my tardiness, but I would beg your kindness in remembrance of my good wishes and efforts to
help the best I can (Ledbetter 1995b:53).
Isabel Patterson died in 1955 but the new generation of archeologists continued to conduct excavations at the Bull Creek site. Sgt. David W. Chase conducted limited excavations as part of his ongoing investigations of the Columbus/Fort Benning region and University of Georgia student Frank Schnell, Jr., with the assistance of personnel from the Columbus Museum conducted more extensive excavations. Details of this work will be presented in the following chapters.

Kelly failed to write that Bull Creek report the following spring. Kelly's role in the Bull Creek saga waned following the 1950 excavations. Mrs. Patterson continued to work with the Bull Creek material with efforts directed towards exhibits containing her beloved dog pots for the new Columbus Museum as well as the Smithsonian Institution (Figure 78). Kelly continued to correspond with Patterson and in a letter dated August 6, 1953 he offered support by guiding a new generation of archeological attendants.

In the meantime, I am enclosing a letter received during July while I was in

Figure 78. Photograp .dated 1957 shOWing a Bull Creek dog pot on display at the
Smithsonian Institution (Columbus Museum files).

90

Chapter 5 Schnell's 1959 Excavations

Nearly a decade following Kelly's brief investigation at Bull Creek, Frank Schnell, II. organized and successfully completed a systematic excavation project at Bull Creek. A manuscript which details those excavations, The Bull Creek Site in Western Georgia (Schnell 1963), includes a substantial body of background material and ceramic analysis that has been in-

obtained from a vocational rehabilitation program operating in Columbus for juvenile delinquents, under the direction of the Columbus Craft Shop, Inc. A total of two to three days out of each week was spent excavating on the Bull Creek Site . . . . A total of 950 square feet was excavated in two areas, XU-I and XU-II, in stratigraphic sequence, along with profiling several banks and relocating
excavation units (Schnell 1963).

corporated elsewhere in this report. The following paragraphs relate to the implementation and results of the 1959 field investigation. Pertinent information has also been extracted from Schnell's field notes curated at the Columbus Museum.
Schnell excavated intermittently at the Bull Creek Site during the summer of 1959. His goals were to:

Figure 79 shows a 1957 property map that encompasses the Bull Creek site. The city property markers on the plat were used as fixed reference points for Schnell's grid. The locations of the 1959 excavation units is shown in Figure 80. A series of excavation photographs are shown in Figure 81. These photographs relate primarily to the major excavation unit located near Kelly's old excavation unit 4.

. . . obtain a larger stratigraphic sample ..

. determine if any structures remained in the penin-

sular portion of the site, and to determine if the

site warranted further extensive excavation before destruction by in-

_______.

"-__ _ L .(h I~INA~O",C o,"'o! ---l:~L ~

dustrial development. The pro-

'1"c t o !Sv

Ttl tT $),....,...

ject was carried

out as a coop-

erative agree-

ment between

the University of

Georgia Labo-

ratory of Archa-

eology, who pro-

vided the supervision and equipment and the Co-

- \ .... _--

.....

'-"'\ ~<>v~

lumbus Museum of Arts and Crafts, which provided labor. This labor was

,
'\ "~
!j

Df'lAWII\lG

e NO

I~~~ A.

p'"'Of"!!..f'l1v ...,...AIIl~ .. ~P.

ClTV 01" COl.U"'!W~ TO .J W ""0001'10" ~ PAl'IT Of \..UolD \..OT~ ~o, ~'. ':C~f.'Tll. ""'E'!I'Ef'I".'

CITv OF C.Ol..UMe:.uS, (;,t.Or'lGI.o..

o"".et 0" crrv ["G'NEt1"l

I'It""~IQH'!>

~c...ve I". zoo

,;:~ ..~:.'":. ;::;:~. ~ O'!.Ct .. M.'" 1':.'~7

Figure 79. Copy of a 1957 plat of the J. W. Woodruff property containing the Bull Creek site (Schnell's main excavation unit is indicated).

91

The Bull Creek Site

APPROXIMATE LOCATION PROFILE2
o ~/
X-I

\ APPROXIMATE LOCATION PROFILE 3

N
!

0

20

40

I

FEET

GRID POINT N500
ESOO
FIRST USED IN X-2
Figure 80. Plan map showings areas of Schnell's 1959 investigations (adapted from field map on file, Columbus Museum).
92

Schnell's 1959 Excavations Figure 81. Series of photographs showing the 1959 excavations in progress. Field director, upper left; remainder show XU 2 (Columbus Museum files).

The Bull Creek Site
Schnell prepared a series of profile drawings, taking advantage of earlier excavation units which had never been filled. Figure 82 shows a profile made in the railway cut just south of Patterson's House Number 1. This is Profile Number 3 shown in Figure 2. The site's occupation surface is shown to be 1.0 to 1.5 feet below the present surface and overlain by alluvial sand. A portion of the profile had been worn away by fishermen crossing over this bank from the railway cut to the river. This path is visible in the 1950 aerial photograph (see Figure 72 in Chapter 4).

A second profile drawing (Figure 83), shows the depth of the railway fill covering the portion of the site referred to as the peninsula, which extends above the confluence of Bull Creek and the Chattahoochee River. This profile drawing shows the prehistoric occupationallayer beneath four feet of railway fill and four other strata. Schnell notes that "closer to the confluence of Bull Creek and the Chattahoochee River, this fill becomes much thicker" (Schnell 1963). Additional profile drawings are provided in following sections describing Schnell's excavations.

Fishermen's Path Pit

1l2r-:,.qf [JI]
~
1;::-::,'1

Humus Yellow - Tan Sand Occupation Layer Red - Tan Clay Yellow - Tan Clay

N
Tap Root
9ME 1 XU-II
Profile South of Patterson's House # 1
Schnell: 1959

Figure 82. Profile 3 drawing of bank cut near Structure 1 originally excavated in 1936 (redrafted from Schnell 1963).
94

Schnell's 1959 Excavations

Cache of Charred CornCobs

Limit of Profiling

V':~':'~I
r=-=1 t=:::"=:.I 1':.';:~?:::1
~~ .4....f.....f..l..
~
E3

Railway Fill (Loose Sand - Clay Mixture) Heavy Black & Red Marblized Clay Yellow Sand Old Humus Tan Sand Occupational Layer Sterile Red Clay

-5 -6 -7 -8
-9

-10

9ME 1 XU-II

Profile made on River Bank at West of XU - IT

0

1m

0

3ft

Facing East

Figure 83. Profile drawing showing the railroad fill at a location nearer the mouth of Bull Creek and adjacent to XU 2 (redrafted from Schnell 1963).

Schnell's field notes indicate that an old excavation cut near Lester's Structure 1 was examined in some detail. This appears to be the area shown in Profile 3 (see Figure 82) which corresponds to Kelly's Unit 5 excavated

in 1950. A comparison of Schnell's field note entry of June 29, 1959 and Kelly's description of Unit 5 shows the area to have remained essentially unchanged during that decade.

95

The Bull Creek Site
I tested in a hole shown to me by D.W. Chase on Sunday. There are very good indications of a hard packed sun baked floor. This is an old excavation on the opposite side of the railway embankment from XU-2 and slightly north of the excavation.
An area about 1 5 feet long and 2-3 feet deep was cleared. A floor appears to run the full length with a heavy ash lens intrusive through it (fire basin). There is about one foot of midden above this and at least one foot of midden below it. Above the upper midden is a heavy mantle of tan sand with the humus on top of that. There is about 10-1 5 feet of flat area before the railway fill is reached. The railway fill is also very low here. It seems to be a good possibility for excavation. The black midden is deeper here than any other
place profiled (Schnell field notes 6/29/59).
Coupled with Kelly's observations of 1950, Schnell's description provides evidence for another well-preserved structure that would have been located approximately 20 to 30 m south of Lester's Structure 1. An extremely important observation made by Schnell concerns the presence of foot-thick deposits of midden both above and below the house floor. This suggests relatively great intensity and duration of occupation, meaning that the Bull Creek phase occupation extended over a sufficient length of time to allow rebuilding of structures on top of middens associated with earlier structures.
The text from Schnell's manuscript relating to the excavations follows. Information from the field notes are included.
XU-I consisted of four 5 x 5 foot squares excavated in the vicinity of an old well and between two old excavation units, one made by Kelly (Unit 2) and one made by Chase (Test B). This small excavation was made to determine the merits of excavating on the higher ground siightly away from the river {see Figure 80}. This excavation unit, though small, did show that the upper portion of the peninsular area had been too heavily disturbed both by recent occupation and extensive testing by others, to warrant further excavation.
One feature was located in XU-I. This was a shallow basin shaped pit {Figure 84}. This pit was circular, 4.5 ft in diameter, and 0.2 feet deep. This feature contained only a few Lamar
96

plain sherds. It was located in the southeast corner of Square 1 at a depth of 2.5 ft.
Soil stratigraphy in the square, from top to basic red clay, included 0.6 ft of yellow-brown sand immediately below a thin veneer of humic leaf mold, 0.55 ft of dark brown charcoal-flecked midden, and 0.55 ft of yellow-brown sand, resting directly on the basic clay at a depth of 1.9 ft.
After finding that excavation in the area of XU-I would be unprofitable, operations were moved farther out on the peninsula, to XU-II {see Figure 80}. It was there that major stratigraphic testing was undertaken and where the new crew remained for the rest of the period of excavation. A total of seventeen 5 x 10ft tests were excavated here. This area was the only remaining easily accessible and largely undisturbed area on the site. Although this area had been occupied in recent times, as evidenced by the large amounts of modern material, the talus from the railway cut, built before the recent occupation, acted as a "buffer" and protected the lower, prehistoric occupation zone. {Figure 85} depicts a schematic representation of the area.
A grid system was laid out to cover the XU-II area, using the Huscher variant of the Wheeler system. A datum point (labeled N500 E500) was arbitrarily placed to the west and south of the area to be excavated, making all grid coordinate readings to the east and north of that point. As an example, the first test excavated in XU-II came within the coordinates N490-500, E525-530.
SQUARE I OF XU - 1 GRID
N
A
1ft
:======:'30 em
10p of Pit 2.5 ft Below Surface Baseof Pit 2.7 ft BelowSurface
Lamar Pit
Figure 84. Plan map of pit found in XU-I (redrafted from Schnell 1963) .

Schnell's 1959 Excavations

Creek Bank

River Bank ~ W

E

-1:::',::1
~ ~

Railway Fill Talus& Recent Humus Old Humus Tan Sand OccupationalZone Sterile Clay

9ME 1 XU-II

Schematic Profile Schnell: 1959

o

2m

o

5ft

Facing South

Figure 85. Schematic representation of the soil stratigraphy adjacent to XU-2 (redrafted from Schnell 1963).

Individual points referred to in the excavation notes were also designated aerially by this system. As may be seen by reference to the map {see Figure 80}, this placed the imaginary zero point well to the south and west of the whole site. This system is therefore expandable to cover all future excavations on the site. This grid system was tied into a number of temporary bench marks on the site as well as two permanently placed bench marks of the city of Columbus.
The first feature to be uncovered in XU-II was a pit (Feature 1) located at grid coordinates N494.6, E528.3 in the first 5 x 10 excavated. This pit was defined 2.2 ft below the surface, was 1.5 ft in diameter and 1.2 ft deep. This pit contained only a few Lamar sherds, charcoal flecks, stream pebbles and badly decayed bone

which was unidentifiable other than it was not human. To the north of this feature was another, similar pit, located at grid coordinates N500, E428. This pit (Feature 2) was 1.9 ft below the surface, 2.0 ft in diameter and 0.9 ft deep. The most interesting pit encountered in the XU-II excavation was located at grid coordinates N509.9, E528.3. This pit (Feature 3) was encountered at 2.2 ft beneath the present surface, was 1.6 ft in diameter and 1.1 ft deep.
Feature 3 contained one complete Lamar Complicated Stamped bowl with very slight lip notching. This bowl had apparently cracked in firing and was discarded complete (except for the crack) in a refuse pit. Nesting inside this bowl were several large sherds of Lamar Complicated Stamped and Lamar Plain {Figure 86}.

97

The Bull Creek Site

(Schnell 1963). Figure 87 illustrates the bowl in greater detail.

Schnell's excavations provided the largest, systematically recovered collection of pottery known to exist for the Bull Creek site. The screened collections have formed the basis for defining the composition of the Bull Creek phase ceramic assemblage. Schnell's characterization of the assemblage is included in a later chapter.

Figure 86. Field photograph showIng vessel fragmentS exposed in

The XU-II excava-

excavated Feature 3 (Columbus Museum files).

tions also produced a number

of large sherds and some

The Lamar Complicated Stamped vessel fragments appear to have been a large, straight

partially reconstructible vessel fragments that provide excellent examples of the decorative

and high necked jar, while the Lamar Plain vessel

motifs present at the site. Figures 88 and 89

was a wide, dish-like shallow bowl. The complete Lamar Complicated Stamped bowl proved to be

illustrate three examples.

9.25 inches in diameter and 4.8 inches deep

o3
em
Figure 87. Photograph of Lamar Complicated Stamped Bowl from Feature 3 (Columbus Museum files).
98

Schnell's 1959 Excavations

o

3

~~~-.i!iiiiiiiiii~~i

em

:~.':.;',".' : -, : '. :.:. :.' .... ' .. :', . .

...~,,\ -. <...,.:.r."

...-:l':-.~:. '.'t'!-11"-":":-~'',', ."...'. - : "

~'''1.).:r. \4<'.t.:>':1 ~,~~:'':-''':'I$''\\:J~
.

;

'~\\-:;::::::::::;"'" ~t J~\ :').;'i.:.. i"~....'-.~.;;:,:.~:.~:;;r~.
{~~<~'~2;,:~"

.:..~.' ....

- ---..'..' .. . "

........

... ......

....

....

_---- ... ............. ...

---------- -------

. . ,,'"
.."

Figure 88. llIustrations of two Fort Walton-style vessels from XU-2.

99

The Bull Creek Site

o

3

IPSj5-.j5-;;;;;jjjjjjj~~

em

Figure 89. Example of bold incised vessel fragment from XU-2.

Figure 88 shows two Fort Walton-style vessels. The upper, partially reconstructed vessel is a common bowl form with a typical decorative motif of zone punctations. The punctations are found in alternate bands that encircle the neck of the vessel in an undulating pattern. The second vessel illustrated in Figure 88 is an example of a hand-eye motif. This motif occurs consistently on late Fort Walton sites and has been found on other Bull Creek and Stewart phase sites in the region (Broyles 1962; Hally and Oertel 1977).
The rim sherd illustrated in Figure 89 represents a moderately large constricted orifice vessel (estimated to measure 32 em in diameter). The sherd is decorated by deep and very wide incised lines which average 4 mm in width. The motif is quite similar to that found on an effigy vessel found in a later investigation of Bull Creek conducted in 1981 (see Figure 105 in Chapter 6).

Schnell's field notes indicate that a number of postmolds and possible evidence of a burned house floor were found in XU-IT. The location of the postmolds and the larger pits are shown in a composite plan map as Figure 90. These features indicate proximity to a domestic structure if not a direct presence within the excavation unit. The postmolds may represent screens or supports related in some manner to the pits; elsewhere on the site similar pits were found outside of structures.
According to field note entries dated July 10, 1959, evidence of a possible burned house floor consisted of a hard packed floor in squares N51O-520, E520-525, extending into the northwest corner of the adjacent square to the east. Fragments of daub and charcoal were found above the floor. A possible burned postmold was found in square N51O520, E525-530. An entry dated July 12 notes a possible fire pit in the floor.

100

Schnell's 1959 Excavations

recovered during his excavations provided no evidence to support the presence of both an early Bull Creek phase occupation and a late Bull Creek or Stewart phase occupation. Schnell's conclusions, which will be examined in a later chapter of this report, were that only an early Bull Creek phase occupation was present at the Bull Creek site.

N
~

o

10

20

FEET

Schnell's systematic field techniques provide basic information necessary for interpretation of the Bull Creek site. The fundamental step of establishing a site grid tied to property markers allows the reconstruction of an excavation plan map showing locations of most of the excavations on the site through time. Perhaps the most important result of Schnell's excavation was the recovery of a large pottery sample from a controlled (screened and stratified) context. Schnell's interpretation of those data will form a basis of much of the ceramic examination chapter.

Schnell's inves-

tigations were to be the last

professionally conducted ex-

Figure 90. Plan map showing the location of pits and postmolds in XU-2 cavations in the heart of the

(adapted from field note entry).

Bull Creek site. A quarter of

a century later, work on the

Schnell's findings appear to confirm

periphery of the site would be conducted in

Kelly's 1950s interpretations of a structure

connection with the Columbus Riverwalk

based upon the excavation of nearby Unit 4 in

project. A review of that work will be

1950. The richness of the deposits examined

presented in Chapter 7. However, before the

by Schnell provide ample evidence of extended

recent investigations are detailed, a summary

occupation during the Bull Creek phase. How-

of the minor investigations conducted at the

ever, Schnell's analysis of the ceramics

site will be presented.

101

THIS PAGE LEn BLAHK

Chapter 6 Minor Investigations

While the excavations conducted by Lester in 1936-1937, Kelly in 1950, and Schnell in 1959, provide the bulk of the archaeological data needed to interpret the Bull Creek site, there have been several minor investigations conducted from the mid-1940s into the early 1980s which also provide valuable bits of information. These investigations include additional survey work, testing, excavation of features and site assessments related to proposed development of the property. Some of these studies may be considered peripheral to the excavations conducted in the heart of the Bull Creek site but all provide data necessary to reconstruct the extent of the Bull Creek village. One goal of reviewing this work is to confirm or reject the claim of Lester and Patterson that the Bull Creek village covered an area of three-quarters of a square mile (Lester 1938:27). This chapter will begin with the survey of Frank and Barbara Fenenga conducted in 1945 and continue with the known events relating to the other investigations.
1945 Survey by Frank and Barbara Fenenga
While stationed at Fort Benning, Franklin Fenenga and Barbara Wagner Fenenga conducted a surface survey of the Columbus area. Much of this work focused on sites in the lower Bull Creek drainage. Nine of the 15 sites identified lie near the mouth of Bull Creek. The report titled An Archeological Survey of the Vicinity of Columbus, Georgia (Fenenga and Fenenga 1945) contains line .drawings of numerous artifacts and detailed ceramic descriptions which allows temporal context to be readily determined for all sites. The collections are presently curated at the Columbus Museum. This obscure report provides some of the only survey data for an area which has since been affected by extensive commercial development.

The Fenenga's defined site boundaries by surface scatters of artifacts. Apparently, site boundaries were approximated based upon the extent of surface exposure. In certain instances, Bull Creek for example, this resulted in underestimation of site size and splitting into multiple sites. At the time of the Fenenga survey there was no statewide mechanism for assigning site numbers. The Fenenga's identified their sites by numbers 1 through 15 and named each site. As part of the Smithsonian survey (Huscher 1960), David Chase completed Smithsonian Institution site forms for the Fenenga sites and attempted to reconcile their locations with previously known sites such as Bull Creek. This work was only partially successful. As late as 1994 there were still substantial discrepancies within the State of Georgia Site Files concerning the Fenenga sites (Ledbetter 1994b:26).
The Fenengas identified two site locales within the confmes of the Bull Creek site area (Figure 91). These were identified as the Weracoba Creek site (Site 2) and the Riverside Driving Park site (Site 4). Two additional sites (No.3, the Lumpkin Superhighway site and No.5, the Confluence site) were identified across the channels of Weracoba Creek and Bull Creek respectively. Both of these sites also contain Bull Creek phase occupations (Fenenga and Fenenga 1945: 1631). The Weracoba Creek site represents the most important site locale examined simply because the area was documented by no other investigation. The Fenenga description of the Site 2 area follows:
The site is on the right bank of Werocoba Creek at the point where it is crossed by Lumpkin Superhighway. The site is about 150 yds. in diameter, extending on both sides of the highway, the construction of which has destroyed about 60% of the total area of the site.
103

The Bull Creek Site

I

,I \

\. \

\
\

\1
\\\:~,\ ;~N~~,/

II ""-...:

,,

\ , ~Q""_/

\
\

\
\

,-

,(i'"'-'U-----<, ,

,I

/J

\

\ \\

I

'\

VtI '~~<;:?t~

\

\/ 5 ,,\ ... <,

Figure 91. Copy of a sketch map showing the locations

of Fenenga Sites 2-5 (source Fenenga and Fenenga

1945:12, Map 3).

Local informants say that a party of archaeologists from the University of Georgia excavated here during construction work which would indicate that this is the Bull Creek Village excavated by Lester. If this is the case "Bull Creek" is an unfortunate misnomer . . .. A great quantity of pottery sherds are found throughout the two feet of occupation zone but flint chips, stones and fragments of shell are extremely rare. No bones
were observed (Fenenga and Fenenga 1945:16).
Although the area collected as Fenenga Site 2 lay well to the east of the richest part of the Bull Creek site and Lester's excavations, a substantial amount of material was still found. The Fenengas collected more than 400 sherds and one projectile point from the site. Figure 92 shows examples of Fenenga line drawings of representative sherds from the site. These sherds clearly conform to Fort Walton and Lamar types of the Bull Creek phase,

. ....!!~.....:"':.:.:=...~..:::::.;... J.:
Figure 92. Examples of ceramics from the Werocoba Creek Site (source Fenenga and Fenenga 1945:17)
104

The Weracoba Creek site represents a portion of the Bull Creek village, with evidence of intensive occupation. The locale is enclosed on three sides by a large bend of Weracoba Creek, which would have formed something of a geographical boundary. We know from other investigations that the Bull Creek village extends from the Chattahoochee River to join this locale, but subsequent research has not been conducted to determine if the village extends farther upstream.
The area identified as the Riverside Driving Park site (Number 4) is a portion of the Bull Creek site which surrounds the borrow pit. Artifacts illustrated by the Fenengas are appropriate and most of the information presented conforms with other reports. Their description of Site 4 follows:
The site is at the junction of Weracoba Creek and Bull Creek and is on the right bank of both streams .... The essential frontage of the site is on Bull Creek, being separated from Weracoba Creek by a long sloping bank. The site is about 100 yds. in diameter. Its central portion has been removed by a borrow pit of recent origin which extends into the subsoil (sandy red clay) to a depth of about 4 ft. The soil of the site is dark and it contains a great many potsherds (290 collected) but very few flint chips or other stones. No special features were observed in connection with the site, the maximum depth of deposit of which seems to be less than 24 inches. All of our collections were made from the exposed face of
the borrow pit (Fenenga and Fenenga 1945:26).
The remaining two sites identified near Bull Creek (sites 3 and 5) are both large and contain artifacts of several components. Bull Creek phase sherds are present on both sites but appear less common on Site 5, which lies directly across the mouth of Bull Creek. Both sites were severely disturbed by development in subsequent years. Site 5 was recently reexamined as part of the Riverwalk survey (Ledbetter 1994). The only additional information on Site 3 relates to Chase's attempt to relocate the area in order to prepare a site form. Chase met with minimal success.

Minor Investigations
Much of the site is covered by "boomtown" type establishments-juke joints and the like. Behind this is a rather thick woods where surface finds are not apparent. Did a little testing but with negative results. May have been a bit off the
beam. Will return (Smithsonian Institution site form: 1959).
1950s Investigations of David Chase
Aside from the information contained in an unpublished manuscript titled The Bull Creek Site, A Type Station of the Middle Lamar Culture, which will be cited as (Chase 1957), relatively little is known concerning these excavations. Chase appears to have excavated at Bull Creek with the goal of acquiring a ceramic collection that could be used to refine the chronological sequence for that portion of the Chattahoochee valley. In the mid-1950s, the artifacts from earlier excavations were stored by the National Park Service and the University of Georgia and were not available for study by Chase. Figure 93 shows the site area as it appeared in a 1955 aerial photograph at the time of Chase's work.
A sketch accompanying Chase's manuscript shows the approximate locations of his test pits (Figure 94). Descriptions of profile stratigraphies are presented for two of the excavations (Trench D and Test E) but notes relating to the excavations could not be obtained. The location of the artifacts has not been determined. The artifacts from this site, like others examined by Chase, should have been curated with the Columbus Museum or University of Georgia in Athens. Most of Chase's papers are stored at the Columbus Museum, although some material has also been distributed to the University of Georgia and the Smithsonian Institution. The Smithsonian comes into play because Chase worked closely with Harold Huscher and the River Basin Survey project. In all probability, and with further effort, the artifacts and additional notes from Chase's excavations will be found at some institution. Because most of Chase's manuscript deals with analytical and theoretical concepts unrelated to fieldwork, only a portion is presented in this chapter.
105

The Bull Creek Site
Figure 93. Portion of a 1955aerial photograph of the Bull Creek site area (Columbus Museum
files). 106

....
" ,"
,', ,'.
. .;'
'," ':','
.... :
....

rr ,~. , ,
~ / OldExcavation
'-~j r--l

I

I

I

I

-:

I

1

I\ JI

-:

'.':
,','

Minor Investigations

J
GN

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

Figure 94. Copy of sketch map prepared by Chase showing location of excavation units (the landmarks are not drawn to scale or accurately portrayed).
107

The Bull Creek Site
Based upon Chase's sketch map, he excavated a total of four test pits and one trench. The tests were excavated adjacent to the erosional feature or gully that parallels the railroad embankment. One of the units (Test B) would have been rather close to Kelly's Unit 2 of 1950. The introductory part of Chase's manuscript provides an intriguing glimpse into the historically pervasive documentary vacuum which has always enveloped the Bull Creek site. As noted, at the time of Chase's writings, most Bull Creek documentation was inaccessible. Chase's comments on the earlier investigations apparently were based on the brief published account found in Patterson's 1950 paper in an Early Georgia. Chase had little or no contact with Isabel Patterson while working at Bull Creek (David Chase, personal communication, 1995). Selected paragraphs from Chase's manuscript follow. Sections of the manuscript that discuss ceramic attributes and cultural contexts will be presented in a later chapter.
This site, located between the Victory Highway and the Chattahoochee River some four miles south of the city of Columbus Georgia, was initially explored by Mrs. Isabel Patterson during the early 1930s. This venture was launched by the discovery of an exposed burial on the south bank of the Bull Creek mouth not far from its junction with the Weracoba Creek. During the course of this exploration, Mrs. Patterson and her assistants explored several burials and house sites. In at least two of the former, effigy dog pottery vessels were found. The bulk of the utility pottery fell into a recognizable culture product of the Lamar Culture. Predominant among the sherd types found at the time were Lamar Complicated Stamped, Lamar Plain and Lamar Bold Incised. Details or notes relating to house plans or post patterns are not known to this writer since the notes in question have never been published nor are they readily available at this time.
Sporadic testing was conducted from the autumn of 1955 to the present (1957) on this site not only to confirm the findings of Mrs. Patterson but also to determine stratigraphy, culture change or any factor not disclosed by her during the course of her explorations ....

Pipes are infrequently found, at Bull Creek only three stems and one bowl fragment came to light, hundreds of pottery disks and one of stone appeared. This is not surprising since the disks prevailed from earliest Lamar times into historic period. Objects of shell were absent as were those of bone. Occasional areas yielding refuse animal bones were found. No projectile points were found during our explorations at Bull Creek, but we suspected that these would be of the small isosceles triangle found elsewhere on sites of the same period and culture . . . .
We suspect that Bull Creek represents an early habitation site of the Middle Lamar Period due to the presence of a small amount of typical Early Lamar pottery included in the lower levels of refuse. At the time of the occupation of this site, which was quite a large town, the Lamar culture had achieved domination of the middle Chattahoochee Valley and were thus able to move from the large inland sites of Patterson and Singer to the River way which they finally controlled. No mounds are associated with Bull Creek and it is suspected that although the tradition of mound building had not perished by Bull Creek times, it was definitely on the way out . . .. Whether the dog pots found by Mrs. Patterson represent a relationship to the Cult is a matter of pure speculation. The general opinion is that they were not.
Bull Creek will be destroyed by the end of 1963 by the River Development Program. This is somewhat unfortunate since much can still be learned from this large site about the Lamar people and their climactic society (Chase 1957).
Chase's excavations were placed to the east of the railroad embankment and beyond the most productive part of the site. The stratigraphy described for Trench D consisted of a humus zone overlying a dark sand (refuse zone) which extended to a depth of 16 inches. Below this, a light sand zone containing charcoal flecks but no artifacts extended to 25 inches. Sterile loam clay lay below. The stratigraphy described by Chase is similar to but more shallow than that described a decade earlier by the Fenengas (Fenenga and Fenenga 1945:26).

108

The second profile described by Chase was for Test E, which lay at the eastern edge of the borrow pit. The stratigraphy here consisted of a thin humus zone above a light grey loam refuse zone which extended to 13 inches below surface. This light zone overlay a dark loamy midden that extended to a depth of 18 inches. Sterile brown clay lay below.
Chase's map shows the locations of a number of old excavations. The excavations to the west of the gully can probably be attributed to the WPA and 1950 excavations. However, the excavations to the east do not appear to match any locations of earlier professionally supervised excavations. These old excavations may be evidence of unrecorded archeological or quasi-archeological digging or quite possibly pothunter looting activity.
Chase's investigations at Bull Creek provided him with sufficient information to characterize many aspects relating to the Lamar period at the site and within the region. The nature of these characterizations will be further examined in a final interpretative chapter of this report.
1960 Assessment by Huscher as Related to Construction of the Walter F. George Reservoir
While Harold Huscher did not personally investigate the Bull Creek site, he did include an assessment of the potential impacts of reservoir construction in his 1960 report Appraisal of Archeological Resources of the Walter F. George Reservoir Area, Chattahoochee River, Alabama and Georgia (Huscher 1960): Relying to a great extent upon the work of Chase coupled with the limited amount of available documentation, Huscher briefly reviewed the history of the site and made a series of recommendations concerning the site. Huscher and Chase were also responsible for preparing the first official site form for Bull Creek. Huscher's assessment follows:

Minor Investigations
9Me1, the Bull Creek Site, is an important Lamar site, regarded as the type site for the Columbus (Georgia) area variant of Late Lamar. The site lies on a point of terrace just north of the mouth of Bull Creek, at elevations between 230 and 240 feet, which throws it into the A3 (important but will not be inundated) category; however, since it lies in the most probable area for harbor development, the dredging of harbor facilities and turn-around basin, and the consequent commercial exploitation, it is regarded as of the highest priority. As indicated under "Previous Archeological Work," the site was excavated extensively in 1936 by Mrs. Wayne Patterson and Mr. Frank Lester, in cooperation with the University of Georgia. Additional excavations were carried out in 1950 by A.R. Kelly of the University of Georgia, assisted by Joseph Mahan. In 1959, Frank T. Schnell, Jr., under a cooperative arrangement between the Columbus Museum of Arts and Crafts and the University of Georgia, mapped the site in detail, relocating, where possible, check points of the previous excavations, and opened two additional squares, uncovering part of a house pattern. The site
should have additional work (Huscher 1960:106).
Huscher was one of many archeologists to stress the importance of fullscale investigation of Bull Creek before complete destruction took place.
We also have records for three additional site investigations conducted in the 1960s. One represents something of a student's project, which recorded information from a looter's pothole. The second relates to archeological remains exposed during commercial development of the area north of the previous Bull Creek excavations, an area to be known as the Go-Kart site. The remains found at Go-Kart by personnel of Columbus Museum are associated with a large historic Creek settlement of the early nineteenth century, which partially overlies the late Mississippian Bull Creek village. A few years later, a second large Creek feature was exposed during construction on the north side of Victory Drive. The area containing the second feature became known for a short time as the Dolly Madison locale after the factory which was being constructed at the time.
109

The Bull Creek Site

Following excavation of the second feature at Dolly Madison, Frank Schnell, Jr., prepared a manuscript describing the excavations (Schnell 1970). Schnell combined the two locales within a single site, 9Me50, which he named the Victory Drive site. In that manuscript, Schnell noted the probability that 9Me50 and 9Me1 overlapped (Schnell 1970).
o XU-B
9Me50

Bull Creek Site

\

9Mel

1960 and 1969 Investigations at the Go-Kart and Dolly Madison locales.

Prior to 1960, excavation at the Bull

Creek site had been limited to the area

extending north from the mouth of Bull Creek

to the area of Lester's House Site 2, approx-

imately 300 m. Lester had inferred that the

village area extended over a much larger area,

but the village portion of the site had been

generally plowed away

(Lester 1938:27). In

1960, a large pit dating

to the historic Creek

N
~

o

50

m

period and a number of postmolds were uncovered south of Victory Drive but well to the north of Lester's excavations. This area became known as the

Go-Kart site. In 1969,

another pit was found

north of Victory Drive

and was named the

Dolly Madison site.

This area also produced

Creek material.

Schnell noted a pro-

bable overlap between

the two areas and

assigned a single site

number, 9Me50.

Figure 95 shows

Schnell's map of

9Me50 and the

locations of the

excavated features.

Figure 96 illustrates the

manner of the initial

discovery of the Go-

Kart features.

Figure 95. Map of the Victory Drive site showing the Feature locations (adapted from Schnell 1970).
110

Minor Investigations
Site of Indian Village Unearthed By Kart Race Track Developers

Kart racing, a modern

fad of a modern people, is

indirectly responsible for a

new archaeological find here.

On the site of what is

soon to be a kart racing

track, the builder yesterday

literally "stumbled across"

the remains of an Indian

village on Victory Drive.

Joe Mahan, curator of

the Columbus Museum of

Arts and Crafts, said the

village is definitely a part of

the Coweta settlement, the

main part of which has been

uncovered previously in

Russell County.

The village which began

taking shape yesterday is

directly across the river from

the Russell County village.

NealWickham, president

of Go Kart Sales Inc., who is

developing the tract across

from Municipal Airport, was

supervising a work crew

yesterday morning in the

building of the race track.

He said he kicked a clod

of dirt and saw a piece of

pottery.

He scratched

around and found other

pieces of pottery.

Wickham, who is a

member of the Coweta

Memorial Assn., a group

devoted to the study of local

Indian lore, quickly

recognized that here was

something which needed

further investigation. He got

in touch with Mahan and

Sfc. David Chase, curator of

the Fort Benning Infantry

Museum.

They arrived in short

order and began investigation

of the site. By noon, they and their helpers had unearthed several sacks full of bits and pieces of pottery and several animal bones.
Chase said the site where the excavation was being made was probably a refuse pit, where the Indian families of the village dumped broken pots, dead animals and other trash.
Further out in the field being developed by Wickham, a road scraper was slowly bringing to light other information about the village.

Mahan and Chase pointed to round, dark spots in the red clay.
"Those are the spots where the posts of houses once stood," Mahan said.
The most important find as of mid-afternoon yesterday, was the barrel of a musket, which Chase said was unusually short compared to muskets of the same period, which he estimated to be between 1780 and 1800.

SFC DAVID CHASl:, JOE MAHAN, NEAL WICKHAM, VIEW GUN BARREL. One of artifacts dug up Saturday on site of Kart Racing track.

Figure 96. Facsimile of unattributed newspaper article describing discoveries at the kart racing track (Colwnbus Musewn files).

111

The Bull Creek Site
The "helpers" mentioned in the newspaper piece were University of Georgia archeologists Bettye Broyles and Frank Schnell, Jr. Figure 97 shows Broyles in the process of excavating one of the large pits. Two large pits, which contained material dating to the late eighteenth or early nineteenth centuries, were eventually excavated.
Figure 98 illustrates the distribution of postmolds in the nearby area mentioned in the newspaper article. The map was carefully prepared by Broyles using an alidade and plane table. While the map accurately depicts the positions of the postmolds, there are no reference points or landmarks which allow the exact location to be known or even the proximity of the postmolds to the large pits. The location can only be approximated as the area of XUA as shown in Figure 95.

The map of postmolds covers an area more than 25 m across. By connecting the dots on the map, it is possible to define a series of arcs which may represent some evidence of structures. However, there are no concentrated or well-defined postmold patterns comparable to the Bull Creek site structures. Also, there are no large Lamar trash pits in the area. Based upon a small collection of sherds found in the parking lot area at the track and now curated at the Columbus Museum, it appears probable that any structure represented by these posts predates both the Creek and Lamar occupations.

1960 photograph showing Bettye Broyles excavathig a large pit in tIiekart track construction 112

Minor Investigations

.10

.9
8
.11
.12

16 .15
.14
.17 18

.6 .13 .7
8

.41 .43

38

.44

.46 .45
.48

.39 .40

.37
36 35

47

54

.32 31

Datum 6.

.21

.1

.30 .26
.29

N

.24

~

.25

o12 3

Meters

Adapted from B. Broyles, 1959

Figure 98. Plan map of postmold pattern in the Go-Kart locale (adapted from field map prepared by Bettye Broyles, on file at the Columbus Museum).
113

The Bull Creek Site
A single large Creek refuse pit was excavated as XU-B in 1969 at the site of the Dolly Madison plant. Figure 99 shows two views of the excavation in progress. These photographs clearly depict the massive site destruction relating to the developing commercial landscape of the time.
This large Creek pit produced substantial amounts of aboriginal pottery and Euro-American trade material dating to the early part of the nineteenth century. The feature, which had been partially destroyed when discovered, produced nearly 2,400 sherds representing a minimum of 59 vessels. At least some of these vessels appear to imitate European vessel forms. The contents provide valuable comparative information for similar material found at Bull Creek. Figures 100 and 101 illustrate selected artifacts.
Figure 99. Two views of pit excavation at the Dotty MawsonplalJf (1969
photographs from files of the Columbus Museum).
114

Minor Investigations
"~;'~'~'it~ -';'i~ji;" L;'i.~"~<~~";;"i"d'~~~
3 em
em Figure 100. Selected aboriginal artifacts from the Creek pit in the Dolly Madison locale.
115

The Bull Creek Site
Figure 101. Selected Euro-American artifacts from the Creek pit in the Dolly Madison locale. 116

Minor Investigations

Commoilly, when a very large feature such as a large Creek trash pit is excavated through soils containing occupation remains from earlier times, some intrusive material becomes deposited within the feature. This was not the case with the Dolly Madison feature as there was no earlier pottery in the pit. This fact provides some evidence that this area lay beyond the boundaries of the Bull Creek phase Lamar village.
1960 Investigations by Tyler and Jackson

As with any well known site,

pothunting continued throughout the decades

that Bull Creek was being professionally

investigated. In one instance, information was

recorded by examining one of these potholes.

In January 1969, youthful Columbus

archeologists Jack Tyler and Terry Jackson

prepared a brief report on their

examination of a pothole at Bull

Creek entitled Report on

Clearance of Pothole A, Bull

Creek Site (Tyler and Jackson

1969). The pothole had been

located along Weracoba Creek in

an area which had not been

previously examined through

professional excavation. The

handwritten manuscript also

contains several photographs and

maps which provide some useful

information for locating the pit in

relationship to other features. Photographs of value show the general landscape of that portion of the site as it appeared in 1969,

,'~.:. .......................

.:::;;-

which was prior to intensive

landfilling.

more than one hundred sherds were recovered. One chert flake and one piece of fire-cracked rock were also found. The present location of the collection is unknown. The most important information presented in the brief report deals with location and stratigraphy.
The location of Pothole A (site A) is given as a lightly wooded bluff edge overlooking the junction of Weracoba Creek and Bull Creek. The face of the bluff was eroding severely at that time and the pothole was found on a slight projection only 32 inches from the edge of the cliff. The area below the cliff was thickly covered with brush (Figure 102). Much of the pothole had been refilled with backdirt (Tyler and Jackson 1969). According to the plan drawing, the pothole measured more than one meter across.

Golf

Driving

Range

...--,

-~ .:

-,

. /r. ~~ !;4:'f,J,/ ,

"
'''',, ,

/

"

,/

Trailer Sales

"
/

"I

// ~:.,.,.<.. : " .

The pothole report deals primarily with the methodology of examining the pothole. A few typical Lamar sherds are illustrated in drawings and one general layout photograph. There is no tabulation of artifacts although

N
t
Figure 102. Redrafted sketch map prepared by Tyler and Jackson showing landmarks in the vicinity of Pothole A.

117

The Bull Creek Site

Backdirt was removed from the pothole down to a layer of blue clay which was believed to be the bottom. The profile drawing is difficult to interpret, but a cultural zone appears to be present that extends to approximately one foot below surface. This overlies an apparently sterile orange clay, which in tum overlies the blue clay.
The report prepared by Tyler and Jackson provides a few additional bits of information helpful in interpreting the extent of the Bull Creek village. The existence of the pothole indicates that someone considered the area to be relatively rich in artifacts. The recovery of more than 100 sherds would seem to be confirmation of that fact. The report helps chronicle the nearly continuous erosion of the Bull Creek site during the present century. The report also provides some data for a portion of the site which has been lost to erosion and is now covered by a modem landfill.
1970s Developments
In the following years the Bull Creek site continued to be commercially developed, with only remnants preserved near the bluff edge. The bluff edges were preserved primarily because trees were needed to prevent further bank erosion. Like other sites which appeared to be destroyed, archeologists viewed the site more from a historical perspective than as a site worthy of continued research. Only Frank Schnell, Jr., archeologist for the Columbus Museum sustained a measure of public awareness.

In Muscogee County Georgia, the lake is normally bound by the previously existing river banks. The Bull Creek site seems to be totally unaffected by the flooding of the lake. A potential detrimental by-product of the lake development is the possibility of the construction of a marina on the site. At the moment, the site is owned by an individual conscious of the intrinsic and scientific value of archaeological sites (Schnell 1973).
1981 Excavations by Jack Fuller
In the 1980s, most of the Bull Creek site was graded for commercial development. Because the development was privately funded, there was no professional monitoring. One avocational archeologist, Jack Fuller, recovered a small amount of material and reported his findings to Frank Schnell, Jr., at the Columbus Museum. Mr. Fuller was contacted in 1995 and asked to provide further details to the information on file at the museum. A sketch map (Figure 103) and brief report were prepared. A portion of that report follows.
Since childhood I have known of, and had an interest in the Bull Creek site. In 1980, the John Deere Tractor dealership opened on the old Indian site .... In 1981, this area was further disturbed by people test driving tractors for the dealership.
The Bull Creek Site

Schnell produced one report in the 1970s which dealt with the Bull Creek site from a planning perspective. In his 1973 report A Preliminary Assessment of Archaeological Resources Remaining in the Walter F. George Lake Area, Schnell considered the effects of the lake on the Bull Creek site.

N

GEORGIA

i

Figure 103. Sketch map prepared by Jack Fuller showing the Bull Creek site in 1981.

118

Minor Investigations

One hot day in June of 1981, I noticed a box blade cut. In the cut lay the remains of two small clay vessels. One lay upside down, the bottom sheared away by the box blade, leaving only a ring of rim sherds in the soil. The vessel diameter was about 3.5 inches, with a white slip, large grit tempering and decorated with two incised lines, crudely executed, forming a Lamar Bold Incised design. The other was upright, but at an angle which allowed 1/3 of the rim to be sheared off. This vessel was of comparable size and temper. A poorly executed Lamar stamp design had been applied. This second vessel was recovered {Figure 104}.

On subsequent visits, the destruction of the site quickened. Only one other feature of note was exposed. That being a trash pit, which was also exposed by the tractor equipment. It was shallow with a diameter of about 3 ft. It contained charcoal, ash and pottery. Lamar stamped and incised wares were dominant. The remnants of a large incised bowl with an effigy rim adorno was recovered from the pit (Figure 105).

o

3

H

em

Figure 104. Small complicated stamped vessel from the Bull Creek site.

..:"~., .v.' .':," .

o

3

"

Figure 105. Artist restoration of effigy vessel from trash pit excavated in 1981 (photograph from files of Columbus Museum). Figure also shows enlarged views of head (2 views) and tail.

119

The Bull Creek Site
What the effigy represented is subject to interpretation; I suggest a dog. There was not enough present to restore but the diameter was estimated at 14 inches. The vessel had a white grit tempered slip.
If there is anything left of the site, I think it will be found on the finger of land that extends out south along the river, under the mantle of clay that was deposited for the railroad bed that used to extend across Bull Creek (Fuller 1995).

Only part of collection made in 1981 survives. The collection consists primarily of rims and decorated body sherds (Figure 106 and 107). This information has been incorporated into the larger collection of pottery from Bull Creek and will be presented in Chapter 8. The collection also contains a few pieces of chipped stone, four pottery discs and two pipe fragments (Figure 108).

B A

o

3

em

o

c
E
Figure 106. Examples of Bull Creek phase sherds from a pit excavated by Jack Fuller in 1981.
120

Minor Investigations B

o

3

P""-.

;

em

G

H Figure 107. Examples of incised and complicated stamped sherds from trash pit excavated in 1981.
121

The Bull Creek Site
Figure 108. Examples of Bull Creek phase pottery objects and Archaic bifaces found at Bull Creek in 1981. 122

Minor Investigations

The material in the

Table 1. Characteristics of Vessel from Fuller's collection

collection lends itself to examination

with respect to vessel form. Many

of the sherds are large and vessel profiles and motifs are reconstructible. Fuller's collection

Vessel Type

Plain

Stamped Incised/ Punctate

includes a minimum of 28 vessels.

Most were recovered from a single

Simple Bowl

3

2

feature.

Table 1 lists these vessels

Carinated Bowl

2

o

15

with respect to vessel type and

Jar

o

5

o

surface treatment. The results are

intriguing primarily because of the

high proportion of incised/punctate

vessels. The high proportion of incised vessels and the presence of certain incised motifs suggest this pit was later than most of the Bull Creek site occupations previously examined. The motif shown in Figure 105b, for instance, is a common Stewart phase motif.

The leveling, filling, paving and associated commercial development of the 1980s destroyed most of the Bull Creek site. Except for a few spots at the bluff's edge where trees were left to retard erosion, little evidence of the site remained. It was not until the mid-1990s that the site was again

examined. The results of that work follow.

123

TJDS PAGE LEn BLANK

Chapter 7 1993-1994 Investigations

The most recent archeological investigations at Bull Creek and the contiguous GoKart site were conducted by Southeastern Archeological Services, Inc, (SAS) and will be summarized in this chapter. The primary goals of the 1993 survey were site assessment and determination of site boundaries to aid in the design of a riverwalk park along the Chattahoochee River.
The SAS investigations included survey, testing and excavation, with most of the excavation occurring at the Go-Kart site. By accomplishing these tasks, the route of the proposed Columbus Riverwalk could be planned so as to minimize disturbance to the sites.
As a result of the survey, relatively small but highly significant preserved deposits were found to remain on both sites. Within the defined boundaries of the Bull Creek site, a route was selected that would not impact any preserved deposits. An alternative route was not possible for the Go-Kart site, and data recovery excavations were required. Because the results of data recovery at the Go-Kart site are detailed in a separate report (Ledbetter 1995c), only those results which directly contribute to an understanding of the primary Bull Creek site components are included in this chapter.
By 1993 Bull Creek was more of a memory than an archeological site. Virtually everything known about the site was to be found in Patterson's brief report in Early Georgia (Patterson 1950) or secured in the file drawers of the Columbus Museum. The only remaining person knowledgeable of the site was the Museum's archeologist, Frank

Schnell, Jr. Schnell had maintained the records of past excavations and fortunately retained much of the "oral history" of the site.
Figure 109 shows the route of the riverwalk and the boundary line used to separate the Bull Creek site from the contiguous Go-Kart site. The Columbus Riverwalk survey was actually the first intensive archeological survey of the Bull Creek site. There was no accurate information available for determining the former site limits or for determining the boundary between the Bull Creek site, 9Me1, and the adjacent GoKart site, 9Me50. Multiple site forms existed for the two sites, but depictions of site boundaries were contradictory. Unfortunately, by 1993 very little of the once extensive Bull Creek site was left to be surveyed. A series of photographs (Figures 110-112) illustrate the condition of the site area at the time of the survey.
Much of the area between the South Commons (Columbus Fair Grounds) and Bull Creek has been extensively modified by commercial development. Modification of the old land surface has resulted from grading for leveling and elsewhere extensive landfilling of low-lying areas. Figure 110 shows a portion of the Bull Creek site which has been both graded and filled to create a level surface following severe erosion and gullying. Figures 111 and 112 show the area adjacent to Bull Creek where renewed landfilling (to a depth of approximately 50 ft) was continuing as our survey was in progress. According to one local informant, this landfilling has been ongoing since the 1960s with only limited beneficial results.

125

The Bull Creek Site

N
i

0

200 Feet

e

0

66m

Figure 109. Route of riverwalk through the Bull Creek and Go-Kart sites. 126

1993-1994 Investigations
Figure 110. Photograph of the Bull Creek site as it appeared in 1993, view to the east.
Figure 111. Photograph of landfilling at the Bull Creek site in 1993, view to the southeast. 127

The Bull Creek Site

Figure 112. Photograph taken in 1993 of the approximately 50 ft thick landfill at the edge of Bull Creek.

As previously noted, the riverwalk corridor through the Go-Kart site could not be designed to avoid preserved cultural deposits along the upper edge of the wooded bluff line. Shovel testing and test pit excavation in that area produced evidence of intact cultural deposits consisting of both midden and features. The primary occupations indicated by artifacts from these deposits were Archaic and Early Mississippian (Averett phase). Minor aboriginal components included late Mississippian, Bull Creek phase, and historic Creek. There were also artifacts that dated from the midnineteenth through twentieth centuries (Ledbetter 1994b:69). Data recovery was recommended and conducted by SAS in 1994 (Ledbetter 1995c).
The survey of the Bull Creek site was substantially more involved due to the large area to be examined and the extensive filling.

The survey methodology and results will be briefly summarized from the survey report (Ledbetter 1994b). Figure 113 depicts the site at the time of the survey and shows the locations of all collection and testing units.
A visual assessment of the site in 1993 was far from encouraging. The site area showed the scars of years of neglect and abuse. The impact of the 1920s construction of a railroad bridge across the mouth of Bull Creek were clearly seen. Large borrow pits shown in mid-twentieth century photographs and maps were probably excavated to provide fill dirt for construction of the.railroad levee and trestle. In the mid-century the area was used for recreational racing and several associated structures were constructed. Modern commercial development has affected the site through grading and filling (Ledbetter 1994b:70).

128

1993-1994 Investigations

Graded Field

Collection Area A
\

Riverwalk Route

N
A

o

SOm

Positive Shovel Test o Negative Shovel Test

Figure 113. Plan map the Bull Creek site in 1993 showing locations of SAS shovel tests and surface collections.
129

The Bull Creek Site
A primary task of the present survey was to determine just how much remained of the site since the last investigation nearly 35 years ago. The field archeologists were initially guided through the site area by Columbus Museum's archeologist Frank Schnell. Mr. Schnell provided substantial assistance for interpreting the site by pointing out the general locations of old excavations. Most of the site had been graded and filled since the last excavations but the approximate location of Schnell's 1959 block was ascertained. The primary reference point remaining on the site is the railroad grade and a large depression left from the 1936 excavation of the cemetery.
The Bull Creek site was examined through surface inspection and shovel testing (see Figure 113). A few small pockets of exposed ground were found on the site. These exposed areas were collected separately and designated as separate collection units. Shovel tests were excavated as systematically as possible. The goals of shovel testing were two-fold. First, shovel tests were required to determine the depth of cultural deposits, when present. Second, shovel tests were use to assess the severity of grading and modern filling ...

In terms of surface

Figure 114. Photograph of 1993 shovel testing (Shovel Test 4) .

exposure, there was generally no

correlation with subsurface re-

mains. Across most of the site, artifacts found on

The photograph shown in (Figure 114)

the surface represent either displaced material from

was taken at the location of Shovel Test 4. Shovel

land clearing or residual material remaining after

Test 4 represents the northernmost extent of the

sheet erosion has removed all soil. The conclusion

preserved site area. Because the slumping gully

resulting from the excavation of shovel tests is that

walls to the south of Shovel Test 4 produced

only one portion of the site remained preserved.

ample evidence of an intact site, further shovel

That area lay beneath the old railroad fill. Unfortu-

tests were not excavated in that direction. The

nately, that portion of the site was being actively

large slump shown in the (Figure 115) photograph

eroded away at the time of survey.

begins approximately 5 m south of Shovel Test 4.

130

1993-1994 Investigations
Figure 115. 1993 photograph showing the large erosional feature impacting the Bull Creek site (the midden zone is being pointed out by the archeologist).
131

The Bull Creek Site
The field crew did take advantage of the site erosion to further investigate the site. This investigation included cleaning a section of the gully wall to examine site stratigraphy in the remaining (preserved) portion of the site (Figure 116). This was accomplished by utilizing the area of most recent slumping which was nearly vertical and extended across all soil zones associated with site (Figure 117). Because of the freshness of this slump, little cleaning was necessary to prepare the profile for mapping and photography.

The slumping midden soil at the bottom of the gully was also examined. Several large sherds were found protruding from the boulder-size chunks of soil that had fallen into the gully but much of soil and associated artifacts had obviously
been washed away (Ledbetter 1994:79). Table 2 lists artifacts found by the survey crew in the midden slump at the bottom of the gully. The artifacts were recovered by troweling through large boulder-sized chunks of midden.

1993 photograph of gully wall following profile cleaning. 132

1993-1994 Investigations

0-

)f/

lOOcm

200cm

I

I

,'/ Surface

~1/

300cm I

400cm I

Red Brown Mottled Clay

------ ------------- -IOOcm-

~------------------Yellow Brown Sandy Clay

-200cm-

Old Humus Light Brown Sand

Dark Brown Sandy Loam (Midden)

- - - - - - - - - - - - -------

Medium Brown s a n d = = - - - - - - - - - -

Figure 117. Profile drawing of a portion of the gully wall examined by SAS in 1993.

Table 2. Tabulation of artifacts recovered from gully midden slump from 9Mel.

I 1> . Prbveniellce .

.... Description .

1

<Contents

I

Surface Collection Area A

Midden slump in large gully at southwest corner of site. Material
recovered by troweling large blocks of midden at base of gully.

1 chert Archaic stemmed PP/K 1 chert utilized flake
2 quartz tertiary flakes < 2 em
1 fragment of ground hematite 1 fire-cracked rock
78 Lamar complicated-stamped sherds 1 Lamar vertically pinched rim sherd on
comp. stamped body 2 Lamar noded rims 1 Lamar folded pinched rim sherd 2 Lamar notched rims, plain body 1 Lamar notched and incised rim sherd 4 simple rims on plain body sherds 1 simple rim on comp. stamped sherd 21 plain grit-tempered body sherds 13 plain sand-tempered body sherds 5 burnished sand-tempered sherds 2 Mercier Check Stamped sherds 1 Fort Walton Zone Punctate rim sherd 1 engraved rim sherd
1 incised rim lug Artifact Total: 140

133

The Bull Creek Site

The profile recorded from the gully wall is comparable to the stratigraphy previously recorded for this portion of the site (see Figure 83 in Chapter 5). In the section of the SAS profile drawing shown in Figure 117, various layers of railroad fill greater than one meter in depth cover an old humus zone. The humus zone is compressed by the fill but is conspicuous because of its coloration and high loam content. The humus zone covers a light brown sand which in turn covers a darker brown sandy loam midden zone. Cleaning of the midden zone produced a few Bull Creek phase sherds but no lithics. The midden zone contained flecks of charcoal but no Mississippian features. Of interest, the Archaic projectile point was recovered from the railroad fill of the profile (yellowbrown sandy clay). This indicates the projectile point was brought in with the fill, perhaps from another site area such as 9Me50.

The features (pits) shown in

Figure 117 are related to the railroad.

These features represent holes cut

through the old humus zone that were

filled with the reddish brown clay. The feature fill is homogeneous with

the lowest zone of railroad fill.

Figure 118 shows a photograph of the Figure 118. Photograph of section of profile shown in previous

southernmost feature.

profile drawing (note pit extending from uppermost railroad fiII

into midden) .

Because previous excavations

of the Bull Creek site have already produced

features are present at the periphery of the

significant information, the Bull Creek site is

cemetery that would provide information for

recommended eligible to the National Register

better interpreting the cemetery area. This

of Historic Places. Portions of the site remain preserved that may contribute more information toward an understanding of the region's

area of the site should be stabilized and protected.

prehistory. The preserved portion of the site

bounds the cemetery excavations of the WPA

era. We do not know at present if the entire

cemetery was actually excavated or if related

134

The SAS survey determined that the remainder of the site has been destroyed. The SAS archeologists were instructed to find a route for the riverwalk and a location for the bridge across Bull Creek that would provide no impact to the site. This can be accomplished by routing the riverwalk and bridge across the creek near the location presently occupied by a pipeline. This area lies near the mouth of Weracoba Creek. If this is accomplished, further archeological field work will not be required (Ledbetter 1994:70-84).
Site boundaries for Bull Creek were reconstructed following the 1993 survey

1993-1994 Investigations
(Figure 119). These boundaries, which were prepared in large measure to meet the needs of cultural resource management, represented a combination of former estimates of site boundaries as shown on site forms and the results of the 1993 shovel testing. Essentially the boundary between site 9Me1 and 9Me50 was placed at a large erosional gully. However, subsequent excavations conducted for data recovery of the Go-Kart site, 9Me50, produced evidence that this boundary was not real. These excavations produced evidence that the Bull Creek village actually extended a substantial distance onto the Go-Kart site. The results of data recovery will be briefly summarized.

Figure 119. Tentative approximation of the boundaries of 9Mel and 9Me50 based upon survey results (photograph source, 1950 SCS).
135

The Bull Creek Site
Data recovery excavation in the fall of 1994 was restricted to a narrow, 7 m wide, corridor through the Go-Kart site. Along this corridor, the site was examined by shovel

testing, text pit excavation, backhoe trenching, and machine-assisted block excavation. Figure 120 shows the larger excavation units of data recovery.

I

I

I

I

I

480E

S20E

560E

600E

640E

- 540N -SOON ,.,/ -460N

GravelParking Lot ExistingBuilding

o

30

m

Test Pit c::::l Backhoe Trench
o Isolated StandingTree
---- River Walk Center Line ---Site Boundary

- 400N

'". / .~
.

-360N

- 320N

- 280N

Park.man Cul-de-sac

Figure 120. Plan map of data recovery area of the Go-Kart site showing trenches and block excavations 136

Excavation in the northern portion of the Go-Kart site yielded material and features associated primarily with an early Mississippian Averett phase occupation. The discovered features were postmolds and small pits. There was also a thin scatter of Archaic lithics. Perhaps the most significant contribution of those excavations was the procurement of several Averett phase radiocarbon dates (Table 3). These assays confirm previous age estimates (Schnell and Wright 1993; Ledbetter 1995a) and support a range for the Averett phase between the late tenth and late thirteenth centuries.

1993-1994 Investigations
A few Bull Creek phase sherds were found in the northern part of the Go-Kart site but there was not evidence of intensive occupation. Quite possibly, the northern portion of the data recovery area represents the fringe of the Bull Creek village occupation. Whether this assessment is true or not, the available information from data recovery does indicate a scatter of Bull Creek phase pottery extending a minimum of 600 m north from the mouth of Bull Creek. The area beyond that point has been totally destroyed.

Table 3. Averett phase radiocarbon dates from the Go-Kart site.
.

Beta 78127 Beta 78128 Beta 78129

1020 70 B.P. 820 80 B.P.
920 50

A.D. 1015 A.D. 1235 A.D. 1065, 1075, 1155

-- ...
........
,,
.< .....
A.D. 980-1040
A.D. 1170-1280 A.D. 1035-1195

Excavation in the southern part of GoKart produced cultural remains associated primarily with the Bull Creek phase and historic Creek occupations, along with a continuing scatter of Archaic lithics. Data from these excavations indicate that structures associated with the Bull Creek village extended a minimum of one hundred meters beyond Lester's northern most excavations. Large excavations designated Block B and Block D produced important information relating to the Bull Creek site.
Block B (Figure 121) was excavated at the southernmost edge of the Go-Kart site. The block actually intruded upon the edge of a

large gully which had tentatively been used to define the boundaries between the two site areas. Machine stripping indicated that the area had been filled within relatively recent times but beneath that fill the ground surface was badly eroded. The only cultural features to survive this erosion were a large historic Creek pit and an accompanying square postmold. The large pit was 160 em in diameter and 110 ern deep. Figure 122 shows views of the pit during excavation. Contents of the feature included partially reconstructible vessels, various Euro-American manufactured items (Figure 123), and ,a relatively large amount of Bull Creek phase pottery.
137

The Bull Creek Site

I 595E

I 600E

I 605E

0 0

0

a

0 0 ()

0

0

00

00

Shallow

0

Depression

a

0
Modem Fill Shallow Depression
0

9ME50 Block B
N
~
o
e m
[23 Filled Gully
o Tree

Figure 121. Plan map of block B at the Go-Kart site. 138

335N 330N -

1993-1994 Investigations
Figure 122. site.
139

The Bull Creek Site
9
--- 1 . em
Figure 123. Examples of vessels (large Chattahoochee Brushed jars and undecorated water bottle, greatly reduced in size) and glass and metal beads (greatly enlarged) from the Creek pit. 140

The large Creek pit also produced more than two hundred Bull Creek phase sherds in the feature fill. These sherds were interpreted as material displaced from the surrounding midden during feature excavation and subsequent use. The number of sherds found within the Creek feature indicated a dense midden deposit was once present.
Block D, located to the north of Block B, was particularly important with respect to the Bull Creek village. Block D was located at a point where the riverwalk corridor made a short jog toward the river. Excavation of the block exposed a nearly complete post pattern representing a Bull Creek phase structure (Figures 124 and 125). Excavation produced a substantial number of postmolds and a few small pits associated with the structure. The arrangement of the postmolds suggests a subrectangular structure approximately 7 by 7 m. The perimeter postmolds were relatively

1993-1994 Investigations
large, generally ranging from 15 to 25 em in diameter and as much as 25 em deep. Postmolds which form interior partitions were of comparable size. Large interior support posts were as much as 30 em in diameter and 35 em deep. In most respects, the Block D structure appeared similar to Lester's Structure 1.
The SAS investigations confirmed that most of the Bull Creek and Go-Kart sites have been destroyed. A small strip of both sites survives along portions of the wooded river bluff but these areas are subject to riverbank erosion and may eventually perish. Portions of the cemetery area also remain preserved but that area appears to be even more at risk from erosion. With the exception of a few extremely large and deep features such as that found in Block B, there appears to be little likelihood of discovering intact remains in the remainder of the former site area of the Bull Creek site.

Figure 124. Photograph ofBull Creek phase structure in Block D of the Go-Kart site, view to the west.
141

The Bull Creek Site

I
564E

I 566E

I 568E

I 570E

I 572E

I 574E

I 576E

I 578E

9ME50

Block D

368N-

N
!
o

o

0

2

E

0

m

a

o 0

0

o

00

o

o 00

o

0

o0 0

o

o
o

o0

o

o

o 0 o 00

0

o

0

0 o

00

0

366N 364N362N360N -

o0
o
o

o (J
o
o
o

358N356N354N-

352N-

Figure 125. Plan map of Block D at the Go-Kart site.

There is a slightly greater potential for feature preservation in the area of the Go-Kart site. The data recovery investigations suggest that a series of structures associated with the Bull Creek village extended along the river for

a minimum of 450 m from the mouth of Bull Creek. The results of the SAS investigations with respect to the extent and layout of the Bull Creek village will be pursued further in Chapter 9.

142

Chapter 8 Bull Creek Site Material Culture

Perhaps the weakest aspect of most previous investigations at the Bull Creek site has been the paucity of artifact analyses and synthesis of material remains within a cultural framework of the late Mississippian period, and particularly the Bull Creek phase. In this chapter we will present tabulations of artifacts and interpretations found in the manuscripts and notes of previous investigations. Information pertaining to the material culture of the Bull Creek site will accordingly include the unpublished compilations of data from sundry investigations and new forms of analyses utilizing the surviving collections. An examination of the Bull Creek phase ceramic assemblage will be presented first. Following that, the relatively small amount of information available for other categories of Bull Creek phase material remains will be examined. The latter will be brief because lithic materials, in particular, accounted for a minor part of the Bull Creek phase material culture. Unlike earlier people, it appears that chipped stone projectile points were not being manufactured. The material remains from other components will be noted but not detailed. The examination of other components will be especially concise because that material is covered in detail in a separate report dealing with data recovery of contiguous site 9Me50 (Ledbetter 1995c).

complicated stamped pottery. The site also lay at the northern edge of a Gulf Coast tradition characterized by pottery decorated with Fort Walton zone punctate designs. Figure 126 illustrates the geographic location of the Bull Creek site in relation to the areas of these two ceramic traditions. It should be noted that the areas shown are the approximations of two researchers and subject to variable interpretations.
o

Bull Creek Phase Ceramics
During the late Mississippian period, the Bull Creek site was positioned at the boundary of two groups who developed distinct ceramic traditions. The site lay at the western boundary of a region with a ceramic tradition characterized by Lamar

N
A

o

ISO

kin

Figure 126. Location of Bull Creek showing boundaries of Lamar and Fort Walton ceramic traditions (based upon maps by Williams and Shapiro {1990:5} and Scary {1985:200}).

143

The Bull Creek Site

The pottery from the Bull Creek site is typical of late Mississippian decorative styles and vessel forms found throughout the southeast. However, the combination of ceramic characteristics based upon two ceramic traditions raises the probability of greater diversity of the Bull Creek site assemblage compared to sites located within the heart lands of either tradition. Based upon analysis of ceramics from other sites this would appear to be the case, but as Schnell has noted:

The site has over the years become a favorite example of a certain variant of Lamar. Despite this repeated use as an example, nothing has been published on the site in the way of a comprehensive study of the site and its implication
in southeastern archeology (Schnell 1963).

After three decades, the essence of Schnell's commentary remains. While results of analysis of ceramics from sites such as Rood's Landing (Caldwell 1955), 9Cy51 (Broyles 1962), and Park Mound (Hally and Oertel 1977), have been interpreted as typical of the Bull Creek phase, the actual comparability of these collections to the Bull Creek type site has remained unknown. In recent years, some steps have been taken to better define the characteristics of the Bull Creek phase. Particularly important are efforts to quantify ceramic assemblages (Knight 1994; Schnell 1986, 1990; Schnell and Wright 1993; Scarry 1985) and define the geographical distributions of sites relating to this time period (Hally 1994; Schnell and Wright 1993). Hally's (1994) approximation of the boundaries of the phase are shown in Figure 127.
The work of these researchers has refined our understanding of ceramic types and ceramic relationships. Much of the following presentation of ceramic data will simply help to quantify and hopefully better describe specific attributes of the Bull Creek site pottery assemblage. Before examining the site ceramics in detail, additional background information will be presented.

N
A

o

150

km

Figure 127. Geographical extent of the Bull Creek phase (adapted from Hally 1994:151).
Part of this background material includes the interpretations of earlier researchers, thus providing historical context for the data presented in this chapter. The background material is also meant to impart some of the diverging points of view which have existed, and still exist, concerning this phase. This divergence eventually culminated in splitting the traditionally defined Bull Creek focus into early and late phases (Schnell 1986, 1990).
The blending of ceramic traits from two areas created a basic set of attributes which once distinguished the Bull Creek focus, and now defines the Bull Creek phase pottery assemblage. The basic characteristics of this blending have been described by several researchers but perhaps most concisely as follows:

144

Bull Creek Site Material Culture

Bull Creek phase ceramics can be considered a blend of generic Lamar and generic Fort Walton of this period [late Mississippian]. The zone punctated type Fort Walton Incised becomes the principal Gulf Tradition decorated type, while Lamar Complicated Stamped becomes common as the principal "South Appalachian Mississippian"
type added to the local assemblage (Knight and Mistovich 1984:224).
The observations of past researchers provide a historical perspective relating to the recognition of the rather unique characteristics of the Bull Creek phase ceramic assemblage. In 1951 Kelly noted:
. . . while the rim treatment, paste, and general morphology of the bulk of the stuff at Bull Creek is definitely in the time level and "feel" of Lamar, the specific designs have a Savannah stamped tradition; also there is a lot of pottery that must refer to influences coming up the Chattahoochee from Florida, Fort Walton influences. This means a very interesting variant in Bull Creek of the widespread Lamar manifestation
(Ledbetter 1995b:49).

similarities to Ft. Walton of northwest Florida, to the Lamar Culture of Central Georgia, and to a lesser extent Safety Harbor of the Florida east Coast. There are even some specific ceramic similarities to Moundville, Alabama, and the Dallas Culture which succeeded Hiwassee Island in eastern Tennessee. The type of notched rim strip, for instance, is found both at Moundville and in Dallas associated with effigy rim adornos.
The Later Period Culture at Rood's Landing is neither Ft. Walton nor is it Lamar, but represents rather a borderland culture between the two, apparently more closely oriented toward Ft. Walton. The variant of Lamar Complicated Stamped found at Rood's is particularly interesting in appearing to be directly derived from the older Savannah Complicated Stamped type (Figure 128) . Lamar Bold Incised, as the type is known in central Georgia, is practically absent, and its place is taken by Rood's Incised which more closely resembles Floridian variants of a Ft. Walton time level (Point Washington Incised and Pensacola Incised) ...
(Caldwell 1955:45).

A few years later Joseph Caldwell made very similar observations concerning a related ceramic assemblage at Rood's Landing. According to Schnell (1986) the material described by Caldwell as the Later Period Culture should date to the latter portion of the traditionally defined Bull Creek phase. While the Rood's occupation may date a few decades later than the primary occupation at Bull Creek, Caldwell's observations are still important for interpreting the general time period.
The Later Period Culture, which we know in some detail from our work on the summit of Mound A, shows ceramic

.8

C

r

,-c/ G. 7

o

.,

i

i

JOESIG/f 1I10TII'"S Or LA,4UR CON,P,L/C'ATE.o
AT /?oo.oir LA#~/#G'

Figure 128. Lamar Complicated Stamped Design motifs from Rood's Landing (redrawn from Caldwell 1955: Figure 7).

145

The Bull Creek Site
The first presentation of Bull Creek pottery to the archeological community became the task of Isabel Patterson even as the WPA excavations progressed (Patterson 1936). Patterson's paper highlighted the more spectacular burial finds, but results of preliminary analysis of more than two thousand sherds were summarized.
A preliminary analysis of 2,240 sherds from the Bull Creek site show a surprising resemblance with the 5,000 sherds studied and tabulated from the Lamar Village site on the Ocmulgee. There are no major differences in the field of design or stylistic treatment. In the Bull Creek collection there are a few sherds, a fraction of one percent which are probably related to the Lower Mississippi types exhibited by James A. Ford in his work in this area. A few similar incised and punctate types found at Lamar may be ascribed to the same influence. These two sites, so far removed on different river systems, show such a striking similarity at a time [far removed] considered to be either early protohistoric, just antedating de Soto, or late prehistoric. Forty-nine percent of the Bull Creek sherds are of stamped ware, forty percent plain, two percent incised and one percent incised and punctate. All other designs are less than one percent and ninety
percent of the pottery is grit-tempered (Patterson 1936).
The first attempt at a detailed description of Bull Creek pottery appeared in the mid-1940s (Fenenga and Fenenga 1945). The Fenenga survey, which focused on the lower Bull Creek drainage, resulted in relatively large collections of pottery. Because the Fenengas were not familiar with the ceramic typologies of the region, local type names (ware groups) were used and detailed descriptions were presented. The Fenenga types consisted of Weracoba Grit Tempered, Moon Dairy Sand Tempered, Columbus Fine Sand Tempered, and fiber tempered wares (Fenenga and Fenenga 1945:4-8). These wares generally equate to Schnell's (1963) Bull Creek variant of Lamar (Weracoba Grit Tempered), several types of Woodland and early Mississippian pottery (Moon Dairy Sand Tempered), several types of late Mississippian and historic
146

Indian pottery (Columbus Fine Sand Tempered), and Late Archaic (fiber tempered). The Fenenga's description of Weracoba Grit Tempered provides an excellent characterization for many attributes of the dominant category of coarse grit-tempered pottery found at the Bull Creek site.
The pottery designated as Weracoba Grit Tempered in this report is a thick coarse ware varying but distinctive in color. The temper is of quartzite grit particles ranging in size from 1/2 to 2 mm. The paste used is very fine and smooth. Color is either a greyish white, salmon pink or buff with an occasional dark to black sherd. The surface is characteristically smoothed giving the effect of a fine hard surface repeatedly broken by grit particles. Construction was by the annular method, and lines of fracture are frequently between the smoothed-together coils. Sherds average about 9 mm in thickness. The ware is almost invariably decorated and complicated paddle stamping is the most commonly used motif. The stamping is not clear but appears to be a combination of curvilinear and straight lines. The stamp was used over the entire surface of the vessel. Incising or incised lines in combination with punctate patterns are also present. These designs usually are placed close to the rim. They do not appear in conjunction with paddle stamping but are sometimes supplemented with pinched-up knobs. Another form of decoration restricted to the neck or rim area is a pinched-up or appliqued collar which is decorated with nodes, rosettes and punctations. This collar seems to be an ela-
boration growing out of the folded rim (Fenenga and Fenenga 1945:8).
Following extensive test excavation at Bull Creek in the mid-1950s, David Chase produced the following description of the pottery from the Bull Creek site.
The predominant pottery type found during the recent tests is, without doubt, Lamar Complicated Stamped. This is a thick, gray to black ware, heavily grit tempered with particles of the tempering medium extruding to the surface. In almost every rim of this type, the noded collar is present. Nodes are either applique or else pinched. A minority of sherds lack this collar. Stamping is poorly controlled and random overstamping with poorly cut blocks or paddles seems to be the rule.

The vessel is globular and the stamping is applied over all. The rim is usually everted and the lip squared or rounded. No handles appear with this type. Lamar Plain is either a variant of the Lamar Complicated Stamped but without the stamping, or else appears in the form of small bowls. The lips are occasionally notched. The temper is sometimes grit and in a few instances sand is used. The grit tempered pots do not always show the grit on the surface. The noded collar sometimes appears on this ware and when it does the rim is everted. Other rims indicate straight sided or inverted bowl rims. A third type is Lamar Bold Incised. Rather deep incised, usually curvate lines, appear in the region of the rim. Sometimes these consist of only two or three parallel lines circumscribing the vessel. Interlocking loops also appear on some pots. In the cassuela bowl, which is making its first appearance in Middle Lamar times, the incising is applied to the inner ledge lip. Notched rims often appear with the Lamar Bold Incised vessels. Rim adornos, usually lugs, nodes and more rarely effigies, usually of birds occur on this type of ware. Included in the Lamar Bold Incised is a variant of the Fort Walton Punctate. In actuality, most sherds decorated in this manner cannot be distinguished from the classical Fort Walton punctate type. Another Fort Walton type, acquired by trade perhaps, is the Lake Jackson Plain. This occurs as a minority ware at Bull Creek. Shell tempering is also a minority manifestation. These last traits suggest initial contact with the Fort Walton culture to the south in Middle Lamar times, a relationship which appeared to have gathered strength in later times in view of the predominance of Fort Walton ceramic types on such sites at Engineer's Landing, Kendrick's and Bickerstaff, all post-dating Bull Creek. This may well be a clue as to the origin of the Lower Creek and Appalachicola alliance in protohistoric times.
A very few sherds found were of the burnished black slip variety. This is another ware which became common in later times. These were often incised, never stamped and never with handles. Adornos in this ware are often seen usually as animal or bird effigies. An acquaintance with the Dallas Focus is suspected here. Pipes are infrequently found, at Bull Creek only three stems and one bowl fragment came to light, hundreds of pottery disks and one of stone appeared. This is not surprising since the disks prevailed from earliest Lamar times into historic period ....
Bull Creek was termed 'Middle' Lamar for two main reasons. First, it did not produce the massive, almost undecorated thick sand tempered ware which characterized the Lamar site at Pat-

Bull Creek Site Material Culture
terson. Secondly, it lacked the abundance of fine burnished black ware, shell tempered buff ware and red and yellow slip pottery so characteristic of ceramics of the proto and early historic periods.
We suspect that Bull Creek represents an early habitation site of the Middle Lamar Period due to the presence of a small amount of typical Early Lamar pottery included in the lower levels of refuse. At the time of the occupation of this site, which was quite a large town, the Lamar culture had achieved domination of the middle Chattahoochee Valley and were thus able to move from the large inland sites of Patterson and Singer to the River way which they finally controlled. No mounds are associated with Bull Creek and it is suspected that although the tradition of mound building had not perished by Bull Creek times, it was definitely on the way out. The Neisler Mounds on the Flint and the Rood's Landing Mounds represent some of the last of the great temple mounds of the southeast. By Bull Creek times the Southern Cult has probably all but died out. Whether the dog pots found by Mrs. Patterson represent a relationship to the Cult is a matter of pure speculation. The general opinion is that
they were not (Chase 1957).
A detailed analysis of ceramics recovered from the Bull Creek site during the 1950 and 1959 excavations was initiated by Schnell in the early 1960s but only partially reported in his Bull Creek manuscript (Schnell 1963). Nevertheless, the analysis completed by Schnell is significant and constitutes a substantial part of the site's ceramic data base.
In terms of actual published data, the most detailed information relating to Bull Creek phase ceramics comes from sites 9Cy51 (Broyles 1962) and Park Mound, 9Tp41 (Hally . and Oertel 1977). Both site reports contain thoroughly described, tabulated and illustrated accounts of ceramic collections which generally conform to Bull Creek. Observations concerning ceramics at Park Mound are especially important because of temporal implications. Based upon several attributes, rim fold width in particular, Hally concluded that the Park Mound Bull Creek phase ceramics might date as late as the late sixteenth and early seventeenth centuries (Hally and Oertel 1977:51).

147

The Bull Creek Site
Schnell now believes the ceramic assemblages from 9Cy51 and Park Mound date to the latter part of the traditional Bull Creek phase (Schnell 1986, 1990). According to Schnell's view, the Bull Creek site (9Me1) and the contemporary occupation at Kolomoki (9Er1), are the primary examples of the earlier Bull Creek phase, which should date predominantly to the fifteenth century, while 9Cy51 and Park Mound, and the contemporary occupation at Rood's Landing (9Sw1) are representative of a predominantly sixteenth century Stewart phase (Schnell 1990:67; Schnell and Wright 1993:21).

Schnell's two phases are distinguished by differences in relative proportions of complicated stamped compared to plain, incised and punctate pottery. Bull Creek phase assemblages contain complicated stamped pottery in excess of fifty percent and low counts of incised/punctate wares. Stewart phase collections contain more plain wares (>50%), with complicated stamping dropping to about twenty percent and incised/punctate counts increasing to approximately fifteen percent. Minor amounts of check stamping occur with both phases (Schnell 1990:67-68). Figure 129 shows a graph originally prepared by Schnell (1985) which portrays these differences.

VARIATION IN LAMAR POTIERY ASSEMBLAGES IN CHATTAHOOCHEE VALLEY SITES

70% 60% 50% 40% 30% 20% 10% 0% ....1--11-
9MEI
Incised

9ERI

9CY51 SITES

9SWI

~ Complicated Stamped

9TP41
B~ PIam.

Figure 129. Graph showing differences in pottery collections from Bull Creek phase (9Me1, 9Er1) and Stewart phase sites (9Cy51, 9Sw1, 9Tp41), adapted from Schnell (1985).
148

Bull Creek Site Material Culture

Stewart phase collections are further distinguished by the addition of certain incised motifs similar to Pinellas Incised and the incised pottery of the Avery and Atasi phases to the north and northwest (Knight and Mistovich 1984:224). This would include the type Rood's incised (Caldwell 1955:27). Figure 130 illustrates an example from 9Cy51 of perhaps the most common Pinellas-like motif attributed to occupations of this latter phase (Hally 1994:154). Only one vessel with this motif has been identified at the Bull Creek site. That vessel was recovered from a pit excavated in 1981 and appears to be a rare example of a Stewart phase feature located on the Bull Creek site.

Figure 130. Drawing of typical Stewart phase incised motif (redrawn from Broyles 1962: plate 7).
Primary Pottery Attributes

In the mid-1980s, John Scarry devised a classification scheme for Fort Walton ceramics which encompasses the Bull Creek phase. Scarry's (1985) type/variety system builds upon the earlier Fort Walton typology (Willey 1949) and provides an extremely important tool for interpreting differences in ceramic assemblages within a tight temporal and regional framework. Scarry's scheme has been incorporated into our examination of the Bull Creek site pottery.
In our examination of pottery from Bull Creek, a variety of attributes will be examined which hopefully will provide a characterization of the ceramic assemblage. Throughout this report an attempt has been made to present examples of decorative styles and vessel forms found on the site. The following sections will include more quantified data. The sources of these data are tabulations from previous work, primarily housed in the National Park Service files and files of Frank Schnell, Jr., curated at the Columbus Museum, and a reexamination of the surviving collections from the Columbus Museum and University of Georgia.

While there was no formal report detailing the laboratory analysis conducted following the 1936-1937 excavations at Bull Creek, a very substantial body of data was recorded and tabulated and much of that paperwork has survived. At the regional WPA laboratory in Macon, detailed attribute analysis was conducted using the McKern System, a procedure which provided standardized techniques for examining all of the site collections excavated during the different WPA-era projects. A thick stack of paper now curated at the Columbus Museum contains coding and analysis sheets for 2240 sherds along with some summary sheets.
Copies of the coding instruction sheets used for Bull Creek have been transcribed and are included as Appendix B. The summary sheets are included where appropriate in the present chapter. Even with the coding sheets, some of the attributes used in the 1930s are difficult to understand and interpret. Some of the attribute terminology, for instance, is no longer used. Copies of the original summary sheets which tabulate the primary attributes of the site's pottery are shown in Figures 131 through 133. Keys have been added to explain the codes.

149

t;; The Bull Creek Site o

$TANlI'EO r r o s:

~N~/Sf[L> /'IItTO'I-O~c.
A/No-/."6$.

7" 1 7f.

,.;J-o

Figure 131. Copy of National Park Service summary sheet of 2240 sherds from the WPA excavations at Bull Creek; various sherd attributes.

Bull Creek Site Material Culture

Temper Size:

a) large b) medium c) small

Cross-sectional Appearance of Paste

2) medium fine 4) medium coarse 6) granular 8) contorted 9) compact

(Hardness - no key needed)

Surface Finish

1) very rough 2) rough 3) outer rim 4) glossy 5) polished 6) vermiculated

Execution

1) crude 2) fair 3) good 4) excellent 5) unobserved

(No key for plain through lip)

Figure 132. Copy of National Park Service summary sheet with counts for temper, paste and texture. 151

The Bull Creek Site

/f-/fi/ --3~-

',.': . ::i

~ . ~ . t:J /J~ / _ f) - ~ ../__ A ()

~-~'f~- -; i - -

i-V; tti! /off-:.II~i/. L!..../-.. 7. ;r~--" ., I ~ :'; r'f'f -.. ..-

,!,.---..

, . .-.-.. --- .... .--...i~'I:... ........ -

. ....

_ -:fJ -If tff' -' ~

1.1'

';I/"/1, '

~

.3 I~~

-

I -.........-'--' . .

.

!/fG}#:---I1J

qtt.-------.----..- -

I~" i,f/1:7'!,tJ....-...

. ...-....-----.--.--

I; -At1 -.-

~

?;1 X "
E::

~ '~~--~---':~iY-3~------;.-~-_-~-;--

53?'

: ... _

_{ .A f ,.
,1B.1M ,

If_ : :

/ ' /7'tJ $"...

h;P ~!f.J J~f

C,&.
~ ,~ ~.,_',_,._._.._._--~._._._-_._.._.
1_17-- ._--_._--

('$- 1f.!, 1/ 1 l 3 ; 1

.1' C! r '8.:: .i.*! l if i :3-... .,----. -.-----------------,

v_-i _ - if) l i E I - 5'7-i?" -.-.-..

:.. / /IA)lDL~.

._3 i .. j .._.)t2 ~.~.------- ~ .......i._:~1.

~1

- ~~._.7--L.._L.1_-H~-....-iji-------~----~-~--..--..j--------------.--I.--1.-~---r.. ---~-------.---+-.i:-t-+ -

-

,
i

"""1 ~-L_-.JLJ1J,:--.-----~:.._f--- '---' - .--------.J--.J..-. '."

.! / .. -A--uu-...Ji __. . _J i __.. ____... _ ... _.

! .1. ....-- " __.~

..1-_- A--~PA. ii'-- .: ... :? (.: ..... . .-.. -. - -.-......._: .. 1-..- .----.----.----~

~-E'%.=l=~~~~~JH -=::=-i=--:~~--~ -==---F,-------~

-- ---~;--'jJ~-~t--r--+. ----.---1 . .-------- .--T--~+-d

,

I

.

.,

.,

Key: Decorations: Part of Pot:

Undeciphered 1 sherd from undetermined portion of pot 2 rimsherd 3 rim and shoulder or neck 4 Rim and handle or lug 5 Rim and Body 10 Handle

Quality of Ware: 1 Excellent 2 Good 3 Inferior

Figure 133. Copy of National Park Service summary sheet of 2240 sherds from the 1936-1937 excavations which quantifies attributes of decoration, quality of ware and vessel portion.

152

According to available records, these sherds were apparently analyzed before July 15, 1936 (Walker 1989:15). This would mean that the sherds were recovered from the village area. As previously mentioned, a lack of a descriptive text accompanying the NPS summary sheets makes interpretation of some of the data rather difficult. Fortunately, a number of these pottery attributes were also tabulated by Schnell (1963). Some categories, including temper size and cross-sectional appearance of paste are not precisely defined and cannot be easily compared to other excavations.

Bull Creek Site Material Culture
A number of detailed drawings which illustrate sherds from the site were also generated as part of that WPA analysis process. Figures 134 and 135 illustrate selected examples of these drawings. The drawings are
num- labeled Bull Creek or abbreviated B.C., indi-
vidually numbered using a Park Service bering system, and initialled by the artist. The drawings that survive do not illustrate the full range of Bull Creek pottery, but appear to concentrate upon incised and punctate examples.

.lJt:lll t~eek
*S.2 9.

Bflfle,.e-e/c
;9f 6o~.

S,c. ~/ .
.509/

Figure 134. WPA drawings of Bull Creek phase plain and complicated stamped sherds. 153

The Bull Creek Site
"Bqll ere ek
#775-.
lJu/1 ("r_
./Yo _ 3 _
Co/_ Ga. .
.13. c _ /?8/.9
Figure 135. WPA drawings of incised and punctate sherds from Bull Creek (lower right is shell tempered). 154

Bull Creek Site Material Culture

Probably the most important collection from Bull Creek, in terms of quantifiable data, resulted from Schnell's excavations of 1959. That produced the first screened collection of material from the site (see Chapter 5). The following text and tables are taken from Schnell's Bull Creek excavation manuscript. Illustrations have been added to further define many of the attributes examined.

As may be seen in [Table 4], the basic analysis of the pottery recovered from the 195gexcavations was done from a purely modal standpoint. The following paragraph correlates these modal categories with ceramic types. All of the pottery presented in this table falls within the Lamar series as partially defined in the Southeastern Archaeological Conference Newsletter (I, 2 :4-6, 9-11 ). Some of the type names used here have not been officially designated, but have found their way into common usage ....

Grit Temper BCVL*

o

3

p-;;-

cm

The plain ware

was divided into three

major categories according

to tempering mode; grit, sand and shell [Figure 136].

Sand Temper

For purposes of present

commentary, it might be

mentioned that the grit tem-

pering category is the same

Shell Temper

as the Bull Creek variant of
Lamar [which designates a Figure 136. Enlarged photographs showing tempering agents found at Bull

paste that contains large

Creek (BCVL represents Bull Creek paste variant of Lamar).

particles of garnet bearing

grit extruding from the exterior and sometimes the interior]. The sand tempered category covers a gradient range from very similar to Bull Creek variant of Lamar to a very fine sand tempering. This very fine sand tempering

historic ware which he designated as Coweta Micaceous. The shell tempered category at the Bull Creek Site is primarily a partially leeched form, indicating some, but not considerable antiquity.

finds its analogy in Hurt's (1975) historic and proto-

155

The Bull Creek Site

The complicated stamped ware may be easily assignable to the type Lamar Complicated Stamped, with the exception that it has the Bull Creek variant of tempering. This exception also applies to check stamped, defined by W.H. Sears (1951 :32) as Mercier Check Stamped. The bold incised category falls in more of a coarse sand tempering than of the tempering characteristic of the Bull Creek variant of Lamar. This type is therefore named as Lamar Bold Incised, with no particular distinction being made. The zone incised category finds its stylistic origins in Fort Walton Zone Punctated, but here again, the tempering is characteristic of the Bull Creek variant of Lamar. The punctated category seems to be most closely related to an undescribed punctated type in the original Lamar series. The engraved category is so sparsely represented here that no attempt will be made to discuss it extensively. It will simply be stated that this one example of engraving was

found on a black burnished sherd and is very similar to earlier Mississippian engraved types except for the fact that it is sand tempered rather than shell tempered. Finally, the red painted category, though statistically insignificant, should be mentioned since it is frequently found as a very minor
type in many Lamar sites ... Figures 137 and 138 illustrate examples of the primary types of surface treatment.
In the original analysis, the material was divided stratigraphically to meet one of the original objectives of the 1959 study. This objective was to determine if any differences could be detected stylistically, modally or typologically. No significant distinctions were detected. The material has therefore, for purposes of presentation in this paper, been lumped together into categories only, with no stratigraphic differentiation made.

Table 4. Ceramic counts from Schnell's 1959 excavations.

Plain (undecorated) Grit tempered Sand tempered smoothed surface burnished surface Shell tempered Total Plain
Complicated Stamped Check Stamped
Total Stamped Incised Punctated
Bold Incised Zoned Punctated Punctated Engraved
Total Incised and Punctated Painted
Red Painted Total Painted
Total Sherds in Lamar Series Collection

2460
474 136
8 3078 4903
94 4997
150 99 4
1 254
7 7 8,336

79.92%
15.40% 4.42% 0.26% 100.0% 98.12% 1.88% 100.0%
59.06% 38.98%
1.57% 0.39% 100.0%
100.0% 100.0%

36.92% 59.95%
3.05% 0.08% 100.0%

156

Bull Creek Site Material Culture
o
ps;-
em
Figure 137. Examples of Lamar Complicated Stamped and Mercier Check Stamped (bottom right) sherds from Bull Creek (various collections). The upper left sherd shows typical overstamping.
157

The Bull Creek Site
em Figure 138. Typical examples of incised and punctated sherds from Bull Creek (various collections).
158

Bull Creek Site Material Culture

Schnell also examined the pottery from the 1950 investigations of Kelly.
The ceramic material recovered from Kelly's 1950 excavation [Table 5] was analyzed in order to compare this assemblage with other analyses made of Bull Creek Focus material ... , A total of 3225 sherds were analyzed from Kelly's excavation. This analysis summarization is a combination of all of the units excavated by Kelly. It might be mentioned here that a detailed analysis of the material recovered by Patterson and Lester
in 1936 has yet to be made (Schnell 1963).

Schnell's analysis of Kelly's material shows lower percentages of complicated stamped pottery compared to his 1959 work. Unquestionably, there is variation in percentages of wares from different collections. In part, this represents differing recovery techniques. However, a comparison is instructive. Figure 139 compares the frequencies of these wares from four investigations at Bull Creek (the 1993 SAS collections are included as the fourth collection).

Table 5. Analysis of sherds from Kelly's 1950 excavation (adapted from Schnell 1963).

Bull Creek Variant of Lamar* Plain

1362

42.2%

Bull Creek Variant of Lamar Curvilinear Complicated Stamped

1394

43.2%

Bull Creek Variant of Lamar Check Stamped (Mercier)

41

1.3%

Plain (moderate tempering)

235

7.3%

Incised

67

2.1%

Zone Punctate

53

1.6%

Plain Shell Tempered

13

0.4%

Plain (fine sand tempered

2

0.1%

Burnished Black Incised

4

0.1%

Bull Creek Variant of Lamar Plain Lug

5

0.2%

Bull Creek Variant of Lamar Plain Disc

4

0.1%

Bull Creek Variant of Lamar Complicated Stamped Disc

1

<0.1%

Red Filmed

3

0.1%

Burnished Black Plain

24

0.7%

Brushed (probable Chattahoochee Brushed)

13

0.4%

Negative Painted

1

<0.1%

Burnished Black with Rim Node

2

0.1 %

Simple Punctated

1

<0.1%

Total

3225

100%

*Bull Creek Variant of Lamar designated a type ofpaste which contains large particles of garnet bearing grit extruding from

the exterior (sometimes the interior).



159

The Bull Creek Site
Schnell's tabulations from his 1959 excavations have been used to define the ceramic composition of the Bull Creek phase, at least with respect to surface treatment. Schnell's excavations produced a predominance of complicated stamped wares
(> 50 percent) followed by plain,
incised/punctate and check stamped (Schnell 1990:67).

By accepting Schnell's (1963) assessment that no stratigraphic differences existed for the pottery recovered in the portion of the site investigated in 1959, we may accept Schnell's figures as the definitive characterization of Bull Creek phase pottery with respect to surface treatment. The small SAS collections show strong parallels in percentages to the 1959 excavations.

Comparison of Bull Creek Pottery From Four Excavations

60
50
40
30
20
10
0--'---
Complicated
Stamped

Plain

~ WPA (2090 Sherds)
D Kelly 1950 (3225Sherds)
Schnell 1959 (8336Sherds)
D SAS 1993 (322 Sherds)

Incised/
Punctate

Check Stamped

Figure 139. Comparison of pottery counts from four archeological investigations of the Bull Creek site (the WPA sherd counts exclude the modeled category).
160

Bull Creek Rim Attributes
Bull Creek vessels display a variety of rim forms and rim modifications typical of both Lamar and Fort Walton. The primary rim attributes of Bull Creek pottery have been

Bull Creek Site Material Culture
examined and quantified for the WPA-era excavations and Kelly's 1950 excavations. Figures 140 and 141 show examples of the more common rim forms from Bull Creek. Figure 142 is a copy of the NPS summary sheet of attributes for 280 rims.

B

c

o

3

&.-

i

em

A, folded, pinched; B, folded, cane punctate; C, Applied stip, pinched;

D, applied strip, notched; E-F, applied strip, noded

Figure 140. Examples of appliqued rims from the Bull Creek site.

161

The Bull Creek Site B E

o

3

&.-

em

A, folded; B, thickened, horizontally expanded and notched; C, noded; D-E, vertical lugs; F-G, notched; H, plain, tapered; I, horizontal lug, notched

Figure 141. Examples of additional rim forms from the Bull Creek site.

162

Bull Creek Site Material Culture

- _._~-'- ----_:..-.-----

! _".".. ~ ". ,.;1

"
i

Figure 142. Copy of National Park Service swnmary sheet of Bull Creek rim attributes.

Key:
Lip Shape Code Not Deciphered

The Bull Creek Site

The data presented in the National

cent), and check stamped (0.7 percent). This

Park Service's summary sheet is valuable for

will be examined again in this chapter. These

examining both rim form and vessel de-

figures suggest that the high percentage of

coration. The summary sheet provides counts

complicated stamped represented by individual

of shape and decoration for rims and lips.

sherd counts (approximately 60 percent in

Because the coding instructions (see Appendix

Schnell's 1959 excavations) may result from

B) are not compatible with the category of lip

less than thirty percent of the vessels produced

shape shown on the summary sheet, those

on the site.

tabulations have not been interpreted.

A detailed examination of Bull Creek

According to the summary sheet

rims was also conducted by Schnell (1963).

figures, the most common rim form curves

The results of Schnell's analysis follow.

outward and is decorated with an appliqued

strip. The rim decoration referred to as appliqued is identified on 37.1 percent of all rims. Appliqued rims are formed by adding a strip of clay at the rim or a slight distance below the rim. Generally, this strip is pinched.

To the present, only general statements have been made about the exact nature of Lamar pottery [at the Bull Creek site]. It was decided, therefore, in order to gain a more exact understanding of the nature of Lamar types, and of the Bull Creek Focus in particular, to make a

When the rim collar is wide and extends down

detailed rim analysis. This was done by dividing

from the rim, it is generally referred to as a folded pinched rim and represents a typical Lamar jar style. The decoration category "gridbarfI as used in the summary sheet appears to be the same as check stamped. The

the rims from the 1950 excavation by Kelly, a total of 278 rim sherds, into a number of categories according to rim form and to designate pottery type according to the body portion below the rim [Figure 18]. The following comparative data resulted [Table 6, Figure 144].

remaining ter-

minology appears to

be similar to that used today.
The category of pot decoration is important as it relates to other vessel attributes, discussed later. The summary sheet data may be used to approximate some vessel count traits. The figures comparing pot .decoration and rim decoration show a predominance of plain vessels (57.9 percent), followed by

FOlded,~~:: "i""'~"":""':'''~''~~~~~~~=O=38 26

Overhanging fold

13

Overhanging fold, thinDed 11"1"'1"1''''1''''';' 91

Outeurved lip

13

Thinned,outeurved lip

ThimIed, straight

Roundededge l~~~~~~ggg~~~ Sttaight, rouDded lip

28

Sttaightlipped

14

KeeIed-back Keeled-back, rolled lip

.. s

Keeled-bacl< DCId>ed Rolled lip

Distribution of Rim

Thicl=ed lip

Thickened 0U1Side lip

.s

PiDcbed ]".E,,,~~

Types for All Sherds Recoveredby Kelly

NOlcbed
Flat lip, notched ]z::::::::::S:~~~::J

Noded Roundedlip, notched LughaDdle J~~gg~~=1=1 ==~22

Luted

J~g~~ Lutednotched

6

Lutedpinched

LutedWldifferentiated

PuDctale

Reed punctale ScaUoped ..j:::'.:

-+-

...l-

--'--_ _- - ;

stamped (28.9 percent), incised or

o

10

20

30

40

Number

punctate (12.5 per- Figure 143. Graph showing rim varieties identified from a collection of 278 sherds

(graph produced from figures in Schnell 1963).

164

Bull Creek Site Material Culture

Table 6. Detailed rim analysis of 278 sherds from Kelly's 1950 excavations at Bull Creek.

yi

i

......i<.......

} .......
.... ......

iit . ttt i ......... ........ } ..

'<



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

..:':':

......


i ..
< .:.

.o.: :..,



<

c

.............

....... . .....

1 Folded

25

2 Folded, thinned

29

9

3 Overhanging fold

10

3

4 Overhanging fold, thinned

9

5 Outcurved lip

9

1

6 Thinned, outcurved lip

2

1

7 Thinned, straight

1

8 Rounded edge

23

2

9 Straight, rounded lip

16

1

10 Straight lipped

4

2

11 Keeled-back

1

12 Keeled-back, rolled lip

5

13 Keeled-back notched

4

14 Rolled lip

3

1

15 Thickened lip

1

16 Thickened outside lip

5

17 Pinched

2

3

18 Notched

2

19 Flat lip, notched

5

4

20 Rounded lip, notched

21 Noded

20

2

22 Lug handle

10

23 Luted

2

24 Luted notched

5

25 Luted pinched

8

26 Luted undifferentiated

1

27 Punctate

1

28 Reed punctate

1

29 Scalloped

1

Grand Total

204

30

2

1

4

9

2

4

4

1

3

4

3

1 1 1

25

15

1

26

38

13

9

13

3

1

29

18

14

1

5

4

4

1

5

3

9

2

16

3

22

11

3

6

8

1

1

1

1

4

278

165

The Bull Creek Site

30
25
8' 20
N II
6... 15
1l
~10
5

Rim Type Distribution for Plain Ceramics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 192021 2223 242526272829

':us
'l
II
6... 0)
,t:J
~o

Rim Type Distribution for Complicated Stamped Ceramics
1 Irapl;;! jl~ll71l~ llllrzJ lll~ ~1~ 11r:a 1111111rJ

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 2021 22 23 24 25 2627 28 29

Vl
J : 10
~ ,~, ,~, I : I I I ,Ill, ,

Rim Type Distribution for Incised Ceramics

I IIII I I

I I"',","', , , , I

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 2223 2425 2627 28 29

r J Vl

Rim Type Distribution for Zoned Punctate Ceramics

I

~ '~, : , , , ,", , , ,Bl,m, , I , I I ,",

,I , , , ,I ,

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 2425 2627 28 29

r:J

Rim Type Distribution for Check Stamped Ceramics

I

~ Z 0 rJ I I I I i i I I I I 1 I 1 I I 1 Tj--'1--'1--'1--'1--'1"""'1"""'1"""'1"""'1-r-j-r-j-
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 2627 28 29

Figure 144. Series of graphs comparing rim type and vessel treatment for sherds examined by Schnell (see Table 6 for list of rim types corresponding to the numbers on the graphs).

166

Bull Creek Site Material Culture

Schnell examined rim types in great

Oertel 1977:39). This comparison seems

detail, but his manuscript does not contain

inconsistent with Schnell's dating of Bull

definitions for the rim types. While most of

Creek as an earlier phase and may indicate that

the rim types are easily interpreted, other cate-

certain concepts regarding rim fold width

gories cannot be accurately understood without

might stand further scrutiny, especially with

formal or measured descriptions. There is

respect to Fort Walton influences.

some discrepancy with respect to counts for

ceramic types or pot types. Schnell's counts

A second set of measurements was

indicate that 204 of 278 sherds represent plain

taken while examining all available collections

bodied vessels. This percentage of 73.4

for vessel form data. Sherds of sufficient size

percent is substantially higher than the 57.9

to allow measurement of rim diameter, deter-

percent noted on the NPS summary sheets.

mine vessel shape, and determine vessel de-

This may represent differences in the col-

coration were also measured for width of

lections, differences in analysis techniques, or

folded rims when present. Most of the vessel

simply a typographic error in that manuscript.

form data will be presented in a later portion

A quick reexamination of sherds in Kelly's

of this chapter; however, the rim fold width is

collections, now curated at the University of

pertinent at this point. Table 7 compares ves-

Georgia, produced a figure of approximately

sel rim diameter to the width of folded pinched

53 percent plain vessels, which is more in line

rims for fifty sherds. All sherds represent

with the NPS figures.

typical Lamar constricted rim jars. These data

show that while the smaller jars do display re-

The width of rim folds (applied rim

latively smaller rim folds, there is not a direct

strips) is today considered a critical analytical

correlation indicating increasingly larger rims

measurement with respect to Lamar pottery.

with increasingly larger vessel diameters.

The generally accepted proposal is that these

Beginning at approximately 32 em diameter,

rims get wider through time. Hally's exa-

the rim folds tend to cluster around 21 mm in

mination of folded pinched rims from Park's

width.

Mound led him to conclude that the rims

were sufficiently wide to indicate a late date

(late sixteenth or early seventeenth century) Table 7. Comparison of rim diameter and width of

for the occupation at that site (Hally and

folded pinched rim fold (sample size: 50 rim sherds)

Oertel 1977:39). During the course of the

present project, measurements were taken

Size Range Average Size

on folded pinched rims from various col-

Rim

Width of

Width of

lections. A total of 112 folded pinched rims Diameter Count Rim Fold

Rim Fold

from the NPS collection were measured. That collection also included 59 rims with distinctly applied pinched strips which were generally placed about one em below the lip. This is mentioned because there is

14 em 18 em 20 em 22 em 24 em

1

13 mm

1

16 mm

1

16 mm

2

12-16 mm

2

16-18 mm

13 mm 16 mm 16 mm 14 mm 17 mm

some subjectivity involved in distinguishing sherds with smoothing of the upper portion of an applied strip from sherds that are generally described as folded pinched rims. The folded pinched rims in this collection ranged in width from 11 to 28 mm and

26 em 28 em 30 em 32 em 34 em 36 em 38 em

1

17 mm

4

12-21 mm

2

16-18 mm

8

16-27 mm

8

14-24 mm

8

15-26 mm

3

14-26 mm

17 mm 17 mm 17 mm 21 mm 18 mm 21 mm 21 mm

averaged 20.3 mm, wider than the 18 mm average from Park's mound (Hally and

40 em

5

15-25 mm

21 mm

42 em

3

18-25 mm

21 mm

44 em

1

20mm

20mm

I

--------------------------------------------------~

167

The Bull Creek Site

Two additional attributes of rim decoration, handles and adornos, are inadequately tabulated by the researchers of the Bull Creek site. Chase (1957) noted that handles were not found on Lamar Complicated Stamped vessels at Bull Creek and there are no references in his manuscript to handles on other pottery types. Chase does note that rim adornos, usually lugs, nodes and more rarely effigies, usually of birds, occur on Fort Walton-style vessels and a type of burnished black slip ware (Chase 1957). Schnell's summary tables list 11 lugs while the National Park Service summaries reference six handlesllugs. The handles appear to be vertically and horizontally placed lug handles. An examination of available collections produced several examples of lug handles (see Figure 141, second row) and two or three flattened fragments of pottery which appear to be portions of small strap or loop handles. Large Lake Jackson-style loop handles which appear so frequently on Mississippian sites in the region do not appear to occur at the Bull Creek site. At the Bull Creek site, lug handles appeared to be limited to plain and incised/punctate bowls.

Very remarkable rimsherd, in the form of a woman, with her arms raised to the back of her head. Facial characteristics more or less of conventionalized form. Found 60" below the surface, 27" below the occupation level. 24" N.
6" E. Sta. 4-15L13 (Lester 1937:91).
The old National Park Service photograph fails to do justice to the exceptional features lauded by the preceding description. A search of the NPS collections suggests the artifact has been lost so it cannot be rephotographed.
At Bull Creek, effigy rim adornos appear to be associated with Fort Walton-style bowls. Similar animal style adornos are found consistently on Mississippian sites and have been reported from other late Fort Walton and Bull Creek-related sites in the region (Willey 1949:504; Broyles 1971).

The production of ceramic effigy figures is an important aspect of the Bull Creek pottery complex. This is most apparent in the effigy vessels and pipes found with burials. Effigies are also common on vessels found throughout the village area.

Figures 145 and 146 illustrate examples of effigy adornos from the Bull Creek site. Most of the illustrated effigy objects are animal figures of some type. The object illustrated in Figure 145 has been identified as a human representation. The effigy adorno was recovered from the midden deposits in the cemetery during the WPA excavations and numbered as Find 69. NPS notations describe the object as follows.

o

3

e

cm

Figure 145. Photograph of human effigy adorno (copy of photograph from National Park Service files).

168

Bull Creek Site Material Culture

o

3

em

Figure 146. Animal effigy rim adornos from Bull Creek site collections (size of lower right is unknown).

Vessel Form Attributes
During the course of the present study all available collections of pottery from Bull Creek were examined. As one facet of that examination, all large rim sherds were pulled and selected sherds were analyzed in detail to procure vessel form data that would allow further comparison to other Mississippian site analyses (Figure 147). An attempt was made

to mend vessels and count only one rim sherd per individual vessel. Sherds were included in this analysis if they were sufficiently large to determine rim diameter, rim type, body decoration and vessel profile. Generally, a minimum rim size greater than 4 em was required. Vessel form represents one aspect not sufficiently examined by previous researchers at Bull Creek. Our examination produced data for 219 vessels (Table 8).

169

The Bull Creek Site
Fort Walton Vessels (Scarry 1985)

Coastal Lamar (Braley 1990)

0 CJ C)

A

B

C

C/ CJ C7

0

E

F

0 00

G

H

I

000

,J

I(

~

A) simple bowl; B) carinated bowl; C) cazuela bowl; D) simple restricted bowl; E) collared bowl; F) flaring rim bowl;
G) collared jar; B) evened rim jar; I) evened rim conical jar; J) restricted jar; K) beaker; L) bottle

'C:.7 C )

A

B

00

C

0

00



F

u <::r

G

H

A) simple open bowl; B) carinated bowl; C) constricted rim bowl; D) flaring rim jar;
E) carinated jar; F) flaring rim pot; G) flaring rim plate; B) recessed neck jar

North Georgia Barnett Phase Lamar Vessels (Hally 1986)

c=J 0 C7 t=J Q

A

B

C

0

E

0 Q U0

F

G

H

I

A-B) rounded bowl; C) flaring rim bowl; D) "gravy boat" bowl; E) carinated bowl; F) pinched rim jar; G) Mississippian jar; B) carinated jar; I) bottle

Figure 147. Examples of vessel forms identified from various Mississippian localities.

170

Bull Creek Site Material Culture

Table 8. Vessel forms and counts from Bull Creek.

Vessel Form
Everted Rim Jar Carinated Bowl Collared Globular Bowl Simple Constricted Rim Bowl Open Bowl Effigy Bottle (Dog Pot)

Count
83 36
2 80 15 3

moderately everted to conspicuously flaring
rims (Figure 148). Within the collection, rim
modifications include folded pinched rims (N = 50), applied strip below the rim that is noded (N = 10) or pinched (N = 5), indi-
vidual nodes (N = 6) or lugs (N = 1) at or
below the rim. A small number of jars have unmodified rims (N = 11). Body decoration consists of complicated stamped (N = 65),
plain (N = 17), and check stamped (N = 1).

Total

219

Of the 219 vessels, only four (the three dog pots and one small bowl) are burial objects. The remainder were recovered from the general midden or trash pits and are considered domestic or utilitarian vessels. Each vessel form will be described separately. Measured attributes of individual sherds are presented in Appendix C. That data will be summarized in this chapter.

Everted Rim Jars (N = 83) generally correspond to Hally's pinched rim jar (see Figure 147) and represent an extremely common vessel type on Lamar sites. Our category applies to jars with welldefined neck constrictions and

o

6

,......

i

em

Figure 148. Partially reconstructed Lamar Complicated Stamped everted rim vessel with appliqued pinched rim strip (UGA collection, Bull Creek site provenience unknown).

171

The Bull Creek Site
Within this collection jars account for approximately two-thirds of the vessels decorated with complicated stamping from Bull Creek. The remaining vessels with complicated stamping are simple bowls.
Measurement of rim orifice diameter is the primary means of comparing jar size from different sites. Jars generally reach their greatest diameter a few centimeters below the neck constriction. Relatively few of the Bull Creek rims were of sufficient size to determine maximum vessel diameter but measurements of six sherds showed the increase was only 2 to 4 em greater than rim diameter.
Rim diameters for jars range from 14 to 44 em, Across this spread, 36.1 percent are small to medium jars of 30 em or less and the remaining 63.9 percent are medium to large jars in the 32 to 44 em range. A pronounced peak exists between 32 and 36 em accounting

for 41.0 percent of all jars. Jars account for 37.9 percent of the total vessels in this collection. Figure 149 shows the range of rim sizes for the Bull Creek jars.
Large jars are considered storage vessels, with a rim shape that facilitates pouring and attachment of a protective covering (Shapiro 1990:155). Research by Hally (1986:285-286) indicates the largest jars (40 em and larger) were used for long term storage while medium sized jars were used as general purpose cooking vessels. According to the research of Shapiro the proportions of larger storage jars (> 30 em in rim diameter) within the overall assemblage is an important indicator of the intensity of site occupation. The relative counts for jars compared to other vessel types at Bull Creek and the relatively high frequency of larger jars are comparable to permanently occupied late Mississippian sites (Hally 1986; Shapiro 1990).

EVERTED RIM JARS
12
10
8
~ 6
U 4
2
0'----10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 RIM DIAMETER (em)
Figure 149. Graph showing the range of rim sizes for everted rim jars from Bull Creek. 172

Carinated Bowls (N = 36) are defined as constricted rim vessels with distinct shoulder break, which sometimes forms a well defined projecting keel or carination. Similar vessels from contemporary sites are generally flat-bottomed. The typical Lamar cazuela

Bull Creek Site Material Culture
bowl is included in this category. At Bull Creek, carinated bowls are typically decorated with Fort Walton incised and zone punctate patterns (Figure 150). The lower restored bowl has cane punctation at the shoulder that is not typical of the Bull Creek site.

o
s-.

3

em

Figure 150. Examples of Fort Walton carinated bowls from Bull Creek. 173

The Bull Creek Site
All but one of the carinated bowls in this collection are decorated by incising (N = 11) or a combination of incising and zone punctation (N = 24). The style of the decoration is basically Fort Walton. The presence of Lamar bold incising is questionable. Rim shapes vary from rounded to flattened. Modification includes notching (N = 16) and incising (N = 3). Incised rims are from vessels where the incised body decoration extends to the lip of the rim. The rims of seventeen vessels are not modified. At Bull Creek, rim notching occurs on carinated bowls and simple constricted rim bowls but relative proportions are substantially higher in this vessel type.
Figure 151 shows the distribution of rim diameters for carinated bowls from this collection. The vessel occurs in sizes from 14 to 40 em but there are some indications that some sizes were more common. Almost twenty percent of the vessels measure 24 em in rim diameter. A second peak is evident at 30 to 32 em (30.6 percent of all carinated bowls).
The size distributions of carinated bowls at Bull Creek is similar to that shown

for the Late Lamar Barnett Phase (Hally 1986:275). Hally suggests that carinated bowls were used to cook and serve food that was of liquid consistency and would require a minimum of cooking. Small and large sizes of carinated bowls reflected the quantities of food prepared (Hally 1986:288-289). This vessel type was widely distributed in the midden and trash pits at the Bull Creek site. At least some examples of the vessel type were found on the floor of Lester's second structure.
Collared Globular Bowls (N = 2) are represented by two examples from Bull Creek. The most complete example was found in a trash pit near Lester's Structure 1 (see Figure 34). The vessel form has been described as a deep globular bowl with incurving walls which constrict at the base of a short neck or collar (Willey 1949:498). The previously illustrated vessel is plain with a rim diameter of 26 em. The rim is unmodified. The second vessel is check stamped with a slightly rolled rim and a rim diameter of 36 em. These bowls may have been used as cooking or serving vessels.

CARINATED BOWLS

6
s~ 4
U2
o
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 RIM DIAMETER (em)
Figure 151. Graph showing the range of rim sizes for carinated bowls from Bull Creek. 174

Simple Rounded Bowls (N = 80) occur with essentially the same frequency as jars. The category includes any bowl with a constricted orifice. In some cases the rim is only slightly incurving. This category is

Bull Creek Site Material Culture
distinguished from other simple bowls that slope outward at the orifice. The category combines vessel types similar to Scarry's simple restricted bowl and simple bowl (see Figure 147). Figure 152 shows examples of the range of profile variability.

o

3

I"l~-~--il-_~~

em

,.;:-~.::::.==~::~:.~:~~:::f:~;;:, ...__..,....__...~...,...._.....,,.,.'".:':-::... ;7..

Figure 152. Examples of simple rounded bowls with slight to moderate rim constrictions. 175

The Bull Creek Site
Simple rounded bowls were decorated
with complicated stamping (N = 34), incising
(N = 11), a combination of incising and zone punctation (N = 2) and check stamping (N = 1). The remaining thirty-two bowls were undecorated. Within this collection sixty percent of all vessels are decorated. The percentage of Lamar complicated stamping (42.5 percent) is second only to jars at Bull Creek.
Rim types are highly variable in this vessel category. Fourteen different rim types
were identified, including unmodified (N =
33), notched (N = 22), pinched (N = 2), noded (N = 2), rolled (N = 3), incised (N = 1), applied notched strip (N = 2), applied noded strip (N = 1) folded (N = 3), folded
and pinched (N = 3), folded and notched (N
= 3), and thickened (N = 2). The collection
also contains one T-rim and two rims with effigy adornos.
Rim diameters range from 10 to 42 em with two-thirds being greater than 26 em

(Figure 153). Within this collection, sixty percent of the vessels fall within the range of 28 to 36 em.
Hally's work with Barnett phase vessels suggests that large rounded bowls, which equate in size to the majority of Bull Creek vessels, were used primarily for preparing and serving large quantities of viscous and solid foods but only minimally for heating these foods. Smaller bowls were used for similar functions but in limited quantities (1986:289).
Simple Open Bowls (N = 15) refers to vessels for which the widest part of the bowl occurs at the rim orifice. These bowls differ from the previous category primarily in this one attribute. Figure 154 illustrates a typical open bowl. A restored example from the WPA excavations was shown previously in Figure 152).

SIMPLE ROUNDED BOWLS

10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 RIM DIAMETER (em)
Figure 153. Graph illustrating range of rim diameters for simple rounded bowls from Bull Creek. 176

Bull Creek Site Material Culture

"

-. ~

_- " , ......

...... ~ "

'~- ......... ----------- ~~-

o,........

3
;

em

Figure 154. Example of a simple open bowl from Bull Creek.

Body decoration for open bowls includes complicated stamping (N = 5),
incising (N = 1), and combined incising and zone punctation (N = 1). The remaining eight
bowls are plain. Rims are generally unmodified (N = 6), or thickened (N = 3). Variations include thickened and L-shaped (N = 2), and lug handles (N = 1). One vessel contains incising extending to the lip and one vessel is notched. One vessel contains a simple incised design on the interior lip.
Rim diameters are shown in Figure 150. The range of 10 to 42 cm is similar to other vessel types from the site but unlike the other types there are indications of size groupings. The smaller sized vessels, which range in size from 10 to 24 em include one small vessel from Burial 18, which appears to be a specially produced mortuary vessel. This small bowl, only 10 cm in diameter, is crudely

decorated with a Fort Walton zone punctate design (see Figure 59). That bowl's combination of diminutive size, sloppy decoration and even vessel shape is unique in the site collection. A second small bowl was also recovered from the cemetery, but that vessel has been lost and was not adequately described in the field notes. The small bowls include three plain, two incised, and three complicated stamped vessels. Rims are variable but primarily unmodified (N = 4). The larger bowls are either plain (N = 4) or complicated
stamped (N = 2) with thickened or L-shaped
(N = 3), plain (N = 2), or notched (N = 1)
rims. The one isolated vessel at 28 em is a slightly flaring rim bowl with interior lip incising. Aside from mortuary use for the one example, the differences in sizes may reflect uses similar to large and small varieties of simple rounded bowls.

177

The Bull Creek Site

S~LE OPEN BOVVLS

4
~ 2
o

10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 RIM DIAMETER (em)

Figure 155. Graph illustrating range of rim diameters for simple open bowls from Bull Creek.

Bottles from the Bull Creek site consist entirely of mortuary ves-
sels. Water bottles similar to the category shown by Scarry (see Figure 147) should be expected at the site. Perhaps this omission is the result of sampling error or misidentification of vessel types.

Table 9. Measurements of effigy bottles from Bull Creek.

Vessel Burial Height Length Width

Orifice

Number No. (em) (em) (em) Diameter (em)

V.1I7 3

26.5 30.0 19.0

9.5

V.1I8 7

27.0 27.5 19.5

9.5

V.1I9 16

32.0 28.0 18.0

8.0

This category includes three bottles commonly known as the Bull Creek cemetery dog pots. Figure 156 shows the three vessels and three examples of similar vessels from other sites in the region. The three vessels from Bull Creek and the one vessel from Neisler Mound are considered the only known examples of this one type of negative painted dog effigy vessel (Schnell 1990:67). Selected measurements for the three vessels are shown in Table 9. The vessels will be examined further in the following section which deals with examination of Bull Creek pottery types.

178

BULL CREEK BURIAL 3

BULL CREEK BURIAL 7

Bull Creek Site Material Culture
BULL CREEK BURIAL 16

NEISLER

CEMOCHECHOBEE BELL FIELD

o

3

em

Figure 156. Illustrations of effigy bottles from the Bull Creek site with comparative examples from other western Georgia:" sites.

The Bull Creek Site
Summary of Pottery Types
The pottery from the Bull Creek site might be viewed as something of a hybrid of ceramic styles with some similarities to Lamar and some similarities to Fort Walton. The pottery does not always fit snugly into established type descriptions. Schnell's term Bull Creek variant may be as appropriate as any means for a designation of the pottery series. Basically, this means that vessel types and decorative styles typical of either Lamar or Fort Walton occur on the site, but these vessels are tempered with a distinctive, gametrich grit. The size of the temper varies from 0.7 to 3.0 mm (average 1.4 mm). One problem with this is that we do not know how localized this use of tempering agent is and whether it reflects cultural distinctions. While heavy grit tempering is generally characteristic of late Lamar in western Georgia, the distinctive nature of the site's temper is quite probably a localized phenomenon reflecting the local clay and sand resources.
The recognition of a distinctive Bull Creek variant of temper does not mean that every vessel on the site was tempered the same. At the Bull Creek site, some vessels of all pottery types are tempered with finer grit or sand. This applies most strongly to the incised wares. Still, there needs to be some means to improve upon the use of purely descriptive terminology for analyzing the pottery from the site. As expected, our examination of vessel form clearly indicated that certain surface treatments (decorations) were more common on certain vessel types. This is consistent with other research where the combined attributes of decoration and vessel form have been used successfully in the examination of pottery types (Shnell et. al 1981; Scarry 1985).
Schnell's 1963 manuscript references an appendix that contains formal descriptions of the pottery types found at Bull Creek. Unfortunately, this appendix could not be found
180

and we are limited to the more general characterizations found within the manuscripts of Schnell and Chase. In the following summary, descriptive categories used for Bull Creek pottery will be examined more closely within the context of formally defined pottery types. For the most part, the previously cited descriptions by Chase and Schnell have adequately defined these pottery types, particularly when referenced to the original type descriptions. Lacking in the descriptions of Chase and Schnell are adequate comparisons to Fort Walton ceramics. For this reason, the work of Scarry (1985) will be incorporated into these descriptions.
Plain wares from the Bull Creek site may be considered variants of Lamar Plain or Lake Jackson Plain (Willey 1949:458). As previously noted, Schnell (1963) divided plain wares from Bull Creek into three categories according to tempering. These categories consisted of grit (79.9 percent), sand (19.8 percent) and shell (0.2 percent). Schnell implied that the heavy grit tempered category should be referred to as Bull Creek variant of Lamar. Schnell also noted that the finest sand tempered was similar to historic and protohistoric ware designated as Coweta Micaceous (Hurt 1975). Since we now know that the Bull Creek site does contain an historic Creek component, the association of this pottery type with the Bull Creek phase might be questioned. Schnell noted that the few shell tempered sherds were partially leached, indicating some, but not considerable antiquity (Schnell 1963). There is the possibility that the shell tempered sherds postdate the Bull Creek phase (Abercrombie phase).
There are several characteristics associated with rim modifications at Bull Creek which are common on Lake Jackson Plain vessels but uncommon on Lamar vessels. Rim notching and applied rim strips that are either pinched or noded are particularly common on Lake Jackson Plain vessels (Willey 1949:459).

Bull Creek Site Material Culture

Scarry applies the type Lake Jackson Plain varieties Ingram and Tallahassee to the Bull Creek phase (Scarry 1985:221). Variety Tallahassee contains small amounts of grog temper but has not been recognized in any examination of pottery from the Bull Creek site. Lake Jackson Plain variety Ingram, which does apply, is described as a variety with coarse grit and smoothed surfaces, common to the Rood and Bull Creek phases and in the Appalachicola Valley, the Yon phase (Scarry 1985:221).

Complicated Stamped Wares found on Bull Creek phase sites have been universally described as Lamar Complicated Stamped. This absence of complexity results from acceptance of the proposition that, by definition, any poorly executed complicated stamping should be called Lamar. Schnell's (1963) observation was simply that the complicated stamped ware may be easily assignable to the type Lamar Complicated Stamped, with the exception that it has the Bull Creek variant of tempering (Schnell 1963).

Scarry's variety Ingram is based in part upon the type Ingram Plain which was defined at the Cemochechobee site in Clay County (Schnell et al. 1981:185). The definition for Ingram Plain suggested synonymy with the types Lake Jackson Plain, Pinellas Plain, Coweta Micaceous Utility Ware and a relationship to Lamar Plain. Ingram Plain vessels were primarily open bowls often exhibiting notched lips and sometimes exhibiting applied notched strips. The type was considered "the apparent predecessor of Lamar Plain as it is represented in the Lower Chattahoochee Valley" and further stated that with the addition of pinched folds and fillets, the type probably developed into Lamar Plain of the succeeding Bull Creek phase, but distinguishing the two types is difficult (Schnell et al. 1981:185-188).

Scarry examined complicated stamping more critically, with the caveat that his Lamar Complicated Stamped pottery includes all coarsely tempered complicated stamped vessels found in the Fort Walton area (Figure 157). Scarry notes that Lamar Complicated Stamped pottery probably developed outside the Fort Walton area in the South Appalachian Mississippian style area and was introduced after the Rood phase (Scarry 1985:221-222).

Based upon the results of previous research we would appear to have a choice of what to call the plain ware from Bull Creek. If the site is viewed from a Fort Walton perspective, the type Lake Jackson Plain variety Ingram is appropriate. If the site is viewed from a Lamar perspective, the plain ware may be called Lamar Plain. With respect to vessel form, the Bull Creek site plain wares are associated primarily with simple bowls, accounting for 41.1 percent of those types, and to a lesser extent with jars (20.5 percent).

Figure 157. Examples of motifs for Lamar Complicated Stamped var. Bull Creek (top) and var. Early (from Scarry 1985:223).

181

The Bull Creek Site

Scarry defined two varie-

ties of Lamar Complicated

Stamped that should apply to the

Bull Creek site (see Figure 157).

Lamar Complicated Stamped vari-

ety Bull Creek is a coarse sand-

tempered ware with figure-eight

designs. The type site was not

Bull Creek but site 9Ca51 (Broyles

1962), which Schnell now con-

siders to date to the subsequent

Stewart phase (Schnell and Wright

1993:21). Scarry also defined

Variety Early for pottery con-

taining motifs consisting of sets of

concentric circles, the centers of

which may contain raised dots or

crosses. Variety Early was con-

sidered the dominant type of the

Yon phase in the Apalachicola

Valley and present in the Bull

Creek phase.
At the Bull Creek site, Lamar Complicated Stamping was

o

3

e-- -

em

found on 78.3 percent of the jars Figure 158. Example of coarse grit tempered, bold check stamped

and 41.1 percent of the simple

pottery from the Bull Creek site.

bowls present in the vessel form

analysis collection of 219 rims. Complicated

At Bull Creek, check stamped designs

stamping was not associated with carinated

were identified on three vessel types, jars, col-

bowls, collared bowls or bottles.

lared bowls, and simple bowls. One example

of each was found in a collection of 219 rims

Check Stamping at Bull Creek is

examined for vessel form data.

generally referred to as Mercier Check

Stamped based upon similarity to material

Incised and Zone Punctate Wares

from Kolomoki (Sears 1951). Schnell noted

represent perhaps the most perplexing pottery

that Mercier Check Stamped at 9Me 1 should

styles found at Bull Creek. Various

be viewed as a Bull Creek variant because of

researchers have referred to the pottery as

its distinctive grit tempering (Figure 158).

either Lamar Bold Incised or Fort Walton

Scarry views the pottery as a variant of Leon

Zone Punctate. Schnell (1963) suggested that

Check Stamped, which includes all coarsely

because bold incising at the site occurred more

tempered and boldly check stamped ceramics

with a coarse sand tempering than with the

in the Fort Walton area (Scarry 1985:225).

tempet characteristic of the Bull Creek variant

Scarry prefers Leon Check Stamped variety

of Lamar, the name Lamar Bold Incised

Mercier as a name for the bold check stamped

should be used.

pottery (often with checks up to 1 em) of the

Bull Creek and Yon phases.

182

Schnell felt the zone incised and punctate sherds were variants of Fort Walton Zone Punctated, with tempering characteristic of the Bull Creek variant of Lamar. The punctated wares were viewed as more closely related to an undescribed punctated type in the original Lamar series (Schnell 1963). During our examination of available collections we found these distinctions difficult to replicate.
Chase's efforts to distinguish the two styles were rather vague. Chase identified Lamar Bold Incised by deeply incised, usually curvate lines, sometimes consisting of only two or three parallel lines circumscribing the vessel or with interlocking loops. Chase included a variant of the Fort Walton Punctate within Lamar Bold Incised which for most sherds could not be distinguished from the classical Fort Walton punctate type (Chase 1957).
Hally found that a distinguishing factor of Lamar Bold Incising at Park Mound was simply a greater width of Lamar incised lines compared to Fort Walton incising (Hally and Oertel 1977:45). Using the criteria of greater line width Hally was able to identify only 14 Lamar Bold Incised sherds from a collection of 148 incised sherds from Park Mound (Hally and Oertel 1977:45).
While the more southerly located Bull Creek site might be expected to produce even fewer typical north Georgia Lamar incised sherds, an effort was made to test Hally's conclusion that Lamar Bold Incised has measurably wider lines than other types of

Bull Creek Site Material Culture
mcismg. Measurements were taken of incising on 61 sherds from the NPS collection. Line widths for 34 incised and punctate sherds which appeared to be Fort Walton styles ranged from 0.7 to 3.0 rnm (average 1.7 mm). . Line widths for 27 incised sherds which lacked punctations ranged from 0.5 to 3.0 mm (average 1.8 rnm). The 27 incised sherds were then sorted, somewhat subjectively, into Lamar-like and Fort Walton-like designs.
Sherds identified as Lamar-like (N = 16)
averaged 2.25rnm wide while the Fort Walton-like sherds averaged only 1.2 rnm wide. This does indicate some difference may exist within the collections, but a substantially larger sample of sherds should be measured before firm conclusions are reached.
The same collection of 61 sherds were examined to address Schnell's assertion that Lamar Bold Incised sherds at the Bull Creek site differed in temper. The sherds were sorted into three temper types. While our sorting criteria probably differ somewhat from Schnell's, in this one small sample, there does appear to be a predominance of coarse sand and fine sand tempering for the incised pottery. However, this difference appears to relate to both zone punctate (probable Fort Walton) and incised (possible Lamar) pottery. A low proportion of Bull Creek variant tempering in the zone punctate pottery would seem at odds with Schnell's implication that most sherds contained coarse temper (see page 156). The possibility exists that apparent tempering differences reflect, to some degree, variation which exists in different collections from various areas of the site.

Table 10. Comparison of temper for zone punctate and incised pottery from Bull Creek
~%'iI

Incised (possible Lamar Incised)

27

Zone Punctate (probable Fort Walton) 34

18.5% 35.3%

44.5% 20.6%

37.0% 44.1 %

100% 100%

183

The Bull Creek Site

The incised motifs associated with the Bull Creek and Stewart phases are clearly different from "typical" Lamar assemblages. These motif differences are obvious when sites from different areas are compared. Figure 159 illustrates the differences in the most common incised motifs from three phases. These illustrations are based upon Hally's research (Hally 1994: 153), but similar work has been conducted by other researchers (Snow 1990; Braley et al. 1986) and found to be an extremely valuable research tool for characterizing local phase assemblages. Importantly, Hally found that within each phase certain motifs occur with greater frequency. This is especially interesting for the Bull Creek phase because Hally notes that one motif (Figure 159, third row, left) occurs on more than fifty percent of the identifiable sherds in his sample (Hally 1994:154). In light of the fact that illustrated published data available to Hally comes primarily from two sites (Park Mound and 9Cy51), which Schnell now believes postdates the occupation of the Bull Creek site, Hally's observations are particularly important for comparing and contrasting incised motifs of the Bull Creek phase and Stewart phase.

@jj)j;
Barnett Phase
~~
(iW(I))

The identification of Lamar and Fort Walton design motifs is only one way of defining types of incised pottery. Other researchers working in the Chattahoochee valley have developed typologies based at least in part on vessel form (Schnell et al. 1981 and Scarry 1985). Basically, these typological schemes provide a means of extending Fort Walton type descriptions to the Bull Creek area allowing greater interpretation of various incised motifs.

Tugalo Phase

Scarry's type-variety approach to the incised wares of the region provides a useful tool for interpreting the variety of designs found at Bull Creek. Table 11 presents Scarry's breakdown of incised wares which could occur at the Bull Creek site. Figure 160 illustrates examples of each.

Bull Creek Phase
Figure 159. Comparison of common Lamar Incised designs from three Lamar phases (redrawn from Hally 1994:153).

184

Bull Creek Site Material Culture

Table 11. Presentation of Scarry's (1985) incised types and varieties of the Bull Creek phase.

IType

.... .'.'.".1. ,,iyarietY tl

Ilf:scriPFi()ri> ' .. -,- -"".",.".

1

Lamar Bold Incised

t
{

>>. .

it ....'

y{

>

Subsumes all vessels tempered with coarse sand or grit that are decorated with a band of broad, deeply incised lines on the upper portion of vessel (carinated bowls and cazuela bowls). Motifs are continuous bands of curvilinear scrolls or lines of pendant loops composed of 5 to 15 lines (Scarry 1985:221).

Fort Walton Incised

ii>
>...........'...'.. ,.,.....

, Subsumes all ceramics in the Fort Walton area with a sandy
y paste with decorations formed by incised lines with
punctation filled zones or occur on punctated backgrounds.

Fort Walton Incised

Fort Walton

Characteristic design is a band of running scrolls (both curvilinear and rectilinear) superimposed on a band of punctations. Vessel forms include cazuela bowls, beakers and bottles. Late but widespread variety (late prehistoric and protohistoric (Scarry 1985:215).

Fort Walton Incised

Cayson

Characteristic design is a band containing simple incised rectilinear step figures that separate alternating plain and punctation filled zones. Vessel form is carinated bowl (Scarry 1985:215).

Fort Walton Incised

Englewood

Characteristic design formed by narrow, punctation-filled bands that form abstract rectilinear figures including complex interlocking maze forms, diamonds, triangles, zigzag bands or connected chevrons. Found in Rood and Lake Jackson phase context but most common in Safety Harbor (Scarry 1985:219).

Fort Walton Incised

Safety Harbor

Characteristic design is curvilinear equivalent of variety Englewood. Vessel forms include beakers, bottles and bowls. Most common in Safety Harbor area (Scarry 1985:219).

Fort Walton Incised

Sneads

Characteristic design is curvilinear equivalent of variety Cayson (Scarry 1985:219).

......',...
Point Washington Incised

..>> . ,.. > ,>

Subsumes all ceramics from Fort Walton area with sandy paste decorated with medium curvilinear incised lines.

Point Washington Incised Point Washington

Characteristic motif consists of a series of horizontal running or interlocking scrolls. Vessel forms include carinated bowls, cazuela bowls, bottles and beakers. Late variety found in Bull Creek and equivalent phases (Scarry 1985:230).

Point Washington Incised Griffith

Characteristic designs are a variety of representational forms, many of which are part of Southern Cult iconography. Vessel forms include carinated and cazuela bowls. Chronologically late, found in Bull Creek and protohistoric sites (Scarry 1985:230).

185

The Bull Creek Site

A

E

..... '. .' .

. ' , ' ,

.. .. ',: :.':. '.' '"
. " , .', .. . . ; . ' " ,

..' :.'

.

.

F
A, Lamar Bold Incised; B, Fort Walton Incised var. Fort Walton; C, Fort Walton Incised var. Cason; D, Fort Walton Incised var. Englewood; E, Fort Walton Incised var. Safety Harbor; F, Fort Walton Incised var. Sneads; G, Point Washington Incised var. Point Washington; H, Point Washington Incised var, Griffith
Figure 160. Examples of Fort Walton incised pottery types and varieties typical of the Bull Creek phase (taken from Scarry 1985).
186

The original definition of Lamar Bold Incised was based primarily on the WPA excavations in the Macon area, and as a result those definitions reflect largely the variety of pottery found in that area (Kelly 1938:47-48; Jennings and Fairbanks 1939). Willey applied those definitions to the Fort Walton area with only slight modifications:
Definition: A grit-tempered ware decorated with broad incised lines, hollow reed punctations, and rarely, dot punctations. Curvilinear designs (scrolls) usually combined with rectilinear elements, particularly horizontal lines between scrolls. Some rectilinear elements stand alone. A row of hollow-reed punctations is usually placed at base of the incised decoration which is a border around upper portion of the vessel. The forms of cazuela bowls, on which the incised designs occur, are often covered on the base with Lamar-type complicated stamping.
Relationship of type: Is related to Fort Walton Incised. May be ancestral, in part, to Ocmulgee Fields Incised and Aucilla Incised (Willey 1949:493).
Willey defined Fort Walton Incised as follows:
Technique: Lines and punctations incised into soft surface of vessel. Lines are deep, wide, and usually rectangular in cross section. Large round dot or square punctations most common. Hollow-reed punctations sometimes used.
Design: Elements are volutes, interlocked scrolls, running scrolls, circles, trifoil figures, crescentic forms, S-shaped and reverse S-shaped figures, rectilinear stepped figures, pendant loops, and triangles. Elements are usually repeated around vessel in a connected design pattern. Dot punctations used as filler for both backgrounds and for design proper. Incised lines sometimes used as
fillers (Willey 1949:460).
Willey also named several other incised pottery types for the Fort Walton area, including Pinellas Incised and Point Washington Incised, which have at times been applied to the Bull Creek phase and have been incorporated into Scarry's scheme.
Distinguishing Lamar incising from Fort Walton incising is to a great extent con-

Bull Creek Site Material Culture
tingent on sherd size. Large sherds are required to distinguish patterns that may be typical of north Georgia Lamar pottery from those of the lower Chattahoochee Fort Walton pottery. Unfortunately there were very few large incised sherds recovered from Bull Creek.
During our examination of the Bull Creek collections, all incised designs were drawn as encountered. Most were too small to provide any meaningful indication of design. Eventually 79 sherds were determined to be sufficiently large to determine all or part of a design. Several of these were vessels reconstructed by the National Park Service. Because of the wide variability present in the collection, it was necessary in many cases to create general design categories. This was especially true with respect to the zone punctate vessels where designs differed on each sherd. General rectilinear and curvilinear design categories were selected when there was substantial variability. Figure 161 illustrates examples of each distinct design pattern found in the Bull Creek collections. Table 12 describes the criteria used to distinguish these motifs.
In examining the incised motifs found at Bull Creek, one point becomes obvious. Most of the designs are common Fort Walton decorative patterns. There is little indication of typical Lamar designs. Design Category H which consists of nested half circles bordered by parallel lines, is most similar to a Lamar Bold Incised motif. The recovered examples appear to contain two to four lines in the design, which would be most typical of an early Lamar incised style.
Recognition of Design Category L is particularly meaningful because it represents the most common form of Pinellas Incised. As previously noted, Schnell now considers this style of incising to be typical of Stewart phase but not Bull Creek phase. At the Bull Creek site, this single sherd was found in a probable Stewart phase feature.
187

The Bull Creek Site
A

'.. .-z;:.....~

"'

~

B

.8 D
c

u,

...-,--,y,-r
e.::. f~c: ~

I'

: C: ..... FC..

" ,.., .

&.,.~Cl:".o.

F

G

.. r.

~

.".

" r:
"e



~

IF-

. ~~~= "" . J

1 & ' / '/~ ~'.~~1)

L

N

0)'\)l;l~\: ('(/

o

.~

. "'...

p

Figure 161. Incised designs found at Bull Creek (letters correspond to Table 12).

188

Bull Creek Site Material Culture

Table 12. Incised design categories identified from the Bull Creek site collections.

1c::~t~g8l"yl>

..... ...... I)~S8ripti9fI>

I >1 Count

A

Band of interlocking scrolls (guilloche) with dot (punctate) filler (Fort Walton Incised). 9*

B

Simple curvilinear stepped band alternating with punctate band.

2

C

Parallel undulating bands with alternating punctate fillers.

1

D

Simple continuous undulating band with punctate fillers.

1

E

Complex designs of punctate filled oval and rectilinear elements alternating with clear 17*

zones or elements.

F

Rectilinear step elements with alternating punctate-filled and clear zones.

24*

G

Concentric half-circles with a punctate background

1

H

Concentric half-circles bordered by parallel lines.

3*

I

Concentric circles

1

J

Band of interlocking scrolls without punctate filler (Point Washington) .

5*

K

Running or interlocking open scroll.

3*

L

Running Scroll (Pinellas Incised)

1*

M

Simple looping pattern.

1*

N

Multiple parallel lines.

3

0

Simple half circle single line loops at rim.

2

P

Simple discontinuous rectilinear and curvilinear incised elements below rim.

4*

Q

Hand-eye motif

1

*One or more examples recovered from the probable Stewart phase feature excavated in 1981.

Negative Painted Wares relate to painted water bottles at Bull Creek. There are a few references to fragments of painted pottery found in the general midden of the cemetery area excavations (Lester 1937:54), but these sherds were not relocated in our examination of the collections. Schnell does identify one sherd from Kelly's collection.

In 1979 the Bull Creek negative painted pottery was given the type name Nashville Negative Painted variety Columbus and was considered a local copy of similar vessels from the northwest (Williams 1979). More recently Scarry gave the Bull Creek pots a new type status as Columbus Negative Painted variety Columbus (Scarry 1985:213).

The dog pots from Bull Creek included two varieties of painting. The two vessels from Burials 3 and 7 exhibited red spiral designs on a buff background. The third exhibited a black pattern on a reddish background. A third vessel is most similar to a dog pot recovered from Neisler Mound (see Figure 156). All three designs on the Bull Creek dog pots are quite similar running scroll patterns superficially comparable to the rarely encountered Pinellas Incised vessel design category L (Figure 161-L).

Scarry defined Columbus Negative Painted to include all sand-tempered vessels with indirect (negative) painting in the Fort Walton area, as a distinct type from the shelltempered Nashville Negative Painted wares. Scarry noted that the ware has thus far been found only in Rood and Bull Creek phase contexts. Variety Clay was recommended for the negative painted vessels from Cemochechobee, which differ in vessel form and painted design.

189

The Bull Creek Site

Schnell has noted on several occasions that only four examples of the Bull Creek varieties of negative painted dog pots are known to exist (Schnell 1990:69). This conclusion has been reaffirmed through communications of both Schnell and the senior author with individuals knowledgeable of the antiquities market in the region. The recovery of three-fourths of the known examples of this vessel form from a single site, Bull Creek, does represent a unique occurrence.

Bull Creek/Go-Kart site locality also produced other Creek pottery types such as Kasita Red Filmed and Ocmulgee Fields Incised. Excavation at the Go-Kart site produced substantial amounts of earlier Averett phase ceramics. These ceramics are discussed in detail in a separate report of the Go-Kart site excavations (Ledbetter 1995c) and will not be repeated here.

Other Pottery Types found in the Bull Creek collections include anomalous forms of decoration and a few sherds from other occupations. Several sherds with cordage impressions are listed on the National Park Service summary sheets and examples were relocated in those collections (Figure 162 upper). These sherds may be from non-Bull Creek phase occupations, but they may also represent impressions which commonly occur on vessel bottoms.

Cord Marked

The collections also include a single fiber tempered sherd from Kelly's collection (Figure 162, center) and several Chattahoochee Brushed sherds (Figure 162, lower). Excavations conducted by SAS of the Bull Creek site and the contiguous GoKart site indicate a widespread occurrence of Archaic and historic Creek material across these site areas. The fiber tempered sherd reflects the Late Archaic occupation and the Chattahoochee Brushed sherds reflect the historic Creek occupation.

Fiber Tempered

o

3

E-5*e

em

It should be noted that most of the Creek occupation occurred beyond the boundaries of most previous excavations, as is reflected in the extremely small amount of ceramics recovered from these collections. SAS excavations conducted in the larger

Chattahoochee Brushed
Figure 162. Examples of other pottery types from Bull Creek (Full Size).

190

Other Ceramic Objects

Bull Creek Site Material Culture

Examples of ceramic pipes and pottery discs have also been recovered from Bull Creek, but were quite rare in the collections. The low counts of pipes appear to be an accurate reflection of the scarcity of the artifact type on the site, but there is some contradiction with respect to pottery discs. Chase noted "pipes are infrequently found at Bull Creek, only three stems and one bowl fragment came to light but hundreds of pottery disks and one of stone appeared. This is not surprising since the disks prevailed from earliest Lamar times into historic period" (Chase 1957).

Our reexamination of the collections produced two complete pipes from the cemetery excavations and four fragments from the village midden. Of this total, two were recovered from the pit excavated by Fuller in 1981. We were able to record only six pottery discs from the collections. Of this total four were found in the Fuller collection. With respect to ceramic discs, we are left with the possibilities of an error in Chase's assessment of numbers or the actuality that discs were commonly found only in the portions of the site examined by Chase and Fuller.

The most complete pipes and pipe fragments found in the surviving collections from Bull Creek are illustrated in Figure 163. The upper two examples are the two effigy forms recovered from the cemetery (Burials 17 and 18). The lower examples are more simply fashioned examples from the midden.

Pottery discs from Bull Creek ranged in diameter from 30 to 45 mm. Surface treatment was divided between plain and complicated stamping (Figure 163)

One clay bead was reported from the cemetery excavations (Find 49). The artifact was not relocated by the author.

Figure 163. Examples of ceramic pipes and pottery discs from Bull Creek.

191

The Bull Creek Site
Shell and Bone Objects

Chase noted that objects of shell were not recovered from the village midden (Chase 1957). However, beads made from marine shell were relatively common in burial context. Shell beads were found with Burials 3, 18, and 20. A conch shell core was also found with Burial 18. The NPS list of special finds includes one bone awl (Find 74) from the cemetery area.
Stone Objects

Numerous objects of chipped and

ground stone occur in various collections from

Bull Creek, but there are no tabulations. The

Celt

majority of these belong to occupations other

than the Bull Creek phase. For this reason

only a general characterization, with a focus

on the Bull Creek phase, will be presented.

Groundstone includes many objects, such as generalized grinding implements which are shared by many cultures. Objects that are clearly associated with the Bull Creek phase are primarily burial objects. Included are greenstone celts and chisels and stone discoidals or chunky stones (Figure 164). Examples from the cemetery include celts and chisels from Burials 5, 6, 9, and a discoidal from Burial 7. Fragmentary celts and ground pebbles of undefined function were also recovered from midden contexts but no accurate counts are available.
Ground pigment was also recovered from several burials (Burials 6, 28, and 29). The descriptions from the WPA notes appear to indicate red ochre or hematite.

Chisel

Ground Pebble

Little is known concerning other types of ground stone associated with the Bull Creek phase simply because few examples survive in the collections. Listed "Finds" consist primarily of crude chunky stones. There is no available data for large grinding implements comparable to a metate or grinding slab.

Discoidal

o

3cm

15!5!==~

Figure 164. Examples of other ground stone objects from Bull Creek.

192

Chipped Stone Artifacts were substantially less common than pottery at Bull Creek, but apparently did occur in significant numbers. This is reflected primarily in the NPS list of finds which records a number of projectile points. As previously noted, these projectile points date primarily between the

Bull Creek Site Material Culture
Paleoindian and Late Archaic periods. Most are made from quartz and Coastal Plain varieties of chert. Figure 165 illustrates selected examples of projectile points from the WPA collections. The collections also include small amounts of lithic debris and expedient tools which probably date to these earlier occupations.

o,.......--

3

em

Figure 165. Examples of projectile points found on the Bull Creek site which pre-date the Bull Creek phase occupation.

193

The Bull Creek Site
The only chipped stone artifact that can definitely be associated with the Bull Creek phase occupation is a large biface recovered from Burial 6. The large blade is made from a dark chert most similar to material found to the north in the Ridge and Valley province. The biface measures 83 mm in length.
Modern Artifacts
A small number of historic artifacts have also been reported from the Bull Creek site. Some objects, such as a metal thimble mentioned in the NPS notes, may date to the historic Creek occupation. Most artifacts recovered from the site are modern and relate to modern land use. These objects will not be discussed.
In summary, the majority of artifacts associated with the Bull Creek phase are ceramic objects. The collections of ceramics are dominated by vessels of medium to large size. There is, however, a full range of vessel sizes and a moderate range of vessel forms. The remaining ceramics consist primarily of pottery discs and smoking pipes. Ground stone objects are represented by a few well-made ground stone artifacts such as celts, chisels and discoidals. Chipped stone production and tool use appears to have been insignificant during the Bull Creek phase. Little is known concerning objects made from organic materials due to poor preservation. There does appear to have been a significant use of marine shell, which survives primarily in burial context.

o
P"5

3cm i

Figure 166. Chert biface associated with the Bull Creek phase occupation.

194

Chapter 9 Summary and Discussion

Our examination of the records and collections from previous investigations of the Bull Creek site disclosed a wealth of information. While records of most investigations were incomplete, surviving notes, maps, collections, and quite often unpublished manuscripts did allow all past excavations to be documented. The curation of notes, photographs, manuscripts and collections relating to the 1930s excavations of Frank Lester and Isabel Patterson was accomplished through the efforts of numerous personnel of the National Park Service. Preservation of material relating to the other excavations has been the responsibility of personnel of the Columbus Museum. Frank Schnell, Jr., in particular, has diligently preserved records of other excavations at Bull Creek which otherwise could not be documented. The Columbus Museum's files of Isabel Patterson, which includes correspondence and manuscripts, proved to be an extremely valuable source of information. The importance of the Columbus Museum as a curation facility cannot be overstated with respect to the completion of this report.
The primary goals of this project were the publication of the findings of previous investigations and a synthesis of characteristics of the material culture that defines the Bull Creek phase. This has been accomplished to varying degrees of success in the preceding chapters. Our success in researching previous investigations often depended upon the fortunate preservation of a few notes, letters, draft manuscripts and newspaper articles. Figure 167 shows the locations of major excavations outlined in the following paragraphs.
The National Park Service files contain substantial amounts of information and thus made a major contribution to this report. Shortcomings of the NPS records include missing field maps, particularly relating to the

cemetery, a lack of an osteological report, and the absence of a synthetic or interpretive report. Lester's field notes appear to be complete, but his original hand-written notes have not been located. We do not know how much information, if any, is missing in the notes transcribed by NPS personnel. With respect to an interpretive report, the lack of such a final report by A.R. Kelly was the primary motivation for the present project.
The excavation project conducted by the University of Georgia in 1950, under the direction of A.R. Kelly, was extremely important with respect to interpretation of the site, and records of the project survive only because of the public's interest in archeology. Aside from the collections, no records of the excavations survive at the University of Georgia. However, photographs and a brief report were produced by the Columbus Chamber of Commerce and these records have been preserved in the files of the Columbus Museum.
Records of more recent research features the work of Frank Schnell, Jr., who excavated on the site in 1959, and as a graduate student at the University of Georgia, undertook the site as a thesis topic. Schnell did not complete a Bull Creek synthesis due to a change in thesis topic; however, he did accumulate a substantial amount of data, particularly with respect to ceramics. Fortunately much of Schnell's data have been preserved and major portions of that research have been reproduced in the ceramic synthesis of this volume.
Records of other investigators, such as Frank Mulvihill, Margaret Ashley and Frank Schnell, Sr., Frank and Barbara Fenenga, David Chase, Jack Tyler and Terry Jackson, and Jackie Fuller, are found primarily in rare publications and unpublished manuscripts preserved in the files of Columbus Museum.
195

The Bull Creek Site GEORGIA

/

\

A ..

r:

ALABAMA

Kelly (1950) Unit4 and Schnell ----+-~.

(1959) Unit 2

/ - - ",-~

WPA Cemetery Block------lr--4WI\\ and Kelly (1950) Unit 3

N

i

o

300

m

Figure 167. Locations of major archeological excavations at Bull Creek. 196

Accounts of work conducted on the contiguous Victory Drive site were found in a manuscript prepared by Schnell (1970). Those excavations relate primarily to an historic Creek occupation. Site boundaries created by Schnell for the Victory Drive site represent the extent of the Creek occupation. As a consequence of defining a distinct Victory Drive site, the modern site boundaries of Bull Creek differ from the boundaries of the old Bull Creek Village described in the writings of Lester (1938) and Patterson (1936).
Isabel Patterson's major work on the Bull Creek site is a 1950 article in Early Georgia, but a substantial body of related writings are housed in the files of the Columbus Museum. Included are letters to A.R. Kelly and others, drafts of several manuscripts, and several newspaper articles authored by Patterson. Additional records are found in various cultural resource management reports, the most current and site-specific of which was completed as a result of the Columbus Riverwalk project (Ledbetter 1994b).
Our examination of the material culture of the Bull Creek site focused on ceramics. Relatively few artifacts of other types have been collected from the site and little research has been conducted upon material such as lithics and subsistence remains. In the case of lithics, we are dealing with a Late Mississippian component at Bull Creek that made little use of chipped stone. Like contemporary Lamar populations in the Piedmont, it is apparent that small, triangular, chipped stone projectile points were not being produced at this time. While small chipped stone points were commonly found in association with the Early Mississippian Averett component on the contiguous Victory Drive site, there are no confirmed records of the recovery of any of these points in the excavated portions of the Bull Creek site. The only confirmed chipped stone artifact recovered was a large chert biface found in Burial 6 (see Figure 166). A few groundstone celts and chisels were also found

Summary and Discussion
in burial context and broken examples have been found throughout the midden deposits. Few other stone artifacts exist in association with the Bull Creek phase occupation. Small amounts of pigment stone have been recovered, again from burials. The site has produced a number of marine shell artifacts, primarily beads. The majority of shell artifacts have been found in burials. Past excavations recovered sparse quantities of subsistence remains. In the older excavation projects, bone and charcoal were not commonly saved. In the most recent excavations conducted by SAS, no large pits with the potential for preserved organic material were located. Today, subsistence studies remain a primary deficiency in all investigations of the Bull Creek site.
The study of Bull Creek ceramics has been a major concern of this report. An effort has been made to integrate the ceramic studies resulting from earlier investigations with our own analysis, which focused upon vessel form. Problems in this undertaking resulted from variability in the quality of the original data from different sources, incompatibility of analysis categories and terminology, and the loss of some collections, which makes replication of the original work impossible. Also, differences in the frequencies of certain pottery traits, particularly surface treatment, between total sherd counts and minimum number of vessel counts made comparisons difficult. Nevertheless, we were successful in characterizing the Bull Creek pottery assemblage using these sources of data. This study was not exhaustive; however, much of the primary data relating to vessel characteristics have been presented as an appendix and hopefully will be examined further by other researchers. A primary result of our pottery study is the conclusion that the composition of the ceramic assemblage found on the Bull Creek site is consistent with interpretations of Schnell and others and supports the need to distinguish an early (Bull Creek) and late (Stewart) phase (Schnell 1990:67; Schnell and Wright 1993:21).

197

The Bull Creek Site
Our examination confirmed Schnell's interpretation that the Bull Creek site pottery was sufficiently different from sites such as 9Sw51 (Broyles 1962) and the Park's Mound site (Hally and Oertel 1977) to be defined as a distinct phase. At the Bull Creek site, the higher frequency of complicated stamping and the near absence of late Lamar incised and Pinellas Incised motifs does appear to indicate an earlier occupation. Unfortunately, we do not have reliable radiocarbon dates to further verify this interpretation.
Other research issues may also be pursued using the accumulated information. Issues relating to the Bull Creek village, the Bull Creek cemetery, as well as the status of the Bull Creek site within the greater Mississippian community can be addressed.
A greater understanding of the extent and architectural layout of the Bull Creek village can be gained by examining the findings of past research conducted at the site. Unfortunately, the loss of much of the site area to development prior to any thorough archeological survey means that major aspects of this issue cannot be resolved. We have good locational information for the excavated structures, and the locations of other possible structures may be inferred by the recovery of house daub and the discovery of domestic features, such as large trash-filled pits.
Several questions can be addressed with further examination of the Bull Creek cemetery by incorporating the findings of the Lester's 1930s and Kelly's 1950 excavations with modem osteological examination of the human remains. A very important question that remains unanswered is whether the burials represent an actual cemetery laid out separately from the village or whether the burials were within the floors of domestic structures.
Another question relating to the burials is the simple issue of sex and age composition. This question is addressed to some degree with
198

new information from an examination of the surviving remains by Chad Braley. Using Mr. Braley's 1994 observations to supplement the field observations of Frank Lester, a greater level of interpretation is possible.
Finally, we can examine the site from a larger regional perspective. This has been accomplished with the assistance of Adam King, who is currently involved in comparable research relating to the Etowah Mound group in northwest Georgia. The paragraphs relating to this topic have been prepared primarily by Mr. King. The basic research question to be addressed concerns the evolution of the Bull Creek site and the Bull Creek phase as the result of the movement of people or the adoption of basic concepts resulting from interaction of local populations with other groups.
The Bull Creek Village
The most recent investigations conducted on the Bull Creek site confirmed that most of the site has been destroyed (Ledbetter 1994b). Today, only a small strip of the site survives along the wooded river bluff edge above the Chattahoochee River. Data recovery investigations conducted on the contiguous Victory Drive site indicates that a series of structures associated with the Bull Creek Village extended along the river for a minimum of 450 m from the mouth of Bull Creek (Ledbetter 1995).
Based upon the combined work of previous investigations and the recent survey (Ledbetter 1994b), it appears the Bull Creek Village extended for a substantial distance along Weracoba Creek. Minimally, the village extended along the creek as far as the eastern boundary of the Fenengas' (1945) Weracoba Creek site (9Me379). This would put the eastern boundary of the site across Victory Drive and a straight-line distance greater than 500 m from the Chattahoochee River (Figure 168).

GEORGIA

Summary and Discussion

/

\ ;--...

~ ...

."

ALABAMA
BULLCREEK CEMETERY

N

i

o

300

e

m

Figure 168. Map showing estimated extent of the Bull Creek Village.

199

The Bull Creek Site
The aerial extent of the Bull Creek Village may be only partially reconstructed using the results of previous archeological investigations. Because most investigations were conducted in the richest and best preserved areas of the site, most of our knowledge of the site comes from the area nearest the Chattahoochee River and Weracoba Creek. There is little information for the interior portion of the site. There is also no means of accurately determining the eastern limits of the site. For that reason, we cannot determine whether the village existed as a planned "town," with structures encircling an open plaza, or whether the site consisted of a series of structures strung along the two waterways.
Past investigations have produced no information on the presence of ceremonial or sacred structures and relatively little information on domestic structures. There are no confirmed accounts of the presence of a mound on the site. Information concerning domestic structures comes primarily from the WPA and SAS investigations.
Lester and Patterson found portions of two domestic structures. The structures were substantial post-construction houses plastered with daub. The structures were basically square but appeared to have had rounded corners. The only mapped structure (Number 1) appears to have been approximately 6 m wide. A large pit, probably excavated to procure clay for house daub, was located adjacent to the structure. A similarly shaped structure was encountered during data recovery of the Victory Drive site (Ledbetter 1995). That structure measured approximately 7 by 7 m.
While specifics of the village layout cannot be determined, there is ample information to conclude that a substantial number of structures were present on the site. A review of all excavations produced consistent evidence of structures in all areas of the site investigated. In past excavations conducted along Weracoba Creek and in the more interior por-
200

tions of the site, house daub and pits were consistently found. Based primarily upon the work of Chase, Schnell, and Fuller, it is apparent that structures were present along Weracoba Creek and extended into the interior of the site for a minimum distance of 150 m from the waterways.
One of many questions that cannot be answered with available information is the length of occupation. Substantial amounts of midden were deposited across a large site area. The beginnings of this accumulation equate to the Bull Creek phase time period. Apparently, the site continued to be occupied, or was reoccupied, during the subsequent Stewart phase. At present, we are unable to judge the intensity or extent of the Stewart phase settlement.
The Bull Creek Cemetery
The Bull Creek Cemetery was excavated in 1936 and 1937 under the direction of Frank Lester (Lester 1938). Lester exposed 46 burial pits containing the remains of at least 50 individuals in a block excavation that covered approximately 400 m2 (Figure 169). While a sizable block was excavated, there is no evidence that the entire cemetery area was exposed. Today, prospects linger that additional burials are still preserved in the unexcavated peninsula of land that lies above the confluence of Bull Creek and the Chattahoochee River.
Much of Lester's preliminary report focused upon the cemetery excavation and description of individual burials (Lester 1938). Lester's account has been presented in Chapter 3 and additional field note information may be found in Appendix A of this report. Table 13 is a summary of Lester's burial data with new observations based upon the osteological reexamination of the remains by Braley. Observations concerning burial orientations and layout of the cemetery are also presented in this chapter.

Summary and Discussion

4-27 4-28

Find 72

4-25

4-26
B18
I( .,
.'0

~ B19 IJ

B44 0'-'?J).B46
[J
B45
B43

IY'\'
B39
D B42
{}

B5

llf.'\ \,

~

0~)/,,>B7

. 4-21 4-22
'V.~ B14

B36 B40

~ ,- \wl>~_B37 B35

"
-.(9qo;
B28
B16
..,.
~
B26 <Xl

B25
B17
B30o:,_l,'0.<1>'" "~'_ B31

BI3 t.~ B22

B3
~M""'

4-11
Sl
eI B12

4-12
<'~B2

II "I

t;J,

B24

~q B32

., B34

~ B33

0'

4-4

.ot..v...

L33 L31 L29 L27 L25 L23 L21 Ll9 Ll7 Ll5 Ll3 Lli

L9

L7

L5

L3

Ll

4-1

4-2

RI

R3

Figure 169. Plan map of the Bull Creek Cemetery.

The Bull Creek Site

Table 13. Summary of Bull Creek Cemetery

BlinCll.
Depth" ..2 .,

.,;~

1><

>

::-:-:.:':.:-".:..-:-:.

1

unknown

NAINA

skull only

adult male, skull complete

none

2

contracted flexed 67"fNA

skull elongated

adult, sex and age unknown

none

3

contracted flexed 67"fNA

some post-mortem disturbance

an elderly adult male, teeth worn down to pegs in front, secondary dentin exposed

effigy pot, shell beads

4

flexed

44"/20"

female

sex and age unknown

none

poorly preserved, skull gone Fairly robust tibia and femurs, sciatic notch is

5

contracted flexed 72 "/NA except for teeth, no cervical less than 90 degrees, suggesting male. Adult, greenstone celt

vertebrae or scapular

little attrition of teeth - no cranial fragments.

(7")

Left tibia has thickened mid-shaft; possibly a 1 greenstone

healed fracture. Fragments of long bones celt (4") and 2

poorly preserved,

(legs), vertebra, tarsal, maxilla and mandible chisels,

6 flexed, feet folded 78"/42" and severely damaged during

(no cranial frags). Male suggested by

worked flint,

back to pelvis

excavation

ruggedness of long bones. Person had lost first black flint

premolar and both molars (lower right) before knife, red

death. Moderate attrition on teeth.

pigment

skull, right fibula, both fragmented femurs and long bone fragments, effigy pot,

7

Contracted flexed 78"/42"

femora, tibiae

no cranial fragments

discoidal

good shape, possible cranial robust tibia and femurs (male?), no cranial

8

contracted flexed 68"/36"

deformation

fragments

none

9

contracted flexed 66"/32"

poorly preserved

adult, sex unknown, moderate tooth wear chisel 3l/z"

10

skull only

54"fNA

*****

missing

none

Adult male (cranial morphology). Complete

skull is rather small and the mandible and maxilla, fragments of skull,

11 folded feet back to 73"/36"

individual was of

femurs and tibia. Skull shows frontal

none

pelvis flexed

comparatively small stature. deformation. Antemortem loss of lower right

This is a female burial.

first and second molars. 2 rom pit cavity in

lower left third molar.

fragments of skull and long adult male-skull cap has blunt supraorbital

12 contracted flexed 69"/35"

bones

margins (consist of four long bones)

none

skull fragments, teeth,

13

undetermined 40"/26"

cervical vertebrae

missing

none

Skull is slightly misshapen,

terrible infection of the leg - massive

right parietal shows patholo- thickening of the right femur, tibia and fibula.

14

(Old Man)

51 "/34"

gical enlargement -

Fibula is almost unrecognizable as such.

none

contracted flexed

depression present at the Probable adult male based on left femur robus-

lambdoid inter-section-old ticity. No cranial bone, but fragments of

male.

pelvis and long bones are present.

Elderly person, sex unknown, skull-vertebra

skull crushed, short

fragments, mandible and maxilla. Bone in

15 contracted flexed 78"/53" fragments humeri, ulna and

good shape but incomplete. There was

none

radius, pelvis, traces of

secondary dentin exposure on most teeth.

femora and tibiae

Front teeth so worn they are peg-like.

16

double burial NA/36 "

*****

missing

effigy pot

17 contracted flexed NA/NA

*****

Skull cap and three other cranial fragments are total collection. An adult- gracile char-
acteristics of skull suggest female. There is slight cranial modification of occipital.

effigy pipe

consists of a few fragments 18 contracted flexed 61 "/39" of the skull and sections of
femora

missing

bowl, shell beads, core, effigy pipe

202

Summary and Discussion

Table 13. Summary of Bull Creek Cemetery

I taL 1/>

-~

> .: BUriiil . :.>.-.>-.:"."::".-:::::

.......

.: pePtli* :.

.......

<:.

==

female, skull badly crushed, Sciatic notch of > 90_ probable female.

contracted more

no cervical vertebrae, rest of Inventory: fragments of cranial, pelvis, femur,

19 than any found yet 67"/45" the body well preservation tibia, ulna, humerus, phalanges, tarsals,

none

- knees on chest

scapula, vertebra, ribs. No teeth, maxilla or

mandible (cranial modification).

male - skull, cervical

An adult female(?) - broad sciatic notch,

flexed-folded feet

vertebrae, femora, both

gracile skull. Lower right first molar lost string of shell

20

to pelvis

NAINA tibiae, clavicles, scapulae, antemortem, attrition of teeth suggests young seed beads at

pelvis

adult. Some cranial modification.

neck

flexed - feet

skull, tibiae, femora, ulna,

21

folded to pelvis 36"/21 "

and radii

missing

none

skull crushed but excellent adult male in very good condition but there is

22

flexed

54"/40"

preservation

no skull - a very robust person

none

23

undetermined 57"/33"

bone fragments

missing

none

24

contracted flexed 59"/28"

skull, long bones

missing

none

25

folded flexed 52"/21 "

skull and long bones

missing

none

26

extended flexed 79"/41"

skull and long bones

missing

none

27

contracted flexed 75"/41 "

skull and long bones

missing

none

portions of left tibiae and

pigment, bone,

28

flexed

80"/35" femur and the skull with all

adult, but age and sex unknown

stone, shell,

of the facial bones missing

bone awl

29 contracted flexed 88"/47"

long bone fragments

missing

red pigment

flexed - with

fragments of skull, tibiae, adult - Inventory: cranial (moderate tooth

30

knees farthest 49"/11 " femora, small section of the wear), mandible, long bone fragments-contains

none

away from body

pelvis

mandible fragments of two people

triple burial - 2

31 contracted flexed, 61"/23" (see description in Chapter 3)

missing

none

3rd is skull only

portions of skull, tibiae, probable male, based on robusticity, inventory rocks, flint

32

contracted flexed 52"/34" femora, humeri, and ulna consists of cranial, mandible, and long bone chips and

fragments

animal bones

33

contracted flexed 36"17" frag. skull, teeth, long bones

bad shape, probable adult

none

34

contracted flexed 37"/16" fragments of skull and tibia

bad shape, adult?

none

skull badly decayed,

Cranial deformation of occipital. Skeletal

35

contracted flexed 67"/27" fragments humerus, ulna,

remains of a child - sciatic notch suggests

none

lumbar vertebrae, pelvis, male. Unfused ephipheses of head of femur,

long bones

greater trochanter, and acetabulum.

36

contracted flexed 60"/35"

poorly preserved

missing

none

Excellent shape, possibly an adult male, but

exceptionally well preserved fairly gracile vault. Most of cranium intact, no

37 contracted flexed- 86"/36"

deformation - strong mandible. Lost lower

none

face downward

right first molar, lower left second premolar

and first and second molars before death.

Adult male. Very broad dental arcade. Chin is

38

double extended NA/36 " male and female (guess). unusual, shaped like a "snow plow". Complete

none

burial

See Chapter 3.

mandible and maxilla. This burial also

contains an extra mandible (right fragment).

203

The Bull Creek Site

Table 13. Summary of Bull Creek Cemetery

II ilii'M== ...r"..

P

:

8pP



':"",

skull, vertebrae, humeri, Female with extremely thin skull (teen-aged)
ulna, radii, metacarpal bones. no cranial deformation. Third molars not

39

contracted flexed NA/42 " Portions of pelvis, tibiae, erupted, little tooth wear. Collection includes

none

femora, fibula, metatarsal mandible of a very old person - all teeth of

bones

second person are missing from canine back.

40

skull fragments 61"/42"

skull fragments

an adult? sex unknown.

none

fragments skull, clavicles,

41

contracted flexed NA/39 "

pelvis, long bones

adult? sex undetermined

none

42

contracted flexed NA/26 "

well preserved

A child, lacks third molars, roots of second

molars still unfused, unfused ephipheses of

none

femur and scapula.

NA/50" skull, clavicles, vertebrae, an adult, possibly male based on robusticity -

43

extended flexed

femora, tibiae, pelvis

very fragmented

none

44

contracted flexed NA/26 "

complete

an adult?

none

45

contracted flexed NA/33 "

poorly preserved

an adult male

none

46

contracted flexed NA/28 " teeth, long bones fragments

missing

none

unknown

Box 219. Burial? - Accession number 3923323. Two skull caps, 1 male, 1 female or child, both with fronto-occipital flattening.

unknown

Burial? Accession number 39-23326. An adult male with cranial flattening.

unknown

Burial? Accession number 39-23324. Contains remains of two people, one adult with heavy tooth attrition, one adolescent with
unfused roots on third molars (they had partially erupted, however).

unknown

Burial? Accession number 39-23329. Adult, possibly female - sciatic notch of about 90 0

*Depth: First reading is depth below surface (in inches), second is depth below occupation zone (Lester's measurements).

Our reexamination of the Bull Creek burials provided a few new insights. While there are a few contradictions between field notes and our reexamination concerning the sex and age of a few individuals, the observations are generally consistent. In this respect, a most noticeable finding is the absence of infants and small children. The cemetery appears to be composed only of adolescents and adults. Our reexamination also provided information for cranial deformation which was not included in Lester's notes.
The loss of a few individuals is apparent. This may represent a combination of extremely poor field preservation and cura-
204

tion problems. We do not know if all bone was saved during Lester's excavations. Consistent with NPS laboratory procedures of the times, the Bull Creek skeletons were carefully labeled and chemically treated to insure preservation. However, curation problems may have occurred as the skeletons, or portions of the skeletons, were examined by various individuals resulting in displacement or even loss of some remains. We know, for instance, that the skulls of certain burials were carefully restored and studied following excavation. This is noted in a letter from A.R. Kelly to Isabel Patterson dated June 25, 1937.

In regard to the 4 skulls Dr. Childs has. I saw him after receiving your letter. He will repair them as soon as he gets around to it. Since he is using his own materials, plaster, modeling materials, and his skill, more as a labor of love, I hesitate to press him. Also, you should know that 4 skulls restored is hardly a seriation on which to base any important conclusions regarding racial type, or for use in making comparisons with other Indians. They are valuable as a start toward a skeletal collection. I wouldn't however regard their study as an essential part of the report. Burial data should
suffice here (Ledbetter 1995b:19).
Obviously, Kelly fully intended to have the burials examined in detail. In a letter to Patterson dated July 9, 1937, Kelly stated his plans.
I think now that Marshall Newman (Chicago and Harvard) is our best bet, following suggestion from Setzler. This is the young man, incidentally whom I recommended to do Physical Anthropology on Bull Creek skeletal material (Ledbetter 1995b:21).

Summary and Discussion
Records of Newman's work has not been found, if indeed, any was ever conducted. If records do exist, they would be invaluable in light of the present status of the burial remains.
Questions still remain concerning the initial designation of the Bull Creek Cemetery as a discrete burial grounds separate from the remainder of the village area. Lester (1938) was not able to demonstrate the presence of structures in the area. Lester did note the existence of a thin clay layer, which might be structurally related, but his comments provide minimal information on the feature (see Figure 69). Kelly's excavations near the cemetery produced ample house daub, suggesting the presence of structures in the cemetery area (Kelly 1950). An examination of the orientations of the burials provides one means of assessing the possibility of a planned cemetery (Figures 170 and 171).

/

/

/
,/

4-25 4-26

//
/ /
,/

NE Cluster

-, -,

/

///'<,

-,

/'
/

--- ------ '>" ........ ". .

. ...... ' ...........

.........

........ ....

'-.

<,

\
"""'l- \ .
\ .\ \ \

." '\\.\

4-21 4-22
-,

SWCluster

,\ \

\

\

/'

\)

\' /

GraveLine

GN
A

~m

~

o

5 FeCI

.

I
"..."..

r/"""\". \
, \ .................................

4-11
\
\.

4-12

" f,O~~

y,
/

w.

t

/

"

/

-,,,

/ '" y '"

-,
\

/ '"

. '" \ \/-':

43 4-4

/

/

/

/

L33 L31 L29 L27 L25 L23 L21 LI9 LI7 LIS LI3 Lit

L9 L7

L5 L3

LI

4-1

4-2

RI

R3

Figure 170. Burial orientations in the Bull Creek Cemetery.

205

The Bull Creek Site
Northeast Cluster of Burials (N = 27)
Grid North

Southwest Cluster of Burials (N = 13)
Grid North

'$"I
Z
N=4

EtoW ------+----- WIOE

N=4

N=4

Nw I

N=7

Figure 171. Contrasting burial orientations of two burial clusters in the Bull Creek Cemetery.

Burial orientations have been previously examined by Schnell (1990:67). Schnell noted variation in burial orientations with evidence of clustering along a SSE-NNW axis (24 percent) and NW-SE axis (22 percent). These compass bearings reflect Lester's grid north grid which lies approximately 45 degrees from magnetic north (see Figure 170). This means that many of the burials are oriented roughly perpendicular to the length of the peninsula and the channel of the Chattahoochee River.
The plotted orientations of the Bull Creek burials may be interpreted as evidence of both a planned cemetery and house floor burials. Some groups shown in Figure 169 appear to conform to patterns commonly associated with house floors where they would lie parallel to the walls. However, many of

the graves are aligned along a single axis, which could reflect a cemetery layout.
An interesting result of the plotting of orientations presented in Figure 170 is the apparent presence of two clusters of burials. These have been labeled as the northeast and southwest clusters on the map. Table 14 lists the range of orientations for each group. Another distinction is that burial goods were found only in the northeastern cluster.
One interpretation of this patterning is that the northeastern cluster represents a discrete cemetery group with most of the bodies generally placed parallel, while the southwest cluster conforms more closely to patterning evident with house floors. A second interpretation would be that temporally distinct burial episodes are reflected.

206

Summary and Discussion

Table 14. Contrast of Burial Orientations from two burial clusters from Bull Creek Cemetery

(based on Lester's grid north notations)

I

Bl1tiaLOtielltatioIl> >1

NortheastCluster

I

$ol.ltlj\yesti@1llster .

I

North to South South to North East to West West to East Northwest to Southeast NNW to SSE Southwest to Northeast SSW to NNE Southeast to Northwest SSE to NNW ESE to WNW Total Count in Sample

Count = 0 Count = 0 Count = 1 (3.7% cluster) Count = 1 (3.7% cluster) Count = 7 (25.9% of cluster) Count = 1 (3.7% of cluster) Count = 2 (7.4% of cluster) Count = 3 (11.1 % of cluster) Count = 6 (22.2% of cluster) Count = 4 (14.8% of cluster) Count = 2 (7.4% of cluster)
27

Count = 1 (7.7% of cluster) Count = 4 (30.8% of cluster) Count = 4 (30.8% of cluster) Count = 4 (30.8% of cluster)
Count = 0 Count = 0 Count = 0 Count = 0 Count = 0 Count = 0
Count + 0
13

The existence of a discrete cemetery area at Bull Creek remains a possibility. Excavation in domestic structures elsewhere on the site have failed to produce burials, suggesting that graves may not have been commonly placed in house floors. Further insight into this intriguing issue would be gained through excavation and comparison of other Bull Creek and Stewart phase sites.
The Bull Creek Phase from a Regional Perspective
Since the first work was done at the site, researchers involved with Bull Creek have noted the unique character the ceramic assemblage found there. While not always explicitly stated, most would certaiIlly agree that this blend of ceramic decorative characteristics resulted as people, pots, and ideas from different geographic areas mixed. In their discussion of the Bull Creek phase in the Oliver Basin, McMichael and Kellar (1960a) interpreted this mixing in the following manner:

At about 1350-1400, several forces begin to impinge upon Rood's focus, which eventually alters it into a Lamaroid expression. Coming from the south, moving up the Chattahoochee River, Fort Walton begins to appear in many sites, especially incised types and Lake Jackson Plain. This force is so strong that a movement of peoples is probably indicated. Secondly, coming from the east and north is the resurgent stamping tradition, complicated and check-stamped, from type Lamar, Savannah, and Wilbanks sources; again a movement of people is indicated and the Rood's Focus is acculturated into the Lamar tradition, but with the addition of Fort Walton-like traits. The culmination of this is probably to be seen at Bull Creek, on the south side of the city of Columbus (McMichael and Kellar 1960:215).
With the benefits of 35 years of archaeological research, it is now possible to see this mixing of ceramic decorative styles in a somewhat different light. McMichael and Kellar originally saw these changes taking place between AD 1350 and 1400. More recently, Schnell's (1986) definition of the Singer phase has pushed the beginning of these changes back, by approximately one century, into the Middle Mississippian period.

207

The Bull Creek Site
According to Schnell, forms and surface treatments commonly found in Savannah and Lamar period assemblages appear in the Lower Chattahoochee as early as AD 1300. During the Middle Mississippian Period (ca AD 1250-1400) a great variety of raw materials and finished goods were exchanged widely across the Southeast. In fact, during this time the entire Southeast may have been more connected, in terms of movement of physical goods, than at any other time in prehistory. One of the most important raw materials exchanged at this time period, as well as many other periods in prehistory, was marine shell. Chemical sourcing of shell found in archaeological contexts has repeatedly shown that the Gulf Coast was one of the major sources of the widely distributed shell.
Through sourcing a variety of nonlocal materials found at the Spiro site, Brown (1983) has been able to identify one of the Middle Mississippian corridors of exchange through which this shell and other materials and goods moved. This corridor began at the Gulf Coast of Florida, extended northward into the Tennessee Valley, and continued toward the Mississippi River and points westward. The Chattahoochee River was undoubtedly one of the major arteries in this corridor, because it offers a water route that connects the Gulf Coast to a point in the upper Piedmont a short overland distance to the major riverways in the Ridge and Valley. It is likely no accident that two of the major Middle Mississippian centers in the Southeast, Etowah and Lake Jackson, are effectively situated at either end of this segment of the corridor.
Along with the shell and other goods, people and ideas also must have traveled up and down the Chattahoochee, and during the Middle Mississippian period the traffic on this route may have been at its heaviest. In fact the connectedness of the two areas is confirmed by the appearance, in Middle Missis-
208

sippian contexts, of Lake Jackson pottery types at Etowah (King 1991) and Wilbanks-like Complicated Stamped sherds in the Lake Jackson area (Scarry 1985). Also, almost identical embossed copper plates were found in Middle Mississippian burials at both Lake Jackson and Etowah (Jones 1982, 1994). In this context, it should not be surprising that sites in the Lower Chattahoochee exhibit a mixture of ceramic traditions derived from the two areas that the river effectively connected.
The mechanisms that brought about this blending of ceramic decorative characteristics need not have involved the movement of people or even pots. A good deal of archaeological and ethnographic research has been done on the use of decorative style as a means of symbolic communication. Essentially, people do not decorate themselves or their possessions without a purpose, and one purpose decoration can serve is as a means of communicating certain messages (Braun and Plog 1982; Plog 1990; Weissner 1983, 1984; Wobst 1977). Depending on the items decorated and the social circumstances, communication might be aimed at the interpersonal, local, or regional level. Overall the messages communicated are relatively simple-In some sense defining an "us" and a "them". In relatively small-scale, kin-based societies like those of the prehistoric Southeast, those who were not your kin or recognized somehow as part of the social group were feared and mistrusted (Hudson 1976). In the case of the Lower Chattahoochee, the adoption of ceramic decorative characteristics may have been an actively pursued strategy whereby the inhabitants of the area acknowledged and displayed their ties to other people, especially from the north, as a way of making exchanges flow more smoothly. Undoubtedly, this was one of a set of strategies that included intermarriage, gift exchanges, etc., aimed at the same goal.

After about AD 1400, Mississippian ceramic assemblages in the Lower Chattahoochee again look somewhat different from preceding phases. While the Singer phase assemblages consist of Rood phase pottery with some Savannah/Lamar types and modes, the Bull Creek phase assemblage is essentially Lamar with some Fort Walton types and modes. In a sense, the ceramics are changed from a southern decorative tradition with some northern influences to a northern decorative tradition with some southern influences. As before, these changes are probably best understood from a regional perspective.
After AD 1400, the Southeast as a region begins to change. The extensive distribution of goods and the importance of elaborate costumes and displays so prevalent in the Middle Mississippian appear to fade. As Anderson (1990, 1994) has suggested, appeals to ideology characteristic of Middle Mississippian chiefly control may have given way to more secular control in the Late Mississippian period. Whatever the structural change, the connectedness of the Middle Mississippian seems to break down or at least regionalize at this time. Reflecting these changes, after AD 1400 the political geography of the interior Southeast seems to shift. During the Middle Mississippian the landscape was dominated by a few large, widely-spaced political centers such as Lake Jackson, Etowah, Moundville, Spiro, Winterville, and Lake George. In contrast, during the Late Mississippian there appear to have been a greater number of more closely-spaced political centers on the landscape, each apparently controlling a smaller area while intensely competing with neighboring polities for regional control (King and Freer 1995).

Summary and Discussion
During this time, shell remained an important raw material used by inhabitants of chiefdoms in the interior Southeast (see Muller 1989; Smith and Smith 1989), so it seems reasonable to assume that the Chattahoochee was still an important avenue for the movement of shell from the Gulf Coast of Florida to points northward. However, unlike during the Middle Mississippian, the political geography of the region was quite different. The two powerful centers at either end of the Chattahoochee that presumably played an important role in the flow of goods through the valley, had lost much of their influence (Hally and Langford 1988; Jones 1994; King 1991). At the same time, those who lived in the Lower Chattahoochee now had immediate neighbors with whom they certainly had contact, probably both hostile and friendly.
After AD 1400, Mississippian chiefdoms of the Lamar ceramic tradition become established in the Lower Tallapoosa, Flint, and Middle Chattahoochee River valleys (see Hally 1993 for a dating of Mississippian mounds in the area). Thus for the first time during the Mississippian period, the inhabitants of the Lower Chattahoochee were surrounded on three sides by people making Lamar ceramic assemblages. Under these circumstances, the stylistic changes that brought about the Bull Creek phase are probably best seen as adjustments made to material culture as the inhabitants of the area, probably out of necessity, became actively involved in more local social and political matters.

209

THIS PAGE LEFT BLANK

REFERENCES CITED
Anderson, D. G. 1990 Stability and Change in Chiefdom-LevelSocieties: An Examination of Mississippian Political Evolution on the South Atlantic Slope. In Lamar Archaeology: Mississippian Chiefdoms in the Deep South, edited by M. Williams and G. Shapiro, pp.187-213.
1994 The Savannah River Chiefdoms: Political Change in the Lat Prehistoric Southeast. University of Alabama Press, Tuscaloosa.
Anderson, D.G., R.J. Ledbetter and L.D. O'Steen 1990 The Paleoindian Occupation of Georgia. Georgia Archaeological Operating Plans 1-6, University of Georgia, Laboratory of Archaeology Series, Athens.
Ashley, M.E. and F. Schnell, Sr. 1928 Mound and Village Sites in Harris, Muscogee and Chattahoochee Counties, Georgia. Manuscript on File, Columbus Museum of Arts and Sciences.
Braley, C.O., L.D. O'Steen, and I.R. Quitmyer 1986 Archaeological Investigations at 9MC41, Harris Neck National Wildlife Refuge, McIntosh County, Georgia. Report submitted to the U.S. Fish and Wildlife Service, Atlanta.
Brannon, P.A. 1930 Coweta and the Lower Creeks. Arrow Points. Vol. 17 (4) pp. 37-47, Alabama Anthropological Society, Montgomery .
Braun, D. and S. Plog 1982 Evolution of "Tribal" Social Networks: Theory and Prehistoric North American Evidence. American Antiquity 47:504-524
Brown, J.A. 1983 Spiro Exchange Connections Revealed by Sources of Imported Raw Materials. In Southeastern Natives and Their Pasts: A Collection of Papers Honoring Dr. Robert E. Bell, edited by D.G. Wyckoff and J.L. Holman, pp. 129-162. Oklahoma Archaeological Survey, Studies in Oklahoma's Past, No. 11.
Broyles, B.J. 1962 A Lamar Period Site in Southwest Georgia, 9CLA51. In: Survey ofArchaeological Sites in Clay and Quitman Counties, Georgia: 9CLA2, 9CLA7, 9CLA28, 9CLA38, 9CLA51, 9QU25. University of Georgia Laboratory of Archaeology Series Report No.5. University of Georgia, Athens.
Caldwell, J.R. 1955 Investigations at Rood's Landing, Stewart County, Georgia. Early Georgia 2(1):22-49.
Chase, D.W. 1957 The Bull Creek Site, A Type Site Station ofthe Middle Lamar Culture. Unpublished manuscript on file, Columbus Museum, Columbus.
1959a The Averett Culture. Coweta Memorial Association Papers.
1959b The Baird Site. Coweta Memorial Association Papers No.3.
1963 A Reappraisal of the Averett Culture. Journal of Alabama Archaeology 9:49-61.
211

DeJarnette, D.L. 1975 Archaeological Salvage in the Walter F. George Basin of the Chattahoochee River in Alabama. University of Alabama Press, University of Alabama.
Delcourt, P.A. and H.R. Delcourt 1985 Quaternary Palynology and Vegetational History of the Southeastern United States. In Pollen Records of the Late-Quaternary Nortn American Sediments, edited by V.M. Bryant and R.G. Holloway, pp. 1-37. American Association of Stratigraphic Palynologists Foundation.
Elliott, D.T., J.L. Holland, P. Thomaston, and M. Emrick 1995 Historic Preservation Plan for the Cultural Resources on U.S. Army Installations at Fort Benning, Chattahoochee and Muscogee Counties, Georgia, and Russell County, Alabama. Garrow and Associates, Atlanta.
Fairbanks, C.H. 1940 Indoor Archeology, in Prehistoric Culturesofthe Southeast, National Park Service Popular Study Series History No.4. pp 27-32. United States Department of Interior, Government Printing Office, Washington.
Fenenga, F. and B.W. Fenenga 1945 An Archaeological Survey ofthe Vicinity of Columbus, Georgia. Manuscript on file, Columbus Museum of Arts and Sciences, Columbus.
Fuller, J. 1995 Bull Creek Site. Manuscript on file, Columbus Museum.
Hally, D.J. 1986 The Identification of Vessel Function: A Case Study from Northwest Georgia. American Antiquity 51:267-295.
1993 The Territorial Size of Mississippian Chiefdoms. In Archaeology ofEastern Nortn America, Papers in Honor of Stephen Williams, ed. by J. B. Stoltman, pp. 143-168. Archaeological Report No. 25, Mississippi Department of Archives and History, Jackson.
1994 Ocmulgee Archaeology 1936-1986. University of Georgia Press, Athens and London.
Hally, D. J., and J. B. Langford, Jr. 1988 Mississippi Period Archaeology of the Georgia Valley and Ridge Province. University of Georgia, Laboratory of Archaeology Series, Report 25, Athens.
Hally, D.G. and L. Oertel 1977 Archaeological Investigations at the Park Mound (9TP4I) Troup County, Georgia, 1972 Season. Submitted to Heritage Conservation and Recreation Service, National Park Service, Contract No. 14-10-9-900395. University of Georgia, Athens.
Hawkins, B. 1916 Letters of Benjamin Hawkins. Georgia Historical Society Collections, Volume 9, Savannah.
Hodler, T.W. and H.A. Schretter 1986 The Atlas of Georgia. The Institute of Community and Area Development, The University of Georgia, Athens.
Hudson, C. M. 1976 The Southeastern Indians. University of Tennessee Press, Knoxville.
212

Hurt, W.R. 1975 The Preliminary Archaeological Survey of the Chattahoochee Valley Area in Alabama. In Archaeological Salvage inthe Walter F. George Basinofthe Chattahoochee RiverinAlabama, edited by David Dejarnette, Alabama Press, Tuscaloosa.
Huscher, H.A. 1960 Appraisal of Archeological Resources of the Walter F. George Reservoir Area Chattahoochee River, Alabama and Georgia. River Basin Survey Report, Smithsonian Institution.
Jenkins, N.J. 1978 Prehistoric Chronology of the Lower Chattahoochee Valley. Journal ofAlabama Archaeology (24)73-93.
Jennings, J.D. and C.H. Fairbanks 1939 Newsletter of the Southeastern Archaeological Conference, Volume 1 (2). Lexington, Kentucky.
Johnson, J.H. 1983 Soil Survey ofMuscogee County, Georgia. United States Department of Agriculture, Soil Conservation Service.
Jones, c. B.
1982 Southern Cult Manifestations at the Lake Jackson Site, Leon County, Florida: Salvage Excavations of Mound 3. Midcontinental Journal of Archaeology 7:3-44.
1994 The Lake Jackson Mound Complex (8LEl): Stability and Change in Fort Walton Culture. Florida Anthropologist 47(2): 120-146.
Kelly, A.R. 1937 Lamar and Related Site Exploration in Georgia. Paper presented at the second annual meeting of the Society for American Archaeology, New Haven, Connecticut. Manuscript on File (Ace. 125), National Park Service, Tallahassee.
1938 A Preliminary ReportonArcheological Explorations at Macon, Georgia. Smithsonian Institution Bureau of American Ethnology, Bulletin 119, Anthropological Papers, No. 1. Washington.
1950 Report on the Archaeological Possibilities of the Bull Creek Site, Columbus, Georgia, with Recommendations for Development. Unpublished manuscript on file, Columbus Museum.
King, A. 1991 Excavations at MoundB, Etowah: 1954-1958. Master's thesis, Department of Anthropology, University of Georgia, Athens.
King, A., and J. A. Freer 1995 The Mississippian Southeast: A World Systems Perspective. In Native American Interactions: Multiscalar Analysesand Interpretations in theEastern Woodlands, edited by M. Nassaney and K. Sassaman, pp. 338-362. University of Tennessee Press, Knoxville, in press.
Knight, V.J. 1994 Ocmulgee Fields Culture and the Historical Development of Creek Ceramics. In Ocmulgee Archaeology 1936-1986, edited by D.J. Hally, 181-189. University of Georgia Press, Athens.
Knight, V.J. and T.S. Mistovich 1984 Walter F. George Lake: Archaeological Survey of Fee Owned Lands, Alabamaand Georgia. Report of Investigations 42. Office of Archaeological Research, the University of Alabama, Moundville.
213

Ledbetter, R.J. 1994a Archeological Survey of the South Commons. Report prepared for the City of Columbus by Southeastern Archeological Services, Inc., Athens.
1994b Archeological Survey of Riverwalk South Rotary Park to Fort Benning, Georgia. Report prepared for the City of Columbus by Southeastern Archeological Services, Inc., Athens.
1995a A Summary of Radiocarbon Dates for the Averett Phase. The Profile.
1995b Dear Isabel, Archeological Correspondence of A.R. Kelly and Isabel Patterson, 1934-1953. Lamar Institute Publication 33, Lamar Institute.
1995c The Victory Drive Site, Data Recovery of 9Me50, Muscogee County, Georgia. Draft report prepared for the city of Columbus. Southeastern Archeological Services, Inc.
Lester, F. 1936 Letter to A.R. Kelly dated March 15, 1936.
1938 Preliminary Report of Excavation at Bull Creek Village Site, Muscogee County, Georgia. Unpublished manuscript on file Columbus Museum.
Lupold, J.S. and F.T. Schnell, Jr. 1991 Historic Resources at the Falls of the Chattahoochee. Prepared for Uptown Columbus. Columbus.
Martin, J.H. 1874 Columbus, Georgia, from its Selections as a "Trading Town" in 1827 to its Partial Destruction by Wilson's Raid in 1865. John Gilbert, Columbus.
McMichael, E.V. and J.H. Kellar 1960 Archaeological Salvage in the Oliver Basin. University of Georgia Laboratory of Archaeology Series Report No.2, Laboratory of Archaeology, Department of Sociology and Anthropology, University of Georgia, Athens.
Muller, J. D. 1989 The Southern Cult. In The Southeastern Ceremonial Complex: Artifacts and Analysis, edited by P. Galloway, pp. 11-26. University of Nebraska, Lincoln.
Mulvihill, F.J. 1925 Some Indications of Indian Occupancy Along Bull Creek, Near Columbus, Georgia. Arrow Points Volume 10 (3) pp 36-41.
Patterson, I. G. n.d. Notes on the Exploration of the Bull Creek Site, Columbus, Georgia. Unpublished manuscript on file, Columbus Museum of Arts and Sciences, Columbus, Georgia.
1936 Archeological Survey of the Chattahoochee Valley. Paper presented at the 2nd annual meeting of the Society for American Archaeology, Washington, D.C. Manuscript on file, Columbus Museum of Arts and Sciences, Columbus, Georgia.
1937 Bull Creek Village. Unpublished manuscript on file, Columbus Museum of Arts and Sciences, Columbus.
1950 Notes on the Exploration of the Bull Creek Site, Columbus, Georgia. Early Georgia Volume 1 No. 1, pp. 34-40.
214

Plog, S. 1990 Sociopolitical Implications of Stylistic Variation in the American Southwest. In The Uses of Style in Archaeology, edited by M. Conkey and C. Hastorf, pp. 61-72. Cambridge University Press, Cambridge.
Scarry, J.F. 1985 A Proposed Typology for the Mississippian Period Ceramics of the Fort Walton Area: A Type-Variety Approach. Florida Anthropologist 38:199-233.
Schnell, F.T., Jr. 1963 The Bull Creek Site in Western Georgia. Manuscript on file, Columbus Museum.
1970 The Victory Drive Site: An Historic Indian Settlement in the Chattahoochee Valley. Manuscript on file, Columbus Museum of Arts and Science, Columbus.
1973 A Preliminary Assessment of Archeological Resources Remaining in the Walter F. George Lake Area. Manuscript 108 on File, University of Georgia Laboratory of Archaeology.
1979 A Cultural Resources Assessment of a Proposed Park on Bull Creek, Columbus, Georgia. UGA Laboratory of Archaeology Report No. 152.
1985 SGA presentation talk on Bull Creek Phase (May 11, 1985)
1986 Phase Characteristics for the Middle Chattahoochee River. Paper presented at the 1986 Lamar Conference, Macon, Georiga.
1990 Phase Characteristics for the Middle Chattahoochee River, in Lamar Archaeology, Mississippian Chiefdoms in the Deep South, edited by Mark Williams and Gary Shapiro, pp. 67-69, University of Alabama Press, Tuscaloosa and London.
1991 A Cultural Resource Assessment ofthe Proposed Site ofthe River Club, Columbus, Georgia. Manuscript on File, Columbus Museum. Prepared for Jordon, Jones and Goulding, Inc.
Schnell, F.T. Jr., V.J. Knight and G.S. Schnell 1981 Cemochechobee: Archaeology ofa Mississippian Ceremonial Center on the Chattahoochee River. Ripley P. Bullen Monograph in Anthropology and History 3. University of Florida, Gainesville.
Schnell, F.T., Jr., and N.O. Wright, Jr. 1993 Mississippi Period Archaeology of the Georgia Coastal Plain. Georgia Archaeological Research Series Design Paper 3. University of Georgia Laboratory of Archaeology Series Report Number 26, Athens.
Sears, W.H. 1951 Excavations at Kolomoki, Season I. University of Georgia, Series in Anthropology 2. Athens.
1956 Excavations at Kolomoki: Final Report, University of Georgia Series in Anthropology, No.5, Athens.
Shapiro, G. 1990 Bottomlands and Rapids, in Lamar Archaeology, Mississippian Chiefdoms in the Deep South, edited by Mark Williams and Gary Shapiro, pp. 147-162, University of Alabama Press, Tuscaloosa and London.
Smith, M. T., and J. B. Smith 1989 Engraved Shell Masks in North America. Southeastern Archaeology 8(1):9-18.
Snow, F. 1990 Pine Barrens Lamar, in Lamar Archaeology, Mississippian Chiefdoms in the Deep South, edited by Mark Williams and Gary Shapiro, pp. 82-93, University of Alabama Press, Tuscaloosa and London.
215

Sturges, D. 1818 Map of State of Georgia. Published by Eleazer Early. On file, at Surveyor-General's Office, Atlanta, Georgia.
Swanton, J. 1922 Early History of the Creek Indians and Their Neighbors. Bureau of American Ethnology, Bulletin 73, Washington.
Trimble, S. 1974 Man-Induced Soil Erosion on the Southern Piedmont: 1700-1970. Soil Conservation Society of America, Ankey, Iowa.
Tyler, J. and T. Jackson 1969 Report on Clearance ofPothole A, Bull Creek Site. Unpublished manuscript on file, Columbus Museum, Columbus, Georgia.
Walker, J.W. 1989 Ocmulgee Archeology: A Chronology. Southeastern Archeological Center, National Park Service, Tallahassee, Florida.
Weissner, P. 1983 Style and Social Information in Kalahari San Projectile Points. American Antiquity 49 (2):253-276.
1984 Reconsidering the Behavioral Basis for Style: A Case Study among the Kalahari San. Journal of Anthropological Archaeology 3:190-234.
Wharton, C.H. 1978 The Natural Environments of Georgia. Georgia Department of Natural Resources, Atlanta.
Willey, G.R. 1949 Archeology of the Florida Gulf Coast. Smithsonian Miscellaneous Collections 113. Washington.
Williams, S. 1979 Some Negative Painted Pottery in the Southeast. Paper presented at 1979 meeting of the Southeastern Archeological Conference, Atlanta, Georgia.
Wobst, M. 1977 Stylistic Behavior and Information Exchange. In Papersfor the Director: Research Essays in Honor of James B. Griffin, edited by C. E. Cleland, pp. 317-342. Anthropological Papers 67. Museum of Anthropology, University of Michigan, Ann Arbor.
Wood, K.G. 1991 Archeological and Historical Investigations of the Proposed Riverwalk Project, Columbus, Georgia. Prepared for Jordan, Jones and Goulding, Atlanta. Southeastern Archeological Service, Inc., Athens.
1993 An Archeological and Historical Assessment of the Columbus Waterfront from the City Whaif to Oglethorpe Bridge. Prepared for the Columbus Water Works. Southeastern Archeological Services, Inc.
216

APPENDIX A
Transcription of Lester's Field Notes Bull Creek Site
1936

Mar. 12, 1936. Visited and made a sketch map of a Village Site, situated just North of Bull Creek, at its junction with the Chattahoochee River. The site is rather well defined, in that a roadway has been cut through it and the exposed profile shows a very definite midden accumulation, ranging in thickness from about six inches to about sixteen or eighteen inches in the middle, where the exposed bank shows the signs of a house of some type - timbers and quite an accumulation of briquettes. This deposit shows in the bank for a distance of some three to four hundred feet and appears from the exposed surface, to be very similar to the Mossy Oak type site on the Ocmulgee.
The ridge or hummocky places nearby seem to have been caused by a company making a railroad cut and are nothing more than spoil banks.
3-12-36. During the afternoon Mrs. Patterson and I visited a place across the river on the outskirts of Phenoix City, Ala., where a portion of a skeleton had been unearthed - also associated with it were some trinkets, that had the appearance of early made articles. A complete circle to go in the lobe and a little bell-like pendulum below, also some small buckles of pewter.
The hill from which these were taken has been used as a gravel pit and the top has almost entirely been removed to a depth of about 31h feet. The underlying geological formation appears to be a type of red sandy loam, either of the Greenville or Barnesville formation - some indication of mica in it. It is possible that the rise may have a small mound on top but for the most part, at least, it is probable, in fact almost certain, that it is nothing more or less than a natural elevation.
Also visited two sites on the Bussey plantation, in Ala. on the bank of the Chatt. just below the mouth of Bull Creek. Both of these sites had quite a bit of surface pottery and were rather extensive. Supposed to be Coweta towns.
3-13-36. Started excavation on the Bull Creek Site this morning, with a crew of 3 negro boys and myself. As previously mentioned, the site has been cut into by a roadway on the east and the river has made rather deep inroads from the west. The place where work was begun is on this narrow strip of ridge, on a bluff, about 40' above the river.
Control Trench #1 was staked out - trench 5' wide - stks intervalS' - stakes nos. starting at 1-39 and going to 1-52. Worked four sections, the width of the ridge from 1-39 - 1-40 to 1-49 - 1-50. Trench #1 has an orientation of N. 15 deg. W - magnetic.
The top soil is a river silt deposit of yellow sandy loam, containing an unusual amount of mica; it ranges in depth from 2" at sta. 39 to about 10" at sta. 48. No sherds were found in this level. No finds.
Immediately below the yellow top soil is a midden deposit of rather greasy type of "blue clay. " Quite a number of sherds came from this level, however, they were all exceedingly small pieces. The majority of pottery found is plain, hard, grit-tempered ware, rather greyish in color - the pottery also has a rather appreciable amount of mica in it.
Find #1 - a rim-sherd with a raised notched design, about an inch from the top.
Find #2 - a rather peculiarly shaped piece of sandstone, about an inch in diameter, so nearly circular that it might have been used for a chunky stone.
A-I

Immediately below the blue-clay level which was about 2 to 4" thick, is a sort of greyish sandy clay containing quite a no. of briquettes and some sherds. This level appears to be immediately above the baked clay, belonging to the house debris. In both this and preceding level, practically no flints have been found and none with secondary working. Typical profile to date.
3-14-36. Continued excavation on house site, exposing 10' of red sandy clay mixture, presumably roof accumulation, between sta. 1-45; 1-46 and 1-49; 1-50. Area south of site 1-39; 1-40 is of a grayish sand clay-midden deposit. It appears that the scattered edge of the house site is directly under sta. 1-46; in section 1-45; 1-46 to 1-49; 150. A baked clay ridge was exposed, I" deep in red level, this ridge appears to be floor material, and is shaped something on the order of a fire bowl, except that it has a very small degree of curvature. Further troweling should prove more enlightening on this.
The sherds from this and the immediately adjacent level of greyish clay, are very abundant much larger pieces than previously noted.
Immediately adjacent to the bake ridge mentioned above, are other baked areas appearing at irregular intervals and slightly different levels. It is probable that all of these belong to the wall and roof structure. Quite a number of briquettes, showing reed imprint, were also found in this level.
Find #3
3-16-36. Removed yellow silt deposit and portion of black midden deposit from section 1-49; 1-50 to 157; 1-58. Worked black midden to a depth of 1 to 3 in. - sherds plentiful.
Find #5 - a piece of briquette, showing reed imprint, found one inch in reddish, sandy clay of
house debris - 32" S of sta, 1-49 and 21" west.
Find #6 - a rim-sherd, showing a notched design on top of the rim, an incised line just beneath it and a portion of an incised whirl, showing punctate marks in it.
3-17-36. Excavated black, midden deposit from sta. 1-49; 1-50 to sta. 1-57; 1-58, to a depth of 14 to 30 inches. Quite a large number of sherds in this level. Also started excavation of reddish sand and clay (house debris) sta. 1-46; 1-45 to 1-52; 1-51. Briquettes, charcoal and pottery are scattered throuout this level.
Note for 3-17-36. Found a small piece of bone, 3" deep in reddish sand and clay, - extreme dampness made it impossible to take up. Presumable leg bone of deer, one end of which had been charred. Large "pores" in knuckle section. 29" S of sta. 1-50 to 6" W.
3-18-36. Continued troweling house site, uncovering quite a large section of briquette material between sta. 1-49 and 1-50, irregular in shape and thickness. Quite a quantity of charcoal was evident, but no timbers except a short section between sta. 1-49 and 1-51, nine to sixteen inches N 1-49 - 6" E. 11" below surface, in reddish sand and clay. Presumably a section of matting of reed with a round pine pole. Another short section 10" long was found 21" S of sta. 1-51, II" E, II" below surface; this appears to be a short section of a log, about 4" in diameter.
3-19-36. Further excavation of house debris continues to show briquette material in abundance. Started troweling floor section and uncovered a piece of broken pottery, laying flat on the floor, with a few
A-2

fragments of human bones in it. This is located just on the edge of the bank (road side) 19": S - 16" E of sta. 1-50. Immediately adjacent to this (26" S- 12" E sta. 1-50) is a portion of a small timber with a large briquette moulded around it.
3-20-36. Started excavation of a section between 1-39; 1-39L1 and 1-57; 1-57L1, working yellow silt top soil, and a portion of the black midden underneath. Sherds plentiful in the midden level. Also continued troweling on floor, uncovering quantities of briquette material and charcoal. Bad weather prevented much headway.
3-23-36. Excavation of black midden deposit 1-39; 1-39L1 to 1-57; 1-57Ll. Quite a number ofsherds, Find #11, a broken rimsherd with a very small handle found 10" deep in black midden deposit, 15" below surface, 22" S - 30" W sta. 1-49.
Between Sta. 1-45 and 1-47 a distinct line of red is observed presumably the west edge of the house debris.
Continued troweling the floor, struck a small portion of a timber about 4" long, 28" S and 3" W Sta. 1-52, lying flat on the floor, also a moulded briquette, 40" Sand 15" E Sta. 1-52.
3-25-36. Rain! - no work. Made arrangements with postmaster for storage space in new P.O. and possible office space in old post office.
3-26-36 more rain and Friday.
Sat., March 28, 1936. Continued work on Bull Creek #1 house site, work out floor. Also visited and made surface collection from Coweta #3 and site at Abercrombie Mound, located on Mr. Brannen's farm, on the Ala. side of the Chattahoochee, about 3 miles below Bull Creek. Visitors in town, Dr. Kelly, Hanna, Solomon and Tony Waring.
April 6, 1936. Started N.Y.A. crew of light excavating, a 5' test trench, from sra. 1-87; 187Ll to Sta.
1-123; 1-123L1. This section lies just north of the house site already described, and shows a slight ridge running through it. Removed yellow topsoil from the whole section, topsoil ranging from 4 to 20" inches. Very few sherds, one Find (#12), a quartz arrowhead of the modified spinner type, found between Sta.
April 7th, 1936. Rain - no work
April 8th, 1936. Continued work on test trench. Working from yellow topsoil into a very rich black midden deposit. Quantities of sherds were found in this level. 5 finds (#13 to 17) Quite a number of briquettes were found in the level and it is probable that there is a portion of a house between 1-93 and 1-107. It appears that the midden deposits runs out between 1-107 and 1-111, grading into a yellow subsoil.
Midden deposit also contains things of late historical occupancy; mixture probably due to plowing; bricks, glass and other articles of such nature, mixed in with pottery fragments and briquette material.
April 9th, 10th, 11th. Rain!
A-3

April 13. Unable to work in L1 Test Trench, due to wet weather, started excavation of 5' Test Tr. E L3 line stakes from Sta. 1-95L5 to sta. 1-111L5, worked through yellow topsoil and into black midden to a depth of 9" at 1-95L5 and 13" at sta. 1-111L5, yellow topsoil an overage of 4" deep. Sherds plentiful- 3 Finds (# 18,19,20) at a depth of 13" below surface at Sta. 1-99L3, briquettes became fairly numerous between sta. 1-97L3; 1-97L5 and Sta. 1-99L3; 1-99L5. Found the pieces of a small pot in place and it will be possible to reconstruct in original form. Found 29" S - 8" W Sta. 1-99L3, also found a complete rim, pinched or beaded design, 1 ft in diameter, found 13" S - 22" W Sta. 1-99L3. Started removing section between l-L1 and 1-L3 between sta. 1-95 and 1-103, to expose house debris. Sherds plentiful in both levels.
4/14/36. New crew of20 men. Extended Test Trench ofL3, L5 to sta. 1-143L5. Worked to a depth of 13" at 1-111L5 to 16" of 1-143L5. It is noted that the black midden deposit appears to run out between 1-115L5 and 1-119L5. being replaced by yellowish clay loam. Very few sherds found in clay loam area. Also started excavation of trench Lll, L13, starting at 1-109Ll3 worked black humus and midden deposit to a depth of 9", worked to 1-143L13 to a depth of 5" Black midden deposit appears to run out at Sta. 1-127Ll3. Yellowish clay appears to extend beneath it, back to 1-107L13. Found outline of a pit in yellow clay deposit, just below black midden deposit. Find #21 found in pit just above burned area.
April 15-16. Continued excavations of area between sta. 1-91Ll - 1-91L5 to 1-123L1- L5. Briquette material appearing at a depth of IS" below surface or 1-91Ll and extending from 1-91L1 - L5 to 1103L1 - L5. Found a pot, broken, but pieces in place, found 28" from 1-99L3, 59" - sta. 1-99L5 - 44" sta. 1-99L3. Scattered fragments in immediate vicinity, appear to belong to these two pots. Portion of another pot was found 65" SE 1-97L1, 41" NE 1-95Ll. Pot is of plain type ware, resting rim down, surrounded by briquette material. There seems to be an absence of the huge moulded type of briquettes found in the first house.
April 20, 21 -'36. Worked area between sta. 1-91L5 -L9 to 1-103L5 -L9, down to a depth of 13" below surface, going through top soil and into black midden deposit, also worked a portion of house debris striking floor between 1-99L1 and 1-99L5. Briquette material shows from 1-95 to 1-95L5 to 1-103L1; 1-1-99L5. Briquettes seem to be of moulded variety but do not show as clearly as in house #1, because of plowing through them.
Total lack of timbers is also probably due to plowing. Several large pieces of pottery were found just on and above floor level. There is a total absence of trade material but modern brick are found mixed all through black midden deposit. See find #34.
4-22,23,24 -36. Staked off Con. #2, across R.R. cut, from sta. 2-55-56 to 2-91-92. Worked through topsoil and into yellow sandy clay-loam deposit underneath. No heavy accumulation of midden deposit, but sherds rather well scattered throughout plowed soil and down into yellow loam. Found a pit in Con. #2 at Sta. 2-59, showing up just under plowed soil.
Two boxes of pottery have come from this pit (See Pit #2 material), several pieces show a rather unusual design of punctate, incised body, with notched rim. At present the pit has not been completely troweled out.
Also started removing briquette material from floor and have found several large pieces of pottery resting just above floor and presumably belong to some of the pits already described. No timber supports as yet but briquettes show more of the massive size of the first house. Also removed topsoil and a
A-4

portion of black midden deposit from section between 1-103L5 -L9 and 1-115L5 -L9, worked to a depth of 6" below surface. Several boxes of pottery from this section.
The briquettes material shows use of fiber for binding purposes and there seems to be an absence of the matted reed construction -- on the whole it is very similar to the other house except that it has a harder clay floor, either due to packing or baking to a small extent. The briquettes shell off the floor proper very easily and it is thus easy to work out.
April 27 - May 4. Continued to remove briquette material from what appears to be floor of house site #2. Portions of floor appear to be intact and other portions (in can #1 - Ll) show signs of having been plowed into. Small brick fragments and occasional pieces of metal show evidence of having been plowed under. At present no post mould impressions are showing.
Two burials have been located in the side of the bank, about 100 yards from where Bull Creek runs into the Chattahoochee River. One has been washed from place and has fallen into the yellow clay muck down below but the other is still in place; with a portion of the lower leg bones and thigh bones sticking out of the bank it seems that both the feet and the pelvis bone have been washed away. In the first - the skull was removed from position by the person who found it before notifying me.
May 5-6. Continued to work on house site #2. On the whole it seems to have been plowed into rather badly; very little evidences of any floor material except that already described. Working plowed soil off down to a yellowish clay loam, showing signs of briquettes and small pottery fragments. A few burned areas have turned up that may possibly develop into post-holes.
Excavated a 10 X 10 pit over the burial in the bank of Bull Creek and worked down to it to a depth of below the surface. On the whole the body is in good state of preservation, but the skull is broken in on the left side -- pelvis bone and feet are gone. Troweling out has developed that it is a flexed burial -- male (?) skeleton -The body is flexed, knees drawn up almost under the chin, and arms folded, with the left hand between the legs. All of the upper vertebrae are gone but some of the ribs remain in place.
May 7-18. Continued work on House # 2, with very small group of workers. Removed pottery from floor (?) level and continued on down. Ground so hard that necessary to keep wet all the time. Floor appears to be very uneven, sloping toward the southwest. Found baked area in S.W. corner at least 3" below level of that in N.E. corner. Still finding some pottery in floor level and traces of briquette material. Very few post moulds have turned up at the present time. Floor area is very uneven, showing patches of baked material in spots. All indications point to use of hard packed clay floor, with patched places baked in place.
May 18-22. More work on first house site, uncovering floor level between 46-45 and 52-51. Removing briquette material from hard-packed blue clay. Made a plaster mould of a large molded briquette. Started working midden deposit beyond immediate confines of house site, just below level of floor. The underlying clay seems to slope down just beyond house site. Found a large refuse pit, containing quite a bit of pottery in flooring location.
Also found some small calcined bone on floor of house site between Sta. 1-47 - 1-48 to 1-49- 50. (Above drawing) Bones are very tiny and have been burned completely through. More post moulds are turning up in this house and should present a very nice architectural floor plan.
A-5

May 25-26. Continued work on #1 house site; more evidence of post mould impressions. Removed moulded briquette from floor of house. Find #? Very little pottery showing up on house floor. Floor stilI shows hard-packed clay variety, with layer of charcoal just over floor. What post remain in holes are partially charred, but what fiber remains uncharred is not totally rotted. It appears that the posts charred, all around the outside and the portion stilI remaining in natural state is the center of the charred portions. See drawing of house for more accurate location of individual post-moulds.
May 27-June 2. Lowered 15 X 15 pit at burial on down to level of burial #2 and found evidence of another burial pit in north wall. Then extended pit 10' to the north and worked on down, finding pottery in the waterlaid sand. At a depth of approximately 5', an effigy pot was found made in the shape of a dog - painted ware. The pot is unbroken and is an exceptional piece of ware. The "dog" is complete, with eyes, nose, ears, four feet and short stubby tail. The pot is part of the furniture with a burial, found just below the pot. The body is face downward, and have only found one arm.
June 2-8. Continued excavation on burial # 3, working down to remainder of skeleton. Sherds occur in all of fill element down to burial and some small pieces right in place with skeleton. Body is laying on the left side, head pointing South - contracted type of flexed burial - knees drawn up almost under the chin, with the pot resting between knees and chin. The hands are between the lower leg bones, right hand about 3th " below the knee-cap and the left hand about 10" above the head. Location Burial #3 between Sta. 4-14 and 4-13.
6-8 - 6-9. Started 10 X 10 pit, 10' N of burials #3 and 4, worked down to a depth of 7"; no sherds found in level. Found a skull of a dog 7" below surface, 26" N Sta. "A". Started form for removal of burial # 3 as a complete unit.
6-10 - 6-12. Lowered 10 X 10 pit to a depth of 31". Top level w-l, sand 3" to 11" deep. Old sod (historic) lies 6" deep - accumulated sand - 14" deep. Bone sherds in Band C levels. Also removed Burial #3 intact, by pouring concrete block around burial and forcing 3/8" steel plate under concrete. The body was placed on exhibit at M & M Bank and has caused quite a bit of favorable comment. Location of Burial # 3 -- 4-
6-15 to 6-22. In extending pit Northward, to remove burial #3, also ran into another burial, which was rather badly cut by the person making the discovery; the burial was found at a depth of 44" below surface and only 20" beneath bottom of occupation level.
The body is laying on the left side, head facing N.W., body oriented N.E. and S.W. Such bones as remain are in very good state of preservation, with exception of fingers, toes and ribs. The skull, pelvis and sections of the vertebrae were cut into with the shovel but the rest of the body is intact.

I-$~ r-""r.nr---i1-541
Found 29" below surface at 1-58

e / /_.r.~~~-----J/-.$&
\

A-6

1--'17

~-/~ . . . . .~
.:.~ .I
I
I i
;

1-/3

i-- 4-/5LI
I
I
I I
!

. _ - - - - - - - - - - - - -4-,/.

Field Drawing of Burials 3 and 4. A-7

;
I
. _....__._. Ji
'/

The body is on the left side, arms straight down toward the thighs, with legs folded back under the body - knees in front with heels just below the pelvis bone - slightly different from the other type of flexed burial. As the hand bones are gone, it is impossible to tell whether or not they were between the legs but it seems highly improbable, as the distance is too great from elbow to the thigh bones. The bones are comparatively small and it appears to be the skeleton of a woman. Location Burial # 4 (between Sta. 4-15Ll and 4-15).
Lowered 10 X 10 pit, sta. 4-20 to 4-24 - 4-19Ll 60 4-23Ll, to a depth of 78" at 4-23Ll, 64" at 4-20. Sherds plentiful to bottom of occupation level. Found a burial just under sta. 4-21Ll, in West profile wall, and extended pit to 4-19L3 to 4-23L3, working down to burial pit. A greenstone celt was found with the burial, resting just back of the head. Skull practically gone and remainder of bones in very poor state of preservation. Details later. Also worked section between 4-26 - 4-20 to 4-15Ll - 419L1 to a depth of 64" at 4-20. Found a burial in the wall on present level - (64") just under Sta. 417Ll. Also found a burial in wall 24" W. Sta. 4-19Ll Details later.
6-23 -6-24-1936. Rain!
6-25. Troweled out burial # 5. Burial is of the contracted, flexed type. Body laying on left side, head pointing SSE. Bones in a very poor state of preservation. Burial furniture consists of a greenstone celt, approx. 7" long and 3" wide at cutting end - 2" wide at handle end. All of the skull is gone, except the upper and lower teeth; no trace of shoulder bones and only the 3 lower joints of vertebrae. Right arm folded across the body, left arm going between the thigh bones, with hand at heels. Burial is in tan sand, 36" below bottom of occupation level, 72" below surface at sta. 4-2IL3. Found quite a few sherds in pit fill. (See box of section.)
Also lowered 5' X 10' pit, 4-15Ll to 4-19LI-L3, to a depth of32"; sherds plentiful in occupation level.
June 29. Lowered section 4-15Ll -L3 to 4-19Ll -L3, to a depth of? below surface; found 4 burials in section; one, #8, directly under 4-19L3; one, #7, on line of sta 4-19L1 -L3; one, #9, between 4-17Ll and 4-15Ll; and one, #6, on line of sta. 4-17Ll -L3. Detailed locations later.
June 30th to July 3rd. Burial # 7 has with it another of the dog effigy painted pots. This pot however, is rather badly broken, the neck being found down on the pot itself, and the head and three of the feet broken off - all parts are in place though, and the vessel can be restored to its original shape. From all appearances the pot is an exact duplicate of the one found with burial #3. There is also a small greenstone discoidal stone as burial furniture; the stone is perfectly smooth and round, about 1%" in diameter and about 5/8" thick, in the middle. Removed the pot after taking photos because of danger of theft. Greenstone celt with Burial #5 having been stolen. Location and description of burial later (after further troweling).
Burial # 7 is of contracted, flexed type bones, in very poor state of preservation; body laying on left side, oriented N-S, with head to south. Skull has been crushed and remaining bones consist of sections of tibiae, femora and upper arm bones. Appears that hands went between the legs as in other burials but all trace of bones gone.
Burial # 8 is of a flexed type, with feet drawn up under the body, knees pointing away from body. Bones in rather poor state of preservation, not completely troweled out present time.
A-8

July 6th to July 8th. Moving surplus dirt back to R.R. fill. Started excavation of section 4-7L3 to 415L3. Lowered to a depth of 35" at 4-15L3 21" at 4-7L3. Sherds plentiful in occupation level. There is what appears to be the outline of a burial pit between 4-15L3 -L1 and 4-11L3 L1, Burial #10 first appeared at 20" below surface at 4-11L3. Several pieces of painted pottery are noted in the sherds coming from this level. Also staked out extensions of Control #4 and ran traverse lines from Con. #1 to Con. #4
July 9. No Crew!
July 10. Continued excavation on section 4-7L3 to 4-15L3. Lowered to a depth of? at 4-7L3; at 415L3 Outline of a burial pit shows between 4-7L3 and 4-11. Found Burial #11 at a depth of? below surface No burial furniture.
July 13-14-15. Lowered section Sta. 4-16 to 4-16R1 - 4-24 - 424Rl to a depth of 24" below surface, also removed large tree between sta. 4-16 and 4-16R1. Found bones of large animal 10" below surface, with a small seed bead resting with bones (Find #45). Lowered section 4-7L3 -L5 to 4-23L3 L5 to a depth of 4" at 4-L5 and 6" at Sta. 4-23L5. Note continued rain has caused considerable damage to burial 4-5-6-7-8-9-10-11, but bones can probably be salvaged.
July 17, 1936. Continued excavation on section 4-7L3 -L5 to 4-23L3 -L5, to 4-23L3-L5, working to a depth of 36" at Sta. 4-23L5, 9" at sta. 4-11L5, working through occupation level from 4-15L5 to 423L5, approx. 4" in tan sand. Considerable pottery from occupation level. Three finds #'s 46,47,48. Only 3 in crew.
July 20, 1936. Continued excvation in section 4-7L3 -L5 to 4-23L3 -L5, working through occupation level and into tan sand, working to a depth of? below surface. Pottery fairly well distributed throughout the level; so far no trace of burial pits as yet. Also continued section 4-16 to 4-24, working tan sand to a depth of? below surface. Find #49, a small clay bead found 5" deep in tan sand, 40" below surface, 11" E 4-18, on line sta. Find #50 a rimsherd of unusual design, found 6" deep in tan sand, 40" below surface, 35" S. Sta. 4-18, also found an outline of a burial pit between sta.
July 21, 22, 23, 1936. Lowered section 4-16 -4-16Rl to 4-24 -4-24Ll to a depth of 4.20 Sherds plentiful in occupational level. Found a burial #13, located at a depth of below the surface, between Stations 4-16 -4-16Ll and 4-18 4-18Ll. Not troweled out at present; skull broken rather badly when found. Also found a burial #14, between stations 4-18 -418Rl and 4-20-4 20Rl; bones in excellent state of preservation. Body lying on left side oriented E & W with skull pointing E. Body is in the contracted, flexed position with knees drawn up almost under the chin, with hands passing between the thighs and lower legs. Skull has peculiar indentation on back of occipital, unnatural deformation - none of the ribs remain, however a few sections of the vertebrae remain in place; pelvis bone in fair state of preservation and a few bones of the feet are still in place, showing arch of foot and upper toes. Skull is also enlarged on right side, being slightly misshapen.
Began troweling out Burial #6. Burial is offlexed type laying on right side -- Body oriented NWSE, head to NW with knees pointed away from body and feet drawn up under the pelvis bone. Arms go straight down at side and fingers nearly reach to femur. Body in rather poor state of preservation, having been considerably damaged after discovery, by heavy rains. Burial furniture consists of a small green celt, about 41h" long, 21h" at cutting end and I" at butt end; two small green stone chisels. A very nice flint knife of black flint, some small pieces of worked flint and some very peculiar reddish powder in globule form, apparently some ferrous oxide, that was possibly used for paint. The flint articles and the greenstone chisels are resting on the "paint" material. 67Ih below sur. 36" in tan sand.
A-9

Burial #7 is in a very poor state of preservation, nothing remaining except small section of fibula of right arm, sections of both femur and fragments of tibia. From appearances it seems that body was laying on left side, oriented SE-NW, head to S.E. and in all probability the hands went between the legs; only small section of? and radius of right arm found. 78" below surface 47" in tan sand
July 27 to 30. July 27 Moved skeleton from M&M Bank. July 28 to 30 Troweled out burials exposed and began cleaning excavations. Found a dog-effigy type 7" S. sta. 4-15L5, 32" E. 70" below surface, 36" deep in tan sand. Pot is slightly different from other two in that neck design is slightly different, neck being higher and smaller. Body of pot painted red with black design. Vessel in perfect state of preservation except that end of dog's tail broken off and one leg loose, but in place. The legs are slightly longer and place at a slightly different angle.
July 31 -- Rain!. August 3,4,5. Complete troweling of burials 8. 14, also started excavation of section 4-24 - 4-23L5 to 4-28 - 4-27L5.
August 10. Continued excavation of section 4-24; 4-23L5 to 4-28; 4-27L5 lowered to a depth of 52" at sta. 4-27L5 and 29" at sta. 4-28. Sherds plentiful through occupation level.
Also troweled out burial #11 between sta. 4-15L3 -L5 and 4-19L3 -L5. Body is of the flexed type, head pointing SSE. Arm straight at sides and going about midway between pelvis and knee. Thighs point straight downward with legs back toward pelvis. Bones in fair state of preservation, a few of the ribs still remaining intact, also some of the bones of the arch of the foot. No burial furniture. The skull is rather small and to all appearances this is the burial of a female.
August 11 to 14. Continued excavation of section 4-24; Rl 4-23L5 to 4-28; 4-27L5, lowered to a depth of 83" below surface at sta. 4-27L5. Located a burial. #18, immediately below sta. 4-24. Burial not completely troweled out, but burial furniture consists of effigy pipe, Find #64 A, core of 63 conch shell and large beads, 64B, presumably of seed. 65, and a small pot 3 3/4 in. dia. incised and punctate. Burial found 69" below surface. Also removed Burials #11. 5, 6, 8, 9.
Burial #8 is of the contracted, flexed type. Body lying on left side, oriented E-W with head to East. Skull in fair state of preservation, vertebrae of neck and portion of shoulder blade and collar bones remaining intact; left arm folded across body, hand at pelvis bone, right arm straight, going between thighs, hand bones under right tibia. Pelvis, tibia, femur, fibula and bones of feet in excellent state of preservation. Three sections of lower vertebrae remaining intact. No burial furniture. Burial located just under sta. 4-19L3. 44" below occu. in tan sand.
Burial #9 is of contracted, flexed type, bones in poor state of preservation. Oriented SE-NW on left side. Nothing remains of upper body except teeth, vertebrae of neck, section of upper arm bone about 4" long. Lower body remains - small section of pelvis bone, tibia, femur, fibula and small section
of hand between femur. Burial furniture consists of small greenstone chisel, approx. 3 lh " long (Find
#56) Burial located 66" below surface 32" in tan sand between sta. 4-15Ll -L3 and 4-17Ll- L3.
Burial #12 consists of small fragments of skull and portion of tibia and femur of both legs. Bones in very poor state of preservation. Burial is of contracted, flexed type, knees in front of body, oriented NW-SE Head to NW. Found between sta. 4-9 - 4-9Ll and 4-11 - 411Ll at a depth of 35" below occupation level. Body laying rt side.
Burial #10 consists of broken fragments of a skull, all other traces of burial gone. Skull crushed beyond repair, burial located between sta. 4-15Ll -L3 4-13Ll -L3. Found 27" below occupation level in tan sand 54" below surface.
A-lO

t- I
EI!j(/Y Pof ~f"

I- '~ 4-/f t:

D/sctJrdal fto)'Je ..9 N'yY J{ 11)/.

JI.-/1 J-/

;;r).L~ .-0--:----.-. ---_.------ _ _~--------J.~-7/ J-/
. - ._. l- -- ..- - - - -
J-ocaJ;~" aLl rJ;~( :# S I

~I 1;6 ~\

.....J

~.lp o"t-:;

::::

Field Drawings of Burials 7 and 8. A-ll

I
I
I
I
1
I
1I 4'
r
_.....
';'-17'- /

. ~_ ...
N

i
1.5' p
~ I 4-- :J.() J
_II'~~~

4--/.5"1- 5 ----I
I
+-/71-~

Field Drawings of Burials 9 and 14.

A-12

Burial #13 consists of the fragments of a skull, found immediately below large tree - roots have destroyed all evidence of other bones; skull crushed. From appearances it appears that body was lying on left side, oriented N-S with head to S. Only a few fragments of the skull - some teeth and a few sections of the cervical vertebrae, small section of collar bone. No trace of burial pit - no burial furniture. Found 26" in tan sand, 40" below surface.

Burial #15 is in a very poor state of preservation, body laying on left side, oriented NNW -SSE. The burial is of the contracted, flexed type, arms appear to go below the legs, however bones in such poor state of preservation it is almost impossible to determine exact relation. Remaining bones consist of skull rather badly crushed frontal bone, decayed entirely, short section of both humerus, trace of ulnor and radious of right arm, pelvis bone almost entirely gone, and only short section of femur and tibia, no trace of fibulas or any of bones of appendages.

Burial #16 consists of a few fragments of tibia and femora, bones are scattered over a large area and this is really the remains of two bodies; general orientation is south southwest and north northeast. Burial furniture consists of the 3rd effigy dog vessel. 36" below occupation level.

Burial #20 is of the flexed type, vertebrae straight, knees at right angles to vertebrae. Bones in fair state of preservation; remaining bones are skull, cervalic vertebrae, both femora, portion of both tibia, small section of pelvis, portions of collar bones and shoulder blades. None of lower vertebrae, ribs, hands or feet. Body laying on left side, oriented ESE and WNW, head to the ESE. No sign of burial pit. Arms folded across body - folded at elbow. Burial furniture consists of beads around the neck.

Burial #18 is located directly under sta. 4-24. Bones in very poor state of preservation, only portions remaining - very small section of skull, a few of the molars and both femurs. Body appears to be laying on left side, oriented SE-NW, head to SE. Knees drawn up toward chin, and appears to be typical Bull Co. type of flexed burial. Burial furniture consists of a small pot, Find #65, approx. 31h dia., incised and punctate design, rather poor workmanship; several large beads, Find 64B, of shell or seed, and unbroken clay pipe, Find 64A, human effigy type, and a small piece of conch shell core Find #63.

Burial #19 is of the contracted flexed type, body oriented SE-NW, head to SE laying on right

side. Bones in fair state of preservation. This burial is contracted more than any previously found - very

good illustration of flexed type, with arms going between the legs. All bones remain, except that skull

has been crushed and is rather badly decayed. None of the cervalic vertebrae remain, but all of the

thoracic together with pelvic and pelvis bones. The arms are straight to elbow and then are flexed to go

between the legs. Female burial. Complete overall length of burial, 31". No burial furniture. Burial

pit first appears at top of yellow clay loam. Burial found 67" below surface, 45" below occupation level.

14" in yellow clay loam.

.

Burial #17 is of the contracted flexed type, oriented NW-SE head to NW. Body laying on right side with the knees at right angles to the vertebrae and both arms going between the legs. Skeletal remains in rather poor state of preservation - one side of skull crushed and all of metacarpal and metatarsal bones gone. Only a few joints of the lumbar vertebrae remain but 5 of the thoric ribs. Portion of skull was all that could be removed.

Sept. 10 and 11, 1936. Moved dump started backfilling hole and excavating 4-7L5 -L7 to 4-27L5 -L7,
a also worked section 4-14 - 4-14Rl to 4-20 -4-20Rl down to clay loam, located burial (#22) between
sta. 4-16; 4-16Rl and 4-18; 4-18Rl almost immediately below the location of burial #13. The body is

A-13

Bvy/~I #/0

/f-/.s-J..3

.IJ./~L./

J ).7
6-~-

__ - - - - .-.---_... ._-~./~ 1-/

,-----/.r L . r

-If"- 3

----------------------"'---.

t

l
Field Drawings of Burials 19, 10, and 16.
A-14

~__.___. ./

about 6" in yellow clay loam and the burial pit first showed about 6" above the loam in tan sand. Located another burial (#24) almost under sta. 4-7L1 in tan sand. Details on both of these burials later.
Sept. 14 and 15. Began excavation on section 4-7L5 -L7 to 4-27L5 -L7. Worked through water-laid sand, yellow clay, historic occupation, tan sand, and into Indian occupation level. Very few sherds in upper levels - rather plentiful in tan sand and occupation level.
Also began troweling out Burial #22. The body is laying on the right side, oriented N.W. and S.E. with head to N.W. The skull has been crushed on the left side but otherwise the body is in an excellent state of preservation. The pelvis is slightly higher than the head and has been slightly displaced by root growth but is in and excellent state of preservation. Remaining bones consist of - skull, in fair state of preservation - clavicles - scapular - both humeri - ulna- radii - a few of metacarpal bones practically all of the vertebrae - tibiae - femora - fibulae - metatarsal bones and phalanges. Found 54" below surface 40" in tan sand.
Sept. 16 to 22. Continued excavation on section between 4-7L5 - L7 and 4-27L5-L7, working down into yellow loam. Located 5 burials in section. Burial #25 between 4-9L5 -L7 and 4-11L5 -L7. Burial #26 between 4-11L5 -L7 and 4-13L5 -L7. Burials #27 and 28 on line sra. between 4-15L5 -L7. Burial #29 between 4-17L5-L7 and 4-19L5 -L7. Also troweled out Burial #17 - see notes under Burial # 17 heading.
Sept. 22 to Oct. 14. This period has been marred by much bad weather, rain has prevented the working out of burials 23 to 29. Excavation was begun on the section L7-L9 and have worked section from 4IlL7 -L9 to 4-27L7 -L9 to a depth of 75" at sta. 4-27L7 -L9. Burials #30 31 and 32 have been located between 4-7L7 -L9 and 4-11L7 -L9. At the present time it appears that portions of the Burials are in the unexcavated section between L9 and L11. Burial #31 is a triple burial, found between sta. 4-9L7 -L9 and 4-11L7 -L9. These remains are in a very poor state of preservation and it is impossible to determine the exact relationship of the individual bones. It appears, however, that at least two of the bodies were flexed in the typical Bull Creek manner and the other burial was of the skull only. All of the bodies were oriented SE and NW, with the heads to the NW and were buried with the face upward, heretofore in the majority of cases the head has been turned to one side or the other. Also the skull (a) (buried without the body), highest up, is gone with the exception of portions of the back occipital and lower jaw. All that remains of (b), the next, is portions of the complete skull - frontal bones and facial bones missing and a short section of the humerus (c) consists of portions of the upper and lower jaw - traces of both humeric and radii and sections of both tibiae, fibulae, and femora. No burial furniture.
The profile at this point shows a band of clay from 2" to 6" thick between sta. 4-17L9 and 423L9, suggesting a house floor. There is also some trace of a sod development below the general occupation level, though it shows no definite form at this particular point.
No date. Burial #24 consists of broken portions of the skull and fragments of the humeri, femora and tibiae. The body left side, oriented NW-SE with the head to the SE. No burial furniture. It appears that this is the contracted type of flexed burial. Found 28" in tan sand - 59" below surface. 4-7L1
Burial #25 is in a very poor state of preservation. Body laying on right side, oriented NW and SE, with head to NW. Remaining bones consist of portions of skull and fragments of femora - tibiae and fibulae.
From position of femora and tibiae it appears that the legs were doubled back under the body with the knees in front. 52" below surface - 21" in tan sand.
A-IS

JfS 1/ l:;)iZ/OTY (...IG .....-r -.- .
I l . / ' l _ ..I J AI""''7'''7''7

r - ---_ __--- 4-/8 L oca,L/on 0/ Ovr/Q/#.22 l+= 7 r

."f-/8R/

~

~

"

~

)/

t

. nn r~\

.

~~

~

i
i
.J
f-/~R(

L ocaf/O/7 Q/' 8wr/QU *30-3/

-

~~9

. > : c=d

!JurI '0/#.30

L/

~c::>

Field Drawings of Burials 22, 30-31. A-16

Burial #26 is an extended flexed burial - laying on the left side, oriented Nand S with the head to the south. Skeletal remains in very poor state of preservation - consists of fragments of the skull and portions of tibiae and femora. Burial is located just under sta. 4-13L7, 79" below surface, 41" below occupation level in tan sand.
Burials #27 and 28
Burial #27 is located between sta. 4-13L5 L7 and 4-17L5 L7, 75" below the surface, 41" below the bottom of the occupation level. The remains are in a very poor state of preservation, but it appears that the burial is the typical type of Bull Cr. flexed burial, with the knees drawn up toward the face. Remaining bones consist of fragments of the skull and of the tibiae and femora. Oriented SSE - NNW with head to SSE. No burial furniture.
Burial #28 is located only a foot away from #27, 80" below surface, 46" below surface - 5" in reddish brown loam. The body is laying on the back, flexed to the left - oriented ESE & WNW with the head to the ESE. Remains in very poor state of preservation - Remaining bones consisting of portion of left tibia and femur and skull with all of facial bones gone. This is a female burial. Burial furniture consists of a small amount of red pigment 12" N - 9" E 4-15L7. Find #65 on drawing - a piece of mussel shell 9": E and 16" N. #61, a piece of worked bone fragment, presumably an awl. # 62, same location as 61 and 63, a group of animal bones on top of 61 and 62; 64-A small rock, apparently the same as the paint pigment.
Burial #29 is located between 4-17L5 L7 and 4-19L5 L7, 68" below surface - 47" below occupation level, 6" in reddish clay loam. Bones in very poor state of preservation - remaining bones consist of sections of tibiae, femora and a section of the right humerus. Body oriented SE and NW, with head to SE. No trace of skull. Burial is a typical flexed type. Burial furniture consists of a ball of red paint pigment.
Burial #32 is the contracted type of flexed burial, laying on the right side, oriented Nand S with the head to the N. The remains are in a rather poor state of preservation. Remaining bones consist of portions of the skull; tibiae, femora - humeri and ulna.
Burial furniture consists of several rocks - flint chips and fragments of animal bone all placed at the neck. Body is 52" below the surface, 34" in tan sand. (Btw. sta, 4-7L9 -L7 and 4-5L9 -L7)
Oct. 14 to Nov. 11. Burial #33 was located in yellow loam between sta. 4-3L3 and 4-3L5, when lowering section for drainage. The burial is in a very poor state of preservation, laying on the right side, oriented N and S, with the head to the south. Remaining bones consist of portions of skull and teeth and the leg bones; from the position of these bones it appears that it was the contracted type of flexed burial. No burial furniture. 36" below surface, 7" in yellow clay loam, 29" below occupation level.
Nov. 11 to Dec. 9. Worked section from sta. 4-lL9 -L11 -L13 to 4-17L9 -L11 -113. Found a burial #34 between sta. 4-Ll1 and 4-L13. Worked Lll -L13 down to bottom of occupation level. Also troweled out Burial #30. Located another burial just at edge of what appears to be a house. Several finds from this section.
Burial #30 is laying on the right side, oriented NW and SE, with head to the NW. Remaining bones consist of a portion of the skull which is in a very poor state of preservation, and portions of both tibiae and femora and small fragments of the pelvis bone. Body found 46" below surface, II" below the occupation level in tan sand. The legs are flexed back under the body with the knees the farthest away from the body of any part of it. Bull Creek type #2. (at sta. 4-11L9)
A-17

4-1$1.5 'f
I
..". "
Field Drawings of Burials 27, 28, and 32. A-I8

+ 4-JL5

Field Drawings of Burials 29 and 33.

A-19

+
-I-5L//

~
1
24 \1
i~

~
J

<, ~

3/ n

i
~+

4--3/.11

+
4-17/..//

Exacf s-!-qh;..
L0 cat'-I;n /JOt-
./" I
-lIVen.
Field Drawings of Burials 34-35 and 36.
A-20

~
~
N.
J +
4-/5/.11

Burial #34 is found between sta. 4-3Lll -L13 and 4-5Lll -L13 at a depth of 37" below the surface - 16" in tan sand.
The body is laying on the left side, oriented E & W with the head to the W. The remains are in a very poor state of preservation - remaining bones consisting - skull, which is almost entirely decayed, nothing left except of portion of the right mandible with two of the jaw teeth and a portion of occipital, sections of both tibiae. Both of which are slightly higher than the head. This burial was evidently of the contracted, flexed type with the knees slightly higher than the skull and the arms going between the legs. No burial furniture.
Burial #35 is located between sra, 4-15Lll -L13 and 4-17L11 -L13. 67" below the surface, 27" below occupation level in tan sand. The body is laying on the left side, oriented E & W, with the head
toward the east.
The burial is of the contracted, flexed type with the knees drawn up toward the chest and the right arm crooked at the elbow-not going between the legs as is the usual case.
The skeletal remains are in a rather poor state of preservation, the skull is very badly decayednothing is left of the upper body except a short section of the humerus and ulnar of the right arm. The lower body consists of a few fragments of the lumbar vertebrae, most all of the pelvis and portions of both tibiae, femora and fibulae. No burial furniture associated with the burial.
Dec. 10, 1936. Troweled out Burial #35 and began work on cut from 4-11L13 -L15 to 4-27L13 -L15. Working down to occupational level to examine what appears to be a house site.
Dec. 28 to 30, 1936. Lowered section between 4-11L13 -L15 and 4-27Ll3 -L15, to the Indian occupation level - cleared whole area 4-11L9 -Ll5 to 4-27L9-Ll5. While the ova contains a very rich midden accumulation and several buried and charred areas and a few miscellaneous post mould impressions, there is no definite sign of a house site. There is quite a profusion of pottery and a few briquettes.
Began lowering section from 4-11Lll -4-15Lll to 4-11LI5-4-15LI5. Found a burial #36 between 4-11L13 -LIS and 4-13Ll3 -Ll5 with part of the body extending into the wall (L17). Began excavation of area 4-11LI5 -Ll7 to 4-13Ll5 -Ll7 and working down to burial.
Also started cleaning profile about 20' W of 4-27L15 on the west side of the R.R. fill. Worked down through yellow clay of R.R. fill and found the same component parts of the profile developments extending through to bank of river.
Jan. 7, 1937. Burial #36 found between sta, 4-11Ll5 -L17 and 4-15L15 -L17. 60" below surface, 35"
below occupation level. The body appears to have been placed on the back with the head turned to the left. The body is oriented E & W, with the head to the E. The knees were drawn straight up and are about 6" higher than the skull. Traces of the right humerus and both tibiae and the skull remain. No traces of any other bones.
Jan. 13, 1937. Burial #37 found by lowering section between 4-11L9 -LIS and Burial located between sta. 4-17L9 -L11 and 4-19L9 -Lll.
Burial is in good state of preservation. This is the contracted, flexed type of burial, oriented E & W, with the head to the W.
A-21

-_.. - 4r-~-
I .'

: I
~._..
-I- '17~ /I

i
I
I
. 4-.
-I-15t-11

-
Field Drawings of Burials 37 and 38. A-22

It is somewhat different from the previous burials, in that the body was buried face downward, the knees were drawn up evidently touching the chin and the left arm was flexed downward with the fingers on top of the right thigh, which means that the arm or hand neither went between the legs. The right arm is flexed and the right hand was at the face. The body is in a rather good state of preservation. Found 86" below the surface, 36" below the occupation level, in tan sand. No burial furniture associated with this burial.
Feb. 12, 1937. Removed overburden (R.R. fill) from remainder of area between 4-11L to 4-27L from L15 to river bank, working down to occupation level.
Located Burial #38 between sta. 4-19L17 to 4-23LI7, and 4-19L13 to 4-23L13 at a depth of below surface, 36" below occupation level. This is a double burial, both bodies were extended, lying on the right side, oriented NW and SE with the heads to the NW.
A. - This body is in a rather poor state of preservation - the skull is crushed and the parts are scattered. (Note: the arms of both bodies were folded across the waist) Remaining bones of A. consist of skull fragments - broken parts of humeri, ulna and radii, portions of both tibia, femora, and fibula.
B. - Is in a much better state of preservation. Remains consisting of skull, one side of which is rather badly decomposed - sections of both ulna -humeri and radii - both collar bones and parts of both shoulder blades - all of the vertebrae with a few of the ribs attached; part of the pelvis, both tibia femora and fibulae and most of the bones of both feet.
From the comparative size of these bodies it appears that A. was female - B. male.
No burial furniture was associated with this interment.
Feb. 15, 1937. Extension of the cut from 4-11L17 -L27 to 4-19L17 -L27, yielded Burial #39.
This burial was located between Sta. 4-21L17 and 4-21L19 at a depth of 42" below the occupation level. The body was lying on the left side, oriented SW & NE, with the head to the SW. This burial was contracted, flexed type with the legs drawn up toward the body and the head was tilted toward the knees. It was interesting to note, however, that instead of the hands going between the legs as is the usual case in this type of Bull Creek burial, the arms were flexed at the elbow and hands were at the neck. The skeletal remains were in an excellent state of preservation. Remaining bones consisted of - skull in good state. All of the cervical vertebrae, thoric vertebrae, with most of the ribs intact. None of the lumbar vertebrae left. Both humeri, ulnae and radii - all of the metacarpal bones - portions of the pelvis. Tibiae, femora, and fibulae - metatarsal bones. No burial furniture associated with the burial.
March 9, 1937. Burial #40 consists ofa few fragments of the skull found between sta. 4-9L15 -L13 and 4-11LI5 -L13 at a depth of 61" below the surface - 42" below occupation level in tan sand.
Burial #41, found between sta. 4-9L17 -L19 and 4-11LI7 -L19 at a depth of 39" below the occupation level. The body is lying on the right side, oriented E & W, with the head to the W. It is of the contracted, flexed type, but it is impossible to tell the position of the arms, as only a small portion of the left humerus remains. Remaining bones consist of parts of pelvis, tibiae, femora, fragments of skull and small parts of both clavicles. There is also a small piece of a joint of an animal bone, close to the upper part of the humerus, and two or three joints of the lumbar vertebrae. There is no burial furniture associated with this burial.
A-23

Burial #42, was found on line of sta. between sta. 4-17L19 and 4-19L19, 26" below occupation level in tan sand below surface.
The burial is of the contracted, flexed type, laying on the left side, oriented N & S., with the head to the South. The feet are gone but they must have been in contact with the Pelvis with the knees drawn up toward the chest. The left hand went between the thighs, about 2" from the pelvis and the right arm crooked downward at the elbow, the ulnar and radius being vertical. The head is turned to the left, thus making it face downward. The skull itself is in a rather poor state of preservation, the top or back of the head being rather badly crushed. On the whole, the body is in a fairly good state of preservation. Remaining bones consist of skull, vertebrae, some ribs, both clavicles, both ulna radii, humeri, both tibiae, femora and fibulae, two or three fingers and practically all of the pelvis.
From the general characteristics of the bones, it would seem that this is an adult female individual. No burial furniture associated with the burial.
Burial #44 is found between 4-19L19 -L21 and 4-21L19 -L21, at a depth of 26" below the occupation level in tan sand. Oriented E & W with head to W.
The body was buried face downward, with the legs flexed to the left. The arms were straight at the sides and the legs are laying on top of the left hand. The skull was rather badly crushed in the back, but otherwise is in an exceptional state of preservation. The remains on the whole are the best and most complete that have come from the Bull Creek Village, every part of the skeleton being perfectly preserved with the possible exception of a few finger joints, toes and one or two ribs. This was the interment of a male adult individual of middle age.
3-9 to 3-23, 1937. Moved surplus dirt on river side of cut - occupation level exposed as shown in preceding diagram (insert)
Worked section between 4-21L25 -L27 to 4-27L25 -L27, to a depth of 52" below occupation level.
Also troweled out Burials #43 -45. Found a small unbroken vessel, Find #72, at the excavation of section 4-21; - 4-23, at a depth of 27" below occupation level; below surface - 33" N. Sta. 4-17L21 on line of stks. There is no evidence of burial pit in this location and no trace of any bones, but this vessel must have been an article of burial furniture associated with a burial.
Burial #43 is of the extended, flexed type, found 50" below occupation level, on line of stats. between 4-11L21 and 4-11L23. The body was laying on the left side, oriented E & W with the head to the E. The feet are flexed back almost touching the pelvis and the knees are down and away from the body. It is impossible to tell the position of the arms, as nothing is left except traces of them. The skull has been rather badly crushed - Remaining bones consisted of skull - crushed - practically all of the vertebrae - both femora - tibiae, parts of the pelvis and sections of the clavicles.
Burial #45, was found between 4-17L21 -L23 and 4-19L2l -L23 at a depth of 33" below the occupation level. The remains were in a very poor state of preservation. All of the body gone except a portion of the skull and short sections of both tibiae and femora. The body was laying on the left side, oriented E & W, with the head to the E and from the position of the remains of the tibiae and femora it appeared that it was of the contracted, flexed type.
A-24

3-23 -3-30, 1937. Worked area from 4-13L27 - 4-21L27 to edge of river bank, 4-13L33 to 4-21L27 to a depth of 67" below occupation level at 4-13L27 and area between 4-9L21 -L25 and 4-27L21 -L25. Found a Burial #46 between sta. 4-17L23 -L25 and 4-29L23 L25.
Found a bone awl, Find #74, 10" deep in occupation level. 10" N -6" E. sta. 4-13L25. 4-1-'37. Completed excavation of area between 4-13L25 -L27 and 4-21L25 -27, with the possible exception that this will have to be lowered.
Burial #46 was found 28" below occupation level, in tan sand. Nothing remained of the body except the crown of a few teeth and short sections of both femora, tibiae. It was impossible to get any accurate picture of the burial, but it appeared that it was ofthe contracted, flexed type, laying on the left side, oriented N & S with the head to the South.
A-25

Bull Creek

.:~##~!~.:_ Qrim~i~~



2

26-27-28

W-E

R

3

33-34

S-N

L

4

SW-NE

L

5

44-45

SSE-NNW

L

6

50

NW-SE

R

7

46-47-48-49

S-N

L

8

55

E-W

L

9

56-57

SE-NW

L

10

11

SSE-NNW

L

12

62

NW-SE

R

13

S-N

L

14

58 to 61

E-W

L

15

SSE-NNW

L

16

SSW-NNE

17

NW-SE

R

18

64

SE-NW

L

19

64-66

SE-NW

R

20

67 to 69

ESE-WNW

L

21

NNW-SSE

R

22

76-77

NW-SE

R

23

24

SE-NW

L

25

NW-SE

R

26

S-N

L

C.F.

4-10; 4-10R1

4-12; 4-12Rl

C.F.

4-13; 4-14

36"

4-15; 4-16

E.F.

4-13; 4-13L1

20"

4-15; 4-15L1

C.F.

4-21L3

36"

E.F

4-17L1

47"

C.F.?

4-19L1-L3

36"

C.F.

4-19L3

44"

C.F.

4-15L1-L3

32"

4-15L1-L3

27"

4-13L1-L3

F

4-17L3-LS

38"

4-19L3-LS

C.F.

4-7;4-7L1

35"

4-11; 4-11L1

4-17; 4-18

26"

4-15; 4-16

C.F.

4-20; 4-20Rl

34"

4-18; 4-18Rl

C.F.

4-17L1

53"

4-17L3

4-13L3-LS

36"

4-15L3-LS

C.F.

4-9L3-LS

36"

4-11L3-L5

C.F.

4-24

39"

C.F.

4-24; 4-24Rl

45"

4-26; 4-26Rl

E.F.

4-13L1; L3

E.F.

4-17LS

21"

E.F.

40"

33"

C.F.?

4-11L9-L11

28"

4-13L9-L11

C.F.

4-11L9-L11

21 "

4-13L9-L11

E.F.

4-13L9

41 "

A-26

-27
28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

SSE-NNW

2[. - C.F.

4-13L5-L7

41 "

40 17L5-L7

ESE-WNW back flexed

C.F.

4-15L5-L7

46"

to L

4-17L5-L7

78-79

SE-NW

L?

C.F.

4-1715-17

47"

4-1915-17

NW-SE

R

E.F.

4-11L9

11"

82-83-84

NW-SE Triple burial

4-9L7-L9

23"

4-11L7-L9

87-88

N-S

R

C.F.

4-5L7-L9

34"

4-7L7-L9

85-86

S-N

R

C.F.

4-3L5

20"

W-E

L

4-3L11-L13

16

4-5L11-L13

89-90

E-W

L

C.F.

4-15L11-L13

27"

4-17L11OL13

91-92

E-W

L

4-11L15-L17

35"

40 15L15-L17

93-94

W-E

Face

C.F.

4-17L9-L11

38"

downward

4-19L9-L11

95-96-97

NW-SE

R

Double 4-19L13-L17

36"

burial

4-23L13-L17

98-99

SW-NE

L

C.F.

4-21L17

42"

4-21L19

100

Skull only 4-9L13-L15

42"

101

W-E

R

C.F.

4-9L17-L19

39"

4-11L17-L19

102-103

S-N

L

C.F.

4-17L19

26"

4-19L19

E-W

L

E.F.

4-11L21

50"

4-11L23

104-105-106

W-E

face down Semi C.F. 4-19L19-L21

26"

flexed to L

4-21L19-L21

E-W

L

C.F.

4-17L21-L23

33"

4-19L21-L23

No Picture

S-N

L?

C.F.?

4-17L23-L25

28"

4-19L230L25

A-27

Bull Creek
Skull Only Contracted flexed Contracted flexed Folded feet back to
pelvis flexed Contracted flexed Folded feet back to
pelvis flexed
Contracted flexed Contracted flexed Contracted flexed
Skull only Folded feet back to pelvis flexed (Female) Contracted flexed Near #3 (Below tree) (Old Man) contracted
flexed Contracted flexed
(West of #6) Double burial Contracted flexed Contracted flexed
Contracted more than any found yet - knees
on chest (Female) Folded feet to pelvis -
flexed Folded feet to pelvis -
flexed Flexed Fragmentary burial Contracted flexed Folded flexed type Extended flexed Contracted flexed

~4ti~~ ;;]~epth~~i6W~tf~~~

1

2

67"

3

67"

4

44"

No furniture No furniture (1) effigy pot and pierced seed beads No furniture

5

72"

Greenstone celt 7" long

6

78"

Greenstone celt 4" long, 2 small greenstone

chisels, several pieces worked flint, black

font knife, red pigment

7

78"

(2) effigy pot, discoidal stone

8

68"

No furniture

9

66"

Greenstone chisel 31h" long

10

54"

No furniture

11

73"

No furniture

12

69"

13

40"

14

51"

No furniture No furniture

15

78"

No furniture

16

(3) effigy pot

17

18

61"

Small bowl 31h" in diameter, several large

beads of shell, piece of conch shell core,

clay pipe human effigy handled type

19

67"

No furniture

20

21

36"

22

54"

23

57"

24

59"

25

52"

26

79"

27

75"

String of seed beads
No furniture
No furniture No furniture No furniture No furniture No furniture No furniture

A-28

Flexed

28

Contracted flexed

29

Flexed - with knees

30

farthest away from

body

Trible burial - first of

31

this sort found - 2

bodies contracted

flexed, other only the

skull

Contracted flexed

32

Contracted flexed

33

80"

Red pigment & small stoned, mussel shell,

evidently bone awl & other pieces of animal

bone

88"

Ball of red pigment

49"

No furniture

52"

Rocks, flint chips, fragments animal bones,

grouped around neck

36"

Rocks, flint chips, fragments animal bones,

grouped around neck

Bull Creek #1
Finds

Find #1 (39-22198)

3-13-36

Rimsherd with raised, notched design, about Ih" below rim. Found 24" N., 23" W. Sta. 1-46;

8" below surface, I" in black midden deposit.

Find #2 (39-22199)

3-14-36

A piece of a broken arrow-point with haft type end. Found 18" N. 8" E. Sta. 1-47. Ph" deep

in greyish sand, midden deposit, 11" below surface.

Find #3 (39-22200)

3-14-36

Small piece of rimsherd, showing raised, notched design, about 1/4" below rim; Found 14" S.,;

4" W. Sta. 1-48; 14" below surface, 5" in greyish sand, midden accumulation.

Find #4 (39-22201)

3-16-36

Peculiar type of rimsherd, showing notched design on top of rim and incised and punctate design

on the body of the pot. Found 22" N -12" W. Sta. 1-52; 6" deep in yellow sandy clay topsoil.

Find #5 (39-22202)

3-17-36

A quartz arrow point - imperfectly made, with portion of the point section broken off. Found

18" S. Sta. 1-56. 23" below surface, 3" deep, in black midden deposit.

Find #6 (39-22203)

3-17-36

Piece of broken flint, showing secondary working, possibly a portion of a scraper or chisel.

Found 6" N.; 16" E. Sta. 1-49; 13" below surface, 3" deep in reddish sand clay.

Find #7 (39-22204)

3-18-36

A small broken piece of stone, elliptical in shape with a hole drilled through it. Found 22" S;

26" E. Sta. 1-49; 16" below surface, 6" deep in house debris.

A-29

Find #8 (39-22205)

3-20-36

Small fragment of rimsherd with small handle attached Found 29" N., 4" E. Sta. 1-47LI; 12"

below surface, 7" deep in black midden deposit.

Find #8A (39-22206)

3-23-36

Small rimsherd with handle, found 15" below surface, 10 in black midden deposit. 22" S; 30"

W Sta. 1-49.

Find #9 (missing at time-no Park Service Number assigned)

3-21-36

A moulded clay briquette, showing timber mould and use of vegetable fiber, as tempering agent,

found resting on floor of house, 41" S. 5" E. Sta. I-52, 26" below surface, 5" deep in house debris.

Find #10 (39-22207)

3-21-36

A piece of broken celt or chisel, found 3" S., 23" W. Sta. 1-53. 18" below surface, 7" in black

midden deposit.

Find #11 (39-22208)

3-26-36

A hafted scraper of quartzite - fair workmanship of material. Found directly under Sta. 1-47L1,

8" below surface, in black midden deposit.

Find #12 (39-22209)

4-6-36

Quartz arrowhead, spinner type, found 2" deep in yellow topsoil, in 5' exploration trench, 36"

E, 18" S. Sta. i-n n.i.

Find #13 (39-22210)

4-8-36

Piece of broken knife or spear-head, found 4" deep in black midden deposit, 12" below surface.

31" e., 17" S. Sta. 1-107Ll.

Find #14 (39-22211)

4-8-36

Quartzite arrowhead, very poor workmanship, found 8" deep in black midden deposit, 13" below

surface. 46" S, 7" E Sta. 1-IIlLI.

Find #15 (39-22212)

4-8-36

Chunky stone of granite type rock, found 6" deep in black midden deposit, 9" below surface.

36" N., 12" E. Sta. 1-91Ll.

Find #16 (missing at time-no Park Service Number assigned)

4-8-36

Pieces of broken pot, found massed together, 13" deep in black midden deposit. 19" below

surface, 35" E., 4" S. Sta. 1-95Ll.

Find #17 (39-22213)

4-8-36

Portion of rimsherd, with very small handle attached; found 9" deep, in black midden deposit.

12" below surface, 18" E. - 18" S. Sta. 1-91Ll.

Find #18 (39-22214)

4-13-36

Small piece of worked flint, possibly used as an engraving tool. Found in 5 ft test trench, 7"

below surface, in black humus, 7" E., 9" N. Sta. 1-103L5.

Find #19 (39-22215)

4-13-36

A-30

Portion of a rimsherd, made in effigy form, of a human face. Found in black humus, 3" below surface. Found 52" N., 31" W. Sta. 1-95L1.

Find #20 (39-22216)

4-13-36

Chunky stone of granite type rock, rather poor workmanship. Found 4" deep in black midden

deposit, 5" below surface. 27" S.; 42:E. Sta. 1-103L5.

Find #21 (missing at time-no Park Service Number assigned)

4-14-36

Large pieces of two broken pots, found in Pit #1, 16" to 18" below surface, in pit fill. 42" S.;

31" E. Sta. 1-119L13.

Find #22 (39-22217)

4-16-36

A piece of worked quartz - knife or side scraper; found 3" deep in yellow topsoil. 39" S. - 52"

W. Sta. 1-99L1.

Find #23 (39-22218)

4-15-36

Broken piece of quartz arrowhead, imperfectly worked. Found 2" below surface, in black

humus. 8" S. - 52" W. Sta. 1-99L1.

Find #24 (39-22219)

4-15-26

A small quartz scraper, found 6" deep in black midden deposit, 9" below surface. 5" N. - 25"

W. Sta. 1-107Ll.

Find #25 (39-22220)

4-15-36

Broken arrowhead or scraper, poor workmanship, found 8" deep in black midden deposit, just

above house debris, 15" below surface, 22" S. - 2" W. Sta. 1-99L1.

Find #26 (39-22221)

4-16-36

Small pottery chunky stone, found 7" below surface, in black humus. 15" S. - 28" W. Sta. 1-

95Ll.

Find #27 (39-22222)

4-16-36

Small spinner type arrowhead, excellent workmanship. Point of bluish flint. Found 9" below

surface, 4" in black midden deposit, on level of house debris. 43" S. - " E. Sta. 1-99L5.

Find #28 (39-22223)

4-20-36

Broken quartz arrowhead or scraper, imperfectly worked, probably a reject. Found 13" below

surface, 7" deep in black midden deposit. 22" N., 12" E. Sta. 1-91.

Find #29 (39-22224)

4-20-36

A portion of what appears to be a bird effigy of some type. Seemingly a portion of the rim of

a pot. Found 16" below surface, 10" in black midden, directly under Sta. 1-94.

Find #30 (39-22225)

4-20-36

A small pottery chunky stone, found 10" below surface. 5" deep in black midden deposit. 11"

E., 17" N. Sta. 1-99L7.

Find #31 (missing at time-no Park Service Number assigned)

4-20-36

Several large pieces of one pot - plain, with beaded rim design, found 13" below surface, 7th "

in black midden deposit, in what appears to be a small pit. 20" W. - 9" N. Sta. 1-93L5.

A-31

Find #32 (39-22226)

4-20-36

Flint scraper, very excellent workmanship, found 13" below surface, 7 1h" in black midden

deposit. 20" W. - 9" N. Sta. 1-93L5, in what appears to be a small pit.

Find #33 (39-22227)

4-14-36

Broken hafted scraper or knife (flint), found 41" N. - 38" E. Sta, 1-119L13, 3" below surface

in black humus.

Find #34 (39-22228)

4-21-36

Small metal sewing thimble, found 12" below surface, 4" deep in black midden deposit, in what

appears to be slump, near edge of bank. 16" W. - 8" S. Sta. 1-100.

Find #35 (39-22229)

4-22-36

Small flint arrowhead, found 8" below surface, 4" in black midden deposit, 26" N. -15" W. Sta.

1-95L7 on west edge of bank.

Find #36 (39-22230)

4-22-36

Quartzite hand-chopper, found 5" S. Sta. 2-70 on line of stas., 4" below surface in topsoil.

Find #37 (39-22231)

5-14-36

Broken pieces of a pot, found on floor level of house site #2. Bowl and rim found 42" E. - 7"

N. Sta. 1-95L1. Bottom found 32" E. - 14" S. 1-95L1. Pot approximately 9" in diameter, plain design,

with fluted rim.

Find #37A (39-22232)

4-29-36

Small arrowhead, found 11" below surface, 6" deep in midden deposit, 32" W - 12" S. Sta. 1-

99L5.

Find #38 (39-22233)

5-14-36

Broken pieces of a large pot, notched rim, incised and punctated design on body of pot. Pieces

found grouped together on floor level of house site #2, between Sta. 1-97Ll -L3 and 1-99L1 -L3.

Find #38A (39-22234)

5-25-36

Flint arrowhead, poor workmanship, found 12" below surface, 8" deep in midden deposit, 18"

S. - 10" E. Sta. 1-41.

Find #38B (39-22235)

5-11-36

Small quartzite arrowhead, found 27" N. - 40" E. Sta. 1-119L9, 10" below surface in black

midden deposit.

Find #39 (missing at time-no Park Service Number assigned)

5-14-36

Broken pieces of a small pot (approx. 9" dia.) resting on floor level of house site #2, 7" W - 31"

S Sta. 1-99L3. Pot has notched rim and plain body. Body notched due to firing.

Find #40 (39-22236)

5-14-36

Several pieces of a large pot found on floor level of house site #2. Found 24" S. - 18" W. Sta.

1-99L3, 13" below surface.

A-32

Find #41 (missing at time-no Park Service Number assigned)

5-14-36

Portion of broken rim of large pot (13" dia.). Found resting on floor level of house site #2, 13"

below surface - 8" W. Sta. 1-99L3. Pot shows painted or beaded rim and circular stamped design.

Find #42 (missing at time-no Park Service Number assigned)

5-14-36

Portion of rim of large pot (I3 "dia.). Found resting on floor level of house site # 2-13" below

surface, 13" S. 18" W. of Sta. 1-99LE.

Find #43 (39-22237)

4-21-36

Hafted type flint scraper, rather poor workmanship. Found 5" below surface, in humus. 451h"

N. Sta. 1-135L9 on line of stakes.

Find #44 (39-22238)

6-17-36

Small pottery chunky stone. Found 6" N. 6" E. Sta. 4-19L3. 25" below surface, 11" in tan

sand.

Find #45 (this is note on sheet in upper left corner) Small bead

Find #46 (39-22239)

7-17-36

Rimsherd, plain design, with small type handle. Found 3" deep in black occupation level, 30"

below surface, 10" N. - 26" E. Sta. 4-19L5.

Find #47 (39-22240)

7-17-36

Small cylinder of baked clay. Found 36" E. Sta. 4-19L5 on line of station, 36" below surface,

9" deep in occupation level.

Find #48 (missing at time-no Park Service Number assigned)

7-17-36

Small briquette showing end of reed imprint, found 53" N. - 9" E. Sta. 4-19L5. 33" below

surface, 7" deep in occupation level.

Find #49 (39-22241)

7-20-36

Clay bead, found 5" deep in tan sand, 40" below surface, 11" E. Sta. 4-18, on line of sta.

Find #50 (39-22242)

7-20-36

Rimsherd with unusual flanged design on neck of pot. Found 6" deep in tan sand, 40" below

surface, 31" S. Sta. 4-18.

Find #51 (missing at time-no Park Service Number assigned)

7-30-36

Small greenstone celt, 4" long, 2Vz" at cutting end, 1" at butt end, found as.part of burial #6.

Found 67Jh" below surface, 35" in tan sand, 7" S. 7" E. Sta. 4-17Ll.

Find #52 (39-22243)

7-30-36

Small greenstone chisel found 671h: below surface, 35" in tan sand, 17" N. 8" E. Sta. 4-17Ll

part of burial furniture with Sta. burial #6.

Find #53 (missing at time-no Park Service Number assigned)

7-30-36

Small greenstone chisel found as part of burial furniture with Burial #6. Found 671h" below

surface, 35" in tan sand, 26" N. 5" E. Sta. 4-17L1.

A-33

APPENDIX B WPA Coding Sheets

PART OF POT REPRESENTED

1. Sherd from Undet. portion of pot 2. Rimsherd 3. Rim and shoulder or neck 4. Rim and handle or lug 5. Rim and body 6. Rim, Body, and Base

7. Body 8. Body and Base 9. Base 10. Handle 11. Whole Pot 12. Undetermined

TEMPER Material

1.

Shell

2.

Sand

3.

Crushed Rock or Grit

4.

Vegetal

5.

Potsherd

6.

Ground Micaceous stone

7.

Hole Temper

SIZE

A.

Large

B.

Medium

C.

Small

CROSS SECTIONAL APPEARANCE OF PASTE STRUCTURE

1.

Fine

2.

Medium fine

3.

Medium

4.

Medium Coarse

5.

Coarse

6.

Granular

7.

Laminated

8.

Contorted

9.

Compact

Criteria (1.
(2.
(3.

Size of Temper Distribution Relatively homogeneity of paste

HARDNESS

22-2.52.533.5-

Gypsum Fingernail Cryolite Calcite Celestite

4-

Flourite

4.5- Chabazite

5-

Apatite

6.- Adularia

SURFACE FINISH - Inner or Outer Surfaces

1. Very rough - (1. From extrusion of temper or lumpish granular exterior knobby)

2. Rough

(2. From pitting, holeing, intaglic effects from absorption or dissolution of temper

or paste)

3. Smooth

(3. Note by extent to which smoothing process is evidenced by mechanical striation

from the pottery smoother)

4. Glossy

5. Polished

6. Vermiculated - Theta, Type sherd EXECUTION of design

7. Glistening -finish- pseudo glaze

1. Crude

2. Fair

3. Good

4. Excellent

5. Unobserved

AREA OF DECORATION
1. Inner Rim 2. Lip 3. Outer Rim 4. Shoulder

5. Body 6. Undetermined 7. Base
x. Denotes design on inside of sherd

DECORATION A. - POSSIBLY related technologically to surface finish.

Cord Marking

1. Irregular Criss-Cross 2. Checker Board 3. Diamond 4. Straight line series or parallel patterns 5. Single strands occur at sparse intervals 6. Undetermined

SMOOTHED-BURNISHED

COMBED, BRUSHING

1. Combing - Definition 2. Brushing - Definition 3 Scratching - Definition

V. Decoration possibly applies with Grass

Phi - Stoneware

Theta - Grass Impressions

Mottling from Firing

X. Denotes design or Execution on inside of sherd.

DECORATION B. - Specialized stamped or paddle marked technique.

Alpha- Large residual class of general stamped or paddle marked sherds in which specific designs are apparent.

1

6

2

7

3

8

4

9

5

10

Sigma - A general class of crudely stamped or impressed sherds characterized by irregular linear or crisscross impressions whose exact technological origin is not known, but which give a taxonomic resemblance when regarded as a whole.
Delta - A fine line cameo effect pattern (Possibly a large and small division).
Pi 1. Checkerboard design squares 2. Rectangular 3. Diamond
a. Large b. Medium c. Small

Decoration C
Technique a. Incising
1. Narrow 2. Medium 3. Wide
b. Trailing 1. Medium 2. Wide 3. Very ?Wide
c. Engraving
d. Punctate a. Small b. Medium c. Large
(Punctate Design) 1. Solid rounded 2. Cylinder 3. Crescent 4. Stippling 5. Triangular 6. Rectangular 7. Bi-sected cones 8. Undetermined
T.N. Thumb nail
R. Roulette

DESIGN
RECTILLINEER
1. Vertical 2. Rectangles 3. Horizontal 4. diagonal or oblique 5. triangle herringbone 6. cross hatching 7. Zigzag 8. Undetermined 9. Triangle
CURVILINEER
1. Concentric Circles 2. Scroll 3. Guilloche 4. Undetermined

Nodes

1. Extruded 2. Applique 3. Moulded

F. Filming

1

2

3

4

1. Painting Red Yellow orange and black

2. Application of slip

a. put on before firing

b. Put on after firing

Clay-wash; not the same as definite slip preparation

Modeling 1. Fish 2. Frog 3. Owl 4. Human 5. Cat 6. Snake 7. Eagle
1. Fabric Impressed

8. Geometric 9. Naturalistic 10. Whole pot 11. Rim 12. Body 13. Bass 14. Lip 15. Bird
Fabric Impressions 1. plain plaiting 2. plain twining - wicker 3. twilled twining 4. zigzag twilled twining 5. Chevron 6. undetermined

DECORATION D - SECONDARY FEATURES

L.H. Loop Handles a. Attached to lip b. Attached below lip c. Project above lip d. Lateral width e. Thickness f. Height g. Handle projection h. Button knob i. Bifurcate teats or knobs at apex (triple knobs at apex) j. Ribbed or single rib handle k. Nodes on middle handles 1. Depression at handle apex m. Rim raised at handle n. Moulded at base o. Riveted at base p. Handle

S.H. Strap Handles a. Attached to lip b. Attached below lip c. Project above lip d. Lateral width e. Thickness f. Height g. Handle projection h. Moulded at base i. Riveted at base

L. G. Lug Handles a. Length b. Height c. Thickness d. Attached below lip e. Attached at lip f. Attached at shoulder g. Rounded h. Pointed or knob? i. Elongate or flange

R.S. Applique Rim Strip (Rim moulding or luted rim strip or band)

1. Moulded to or joined with lip 2. Moulded to or joined below lip c. Teats or nodes d. Beaded e. Pinched? Indented? f. Medium thumbing or swelling g. Cylinder impressions h. Bi-sected impressions i. Rounded band at lip

j. Crescent k. Stamped 1. Scalloped m. Sigma p. Plain

GENERAL VESSEL MORPHOLOGY

Lip 1. Pointed 2. Narrow and Rounded 3. Flattened a. Flat horizontal b. Flat slopes in c. Flat slopes out 4. Everted a. In b. Out c. Both sides 5. Folded 6. Rounded

LIP DECORA nON 1. Scalloped 2. Notched 3. Parallel grooved 4. Incised 5. Punctate 6. Cylinder Impression 7. Undetermined

RIM SHAPE Pot -

A vessel with curvilinear sides and partially constricted orifice. 1. Straight 2. Slightly flaring 3. Flaring 4. Pointed 5 . Incurving 6. Outcurving rim

Beaker Beaker proper Straight side cup

Bottle Globulat body cylinderical neck

Bowl

A. Straight B. Incurving rim C. Cazuela D. Outcurving rim

Plate

Straight rim Outcurving rim

Dish Pans

Base 1. Conoidal 2. Oval 3. Globular 4. Truncate 5. Flattened 6. Annular

7. Outer Base Legs. a. Three legs b. Four legs
8. Inner Legs a. Three legs b. Four legs
9. Footed
Size 1. Lip - thickness 2. Rim - thickness 3. Shoulder - thickness 4. Body - thickness 5. Rim - height 6. Lip diameter 7. Shoulder diameter or greatest width of body 8. Base diameter (if flattened) 9. Height 10. Neck Diameter 11. Neck Height 12. Undetermined 13. Base Thickness

Execution of Design

Excellent

4

Crude

5

Fair

2

Good

1

Unobserved

5

The Quality of the ware

1. Tenacity - Does pottery hold together or is it brittle, flexible, easily broken? 2. Porosity - is it porous; does it show crackling markedly? 3. Firing - Are firing effects evenly distributed or does ware show in cross section areas showing
greater or less effects of heat? 4. Surface finish - Crude, rough finish show poorer technique in pottery manufacturing. 5. Relation of mass to tensile strength
1. Excellent 2. Good or fair 3. Inferior 5. Very crude

APPENDIX C Bull Creek Vessel Form Data

..
..

2

Schnell 1959

23

Fuller 1981

26

WPA

28

WPA

36

WPA

41

WPA

42

WPA

43

WPA

44

WPA

45

WPA

46

WPA

48

WPA

50

Kelly 1950

51

Kelly 1950

53

Kelly 1950

55

Kelly 1950

69

WPA

70

WPA

71

WPA

72

WPA

73

WPA

74

WPA

75

WPA

77

WPA

79

WPA

82

Kelly 1950

85

Kelly 1950

92

Kelly 1950



32 em 18 em 14 em 40 em 32 em 36 em 34 em 42 em 32 em 36 cm 36 em 38 em 40 em 28 cm 44 cm 28 em 14 cm 26 em 36 em 32 em 26 em 34 cm 28 em 22 em 16 em 16 em 22 cm 26 em 22 em

Applied Strip (Pinched) No Modification
Folded Pinched (13 mm) No Modification
Folded Pinched (20 mm) Applied Strip (Pinched)
No Modification Folded Pinched (25 mm) Applied Strip (Pinched) Applied Strip (Pinched) Folded Pinched (24 mm) Folded Pinched (26 mm) Folded Pinched (15 mm) Folded Pinched (21 mm) Folded Pinched (20 mm) Folded Pinched (12 mm)
Applied Strip (Noded) Applied Strip (Noded)
No Modification Applied Strip (Noded) Applied Strip (Noded) Applied Strip (Noded) Folded Pinched (14 mm) Folded Pinched (12 mm)
No Modification Noded
Folded Pinched (16 mm) No Modification No Modification

Complicated Stamped Complicated Stamped Complicated Stamped
Plain Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped
Plain Complicated Stamped
Plain Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped
Plain Plain Plain Plain Complicated Stamped Complicated Stamped

><

00

Li,
.../: .:

"lil'!!11

97

Kelly 1950

98

Kelly 1950

99

Kelly 1950

100

WPA

101

WPA

102

WPA

103

WPA

104

WPA

113

WPA

120

WPA

121

WPA

122

WPA

123

WPA

124

WPA

125

WPA

126

WPA

127

WPA

128

WPA

129

WPA

130

WPA

131

WPA

132

WPA

133

WPA

134

WPA

135

WPA

145

Schnell 1959

147

Schnell 1959

150

Schnell 1959

154

Schnell 1959

~
34 cm 32 em 36 em 42 em 32 em 38 em 32 em 26 em 26 em 38 em 28 em 32 em 20 em 24 em 36 em 34 em 34 em 34 em 36 em 36 em 34 em 36 em 30 em 32 em 30 em 40 em 34 em 36 em 14 em

/:11

) ~

~I-i-- ill ill

~)

...4t1
<

Folded Pinched (19 mm) Complicated Stamped

Folded Pinched (16 mm) Complicated Stamped

Folded Pinched (23 mm) Complicated Stamped

Folded Pinched (20 mm) Complicated Stamped

Folded Pinched (27 mm) Complicated Stamped

Folded Pinched (14 mm) Complicated Stamped

Noded

Plain

Noded

Plain

Noded/Scalloped

Plain

Applied Strip (Noded) Complicated Stamped

Folded Pinched (20 mm) Complicated Stamped

Folded Pinched (18 mm) Complicated Stamped

Folded Pinched (16 mm) Complicated Stamped

Folded Pinched (18 mm) Complicated Stamped

Folded Pinched (26 mm) Complicated Stamped

Folded Pinched (16 mm) Complicated Stamped

Folded Pinched (16 mm) Complicated Stamped

Folded Pinched (14 mm) Complicated Stamped

Folded Pinched (20 mm) Complicated Stamped

Folded Pinched (15 mm) Complicated Stamped

Folded Pinched (22 mm) Complicated Stamped

Folded Pinched (22 mm) Complicated Stamped

Folded Pinched (21 mm)

Plain

Folded Pinched (23 mm)

Plain

Folded Pinched (18 mm)

Plain

Folded Pinched (25 mm) Complicated Stamped

Folded Pinched (19 mm) Complicated Stamped

Folded Pinched (20 mm) Complicated Stamped

No Modification

Plain



< i 0 /..

> .......

.. i


<

155

Schnell 1959

160

Schnell 1959

161

Schnell 1959

166

Schnell 1959

168

Schnell 1959

171

Schnell 1959

172

Schnell 1959

173

Schnell 1959

176

Schnell 1959

177

Schnell 1959

181

Schnell 1959

183

Schnell 1959

184

Schnell 1959

186

Schnell 1959

187

Schnell 1959

191

Schnell 1959

192

Schnell 1959

193

Schnell 1959

194

Schnell 1959

207

Fuller 1981

208

Fuller 1981

209

Fuller 1981

211

SAS 1993

213

SAS 1993

214

SAS 1993







<

B.

}

28 em 26 em 38 em 24 em 26 em 36 em 40 cm 40 em 18 em 40 cm 34 em 40 em 32 cm 40 em 38 em 42 em 34 em 34 cm 34 em 32 em 24 em 42 em 40 em 32 em 30 em

Applied Strip (Noded) Folded Pinched (17 rom) Folded Pinched (22 rom)
Noded No Modification Folded Pinched (15 rom) Folded Pinched (20 rom) Folded Pinched (25 rom) Folded Pinched (16 rom) Applied Strip (Noded) Applied Strip (Pinched)
Lug Handles Folded Pinched (25 rom)
Noded No Modification Applied Strip (Noded) Applied Strip (Noded) Folded Pinched (15 rom) Folded Pinched (24 rom) Folded Pinched (22 rom) Folded Pinched (16 rom) Folded Pinched (18 rom) Folded Pinched (20 rom) Folded Pinched (25 rom) No Modification

Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped
Plain Plain Check Stamped Complicated Stamped Plain Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped

n >/ < ! it~;;11

a',~Ci(Cd u-ddc
I)i~

~()Wl: of _ > <

> /<1



3

Schnell 1959

4

Schnell 1959

10

WPA

11

WPA

35

WPA

49

Kelly 1950

56

Kelly 1950

57

Kelly 1950

83

Schnell 1959

84

WPA

91

WPA

106

WPA

114

WPA

115

WPA

138

WPA

142

WPA

146

Schnell 1959

148

Schnell 1959

149

Schnell 1959

156

Schnell 1959

157

Schnell 1959

163

Schnell 1959

164

Schnell 1959

167

Schnell 1959

169

Schnell 1959

179

Schnell 1959

180

Schnell 1959

196

Fuller 1981

200

Fuller 1981

24 em 24 en 14 em 16 em 24 em 30 em 36 em 28 em 32 em 26 cm 24 em 24 em 26 em 22 em 32 em 18 em 40 em 30 em 30 em 36 em 38 em 32 em 30 em 32 em 40 em 30 em 34 em 18 em 24 em

Notched Notched
Plain Notched Notched Incised
Plain Notched
Plain Notched Notched Notched Notched Incised
Plain Plain Incised Plain Plain Notched Plain Notched Notched Plain Plain Plain Notched Plain Plain

;

%1l

illl
Incised/Punctate

Incised/Punctate

Incised

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised

Incised

Incised

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised

Incised/Punctate

Incised/Punctate

Incised

Incised

Incised/Punctate

Incised

Incised

II. .
.
201 202 203 204 205 206 212

Fuller 1981 Fuller 1981 Fuller 1981 Fuller 1981 Fuller 1981 Fuller 1981 SAS 1993

hil

. .. .

, ... rUU1..

..

32 em 30 em 24 em 28 em 34 em 26 cm 36 em

Notched Plain
Notched Plain
Notched Plain Plain

Plain Incised/Punctate Incised/Punctate Incised/Punctate Incised/Punctate Incised/Punctate
Incised

.....:

.

. .

...

7 20
... .... . .
5 6 12 14 15 17 18 19 24 25 27

WPA
..
Schnell 1959 WPA
Kelly 1950 WPA WPA
Schnell 1959 Fuller 1981
WPA WPA WPA WPA

...

. ...

.:
.

36 em 26 cm

Rolled Plain

.1

32 em 32 em 20 em 40 em 36 em 24 em 36 em 36 em 42 em 36 em 38 em

Applied Strip (Notched) Plain Plain Noded Plain
Notched Effigy Adorno
Notched Notched
Plain Plain

Check Stamped Plain
=
- Plain Incised Plain Plain Incised/Punctate Complicated Stamped
Incised Complicated Stamped Complicated Stamped
Incised/Punctate Complicated Stamped

TIl fJIii i <I

28

WPA

30

WPA

31

WPA

32

WPA

33

WPA

34

WPA

37

WPA

39

WPA

40

WPA

47

WPA

52

Kelly 1950

54

Kelly 1950

58

Kelly 1950

59

WPA

60

WPA

61

WPA

62

WPA

63

WPA

64

WPA

65

WPA

66

WPA

67

WPA

68

WPA

76

WPA

78

WPA

80

WPA

81

WPA

86

WPA

87

WPA

III !II _ I siriM ~lii@iillliiiWj.iitll

l. illf~~t~.

~ti .....

30 em 26 em lOcm 22 em 38 em 40 em 34 em 26 em 30 em 28 em 26 em 24 em 26 em 34 em 40 em 36 em 40 em 32 em 34 cm 32 em 26 em 36 cm 30 em 22 em 34cm 28 em 32 em 30 em 32 em

Notched Notched
Plain Plain Notched Notched Fold Plain Notched Notched Folded Pinched Rolled Plain Plain Notched Plain Plain Applied Strip (Noded) Plain Folded Notched Plain Notched Notched Notched Notched Notched Notched Plain Plain Plain

Complicated Stamped Complicated Stamped
Plain Plain Complicated Stamped Plain Plain Complicated Stamped Complicated Stamped Complicated Stamped Incised Plain Complicated Stamped Complicated Stamped Plain Complicated Stamped Plain Plain Complicated Stamped Plain Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Plain Plain Plain Complicated Stamped Complicated Stamped

.
88 89 90 93 94 95
96
105 107 111 112 116 117 136 137 139 140 141
143 144 151 152 153 158 159 162 165 170 174

.' .. '... .. .' I

WPA WPA WPA WPA WPA WPA WPA Kelly 1950 Kelly 1950 Kelly 1950 WPA Kelly 1950 Kelly 1950 WPA WPA WPA WPA WPA Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959

32 em 28 cm 16 em 30 em 36 em 30 em 22 em 24 em 30 cm 42 em 28 cm 36 em 28 em 22 em 26 cm 10 em 34 cm 40 em 36 em 28 cm 28 em 38 cm 36 em 36 em 32 em 18 cm 32 em 24 em 18 cm

. .
.

Notched Notched Rolled
Plain Folded Pinched
Plain Notched
Plain Plain Thickened Effigy Adorno Notched Plain Notched Plain Plain Plain Plain Folded Plain Plain Plain Plain Pinched Thickened Applied Strip (Notched) Notched Plain Rolled

Complicated Stamped Complicated Stamped
Plain Incised Complicated Stamped Plain Complicated Stamped Incised Incised Plain Plain Complicated Stamped Incised Complicated Stamped Plain Plain Complicated Stamped Plain Incised Plain Complicated Stamped Plain Plain Complicated Stamped Plain Plain Complicated Stamped Complicated Stamped Plain

II

J2"
.......
1111

........

....... r



l36



I>



178

Schnell 1959

182

Schnell 1959

185

Schnell 1959

188

Schnell 1959

190

Schnell 1959

195

Fuller 1981

197

Fuller 1981

198

Fuller 1981

199

Fuller 1981

210

WPA

215

SAS 1993

216

WPA

32 em 34 em 28 em 18 em 22 em 14 em 18 em 20 em 36 em 22 em 30 em 28 em

Plain Noded Folded Pinched (16 mm) Pinched Plain Incised Folded Notched Folded T-rim Folded Notched Plain

Complicated Stamped Check Stamped
Complicated Stamped Complicated Stamped
Plain Incised Incised Plain Incised Incised Plain Plain

8

WPA

9

WPA

13

WPA

16

WPA

21

Fuller 1981

22

Fuller 1981

38

WPA

108

Kelly 1950

109

Kelly 1950

110

Kelly 1950

16 em 24 em 20 em 10 em 10 em 14 em 18 em 34 em 34 em 36 em

L-shaped Plain
Lug Handles Incised Plain Plain Plain Notched Plain
Thickened

Incised Complicated Stamped
Plain Incised/Punctate Complicated Stamped
Plain Plain Complicated Stamped Plain Plain

..
.. \/ '/.

/.' i} \ '.....'. ,.....

118

Kelly 1950

119

WPA

175

Schnell 1959

189

Schnell 1959

2.~111
42 em 34 em 38 em 22 em

.. '.

. <: "
i
L-shaped Plain
Thickened Thickened

\
.,......
i<

" i
<<:

Complicated Stamped

Plain

Plain

Complicated Stamped

< '.'.'...,.
:S~tl

,i/
,
/
<


i\\
i

217

WPA Burial 3

218

WPA Burial 7

219

WPA Burial 16

i' lily .......

<: <:
'. i< <.i..>.. '.\

<:

9.5 em 9.5 em 8.0 em

30 em long, 26.5 em high 27.5 em long, 27 em high 28 em long, 32 em high

,

Negative Painted Negative Painted
Red?

\:T 1-\
Tl00 i5
/VII ;(1"11
i3~

THE BULL CREEK SITE, 9MEl MUSCOGEE COUNTY, GEORGIA
R. Jerald Ledbetter

Georgia Department of Transportation Office of Environment/Location Atlanta, Georgia
OCCASIONAL PAPERS IN CULTURAL RESOURCE MANAGEMENT #9 1997

Georgia Department of Transportation Occasional Papers in Cultural Resource Management

The Georgia Department of Transportation's (DOT) Occasional Papers in Cultural Resource Management series consists of archaeological research sponsored by the Georgia DOT. These reports have been produced by Georgia DOT in-house archaeological staff and by archaeological consultants under contract with the Georgia DOT. Each report within the series presents information about specific archaeological sites considered eligible for the National Register of Historic Places that would be affected by transportation projects.

Each report addresses research questions included in the Scope of Work for each project and the conclusions and interpretations contained therein reflect the theoretical orientation, background, and assorted biases of the authors. Each manuscript has been prepared as a result of a contract with Georgia DOT. The reports are distributed by the Office of Environment/Location, Georgia Department of Transportation.

For a copy of any or all of the reports, please indicate the specific report; there may be a minimal charge for copying if the report is out of print.

Georgia Department of Transportation Occasional Papers in Cultural Resource Management

No. I - Archaeological Investigations at 9CK(DOT)7, Cherokee County, Georgia; by WilliamR. Bowen (1982).

No. 2 - Cagle Site Report: Archaic and Early Woodland Period Manifestations in the North Georgia Piedmont; by Morgan R. Crook, Jr. (1984).

No.3 - Lowe Site Report: A Contribution to Archaeology of the Georgia Coastal Plain; by Morgan R. Crook, Jr. (1987).

No.4 - Rush: An Early Woodland Period Site in Northwest Georgia; by W. Dean Wood and R. Jerald Ledbetter (1990).

No.5 - A Few Visits in Prehistory: Data Recovery at 9RH18, Randolph County, Georgia; by Christopher T. Espenshade (1993).

No.6 - The Pig Pen Site: Archeological Investigations at 9RI158, Richmond County, Georgia; by R. Jerald Ledbetter (1988).

No.7 -

Data Recovery at Lovers Lane, Phinizy Swamp and the Old Dike Sites, Bobby Jones Expressway Extension Corridor, Augusta, Georgia; by Daniel T. Elliott, R. Jerald Ledbetter, and Elizabeth A. Gordon (1994).

No.8 - The Victory Drive Site, 9ME50, Muscogee County, Georgia; by R. Jerald Ledbetter (1997).

No.9 - The Bull Creek Site, 9MEI, Muscogee County, Georgia, by R. Jerald Ledbetter (1997).

No. 10 - An Archeological Survey ofthe Shoulderbone Tract, Hancock County, Georgia; by Thomas J. Pluckhahn (1997).

Cover illustration: Tobacco Pipe from WPA Excavations of the Mississippian Period Bull Creek Cemetery.

Preface
The passage of the Intermodal Surface Transportation Efficiency Act (ISTEA) in 1991 required states to set aside at least ten percent of Federal funds for transportation enhancement projects. In particular, the Transportation Enhancement Activities (TEA) provision of ISTEA provides the opportunity to fund a range of non-traditional projects, including historic preservation and conservation activities with a direct relationship to the intermodal transportation system. Within the state of Georgia, these projects range from parkways and bikeways to restoration of historic railroad depots, lighthouses, and covered bridges and involve a variety of sponsors, including cities, counties, and preservation groups.
One category of projects eligible for Transportation Enhancement funds is archaeological planning and research. The present study is a synthesis of previous archaeological investigations of the Bull Creek (9ME1) site by Southeastern Archeological Services, Inc. for Columbus/Muscogee County Consolidated Government. A portion of this National Register eligible archaeological resource lies within the area of potential effect of the proposed Columbus Riverwalk project. This report was prepared in compliance with Section 106 of the National Historic Preservation Act of 1966, as amended, as a measure to mitigate the proposed project's adverse effect to the archaeological resource.
The Columbus Riverwalk project is part of Columbus/Muscogee County Consolidated Government's Alternative Transportation System Plan, a portion of which is funded by ISTEA. This project illustrates the investment by state transportation agencies, the Federal government (Federal Highway Administration), and local governments in respecting and preserving the past in building for the future. The Georgia Department of Transportation is pleased to publish this archaeological data recovery report as the ninth in its Occasional Papers in Cultural Resource Management Series.
Elizabeth C. Shirk Staff Archaeologist Georgia Department of Transportation Atlanta, Georgia August, 1997

THIS PAGE LEFT BLANK

Abstract
This report is a compilation and synthesis of previous investigations at the Bull Creek site, 9Me1, located near Columbus in Muscogee County, Georgia. The Bull Creek site is the type site for the Late Mississippian period Bull Creek phase. The site was extensively excavated during the WPA period under the direction of Dr. A.R. Kelly. The WPA excavations of 1936 and 1937 encountered well preserved structural remains and an extensive cemetery area. Subsequent excavations of the site were conducted by Kelly in 1950, David Chase in the 1950s, and Frank Schnell, Jr. in 1959. There have also been other minor excavations on the site. Unfortunately, none of these excavations have been adequately reported.
The Bull Creek site has been severely damaged by land clearing and filling related to commercial development and most of the site has been destroyed. The area formerly containing the Bull Creek site lay within the route of the proposed City of Columbus Riverwalk Project. An intensive archeological survey of the portion of the Riverwalk route which crossed the Bull Creek site was conducted by Southeastern Archeological Services, Inc. in November 1993. As a result of that survey, a determination was made that only a small part of the site remained preserved. A route for the riverwalk was designed to avoid the preserved site area.
The Bull Creek site and a bordering site named the Victory Drive site, 9Me50, were recommended eligible to the National Register of Historic Places. While portions of both sites remain that can provide significant new information, the Bull Creek site was felt to have historical importance because of the extensive WPA-era excavations and its role in defining the Bull Creek phase. Because the continued lack of reporting of previous excavations at the National Register-eligible Bull Creek site constitutes an adverse impact to the site, the logical course of action to mitigate against this adverse impact was to insure that existing notes and records would be assembled into an accessible, final report. This document is the final, comprehensive report.
Efforts to accomplish these tasks met with varying degrees of success. Records and collections of different investigations varied in quality and completeness. However, some information was available for each of the investigations. As a result, we have been successful in assembling a volume that details the history of past investigations. Perhaps more importantly, the volume presents the ideas and site interpretations of past investigators gleaned from field notes and unpublished manuscripts. Reporting of past investigations together with re-analysis of existing artifact collections, has allowed the relevance of the Bull Creek site to be displayed and its role in the Mississippian period of the Southeast to be addressed.
iii

THIS PAGE LEn BLAHK

Acknowledgements
This report is predominantly a compilation of unpublished research relating to a very important archeological site. The Bull Creek site is the location of historically important WPA-era investigations and the type site for the Late Mississippian period Bull Creek phase, yet it has not been well reported. As such, a thorough accounting of all past investigations was necessary. This report should be viewed as recognition of the accomplishments of past researchers Isabel Patterson, A.R. Kelly, Frank Schnell, Jr., and others who salvaged significant information from the site. The Columbus Riverwalk Project provided the opportunity to investigate the site in the field and allowed a re-examination of the documents and collections from previous investigations conducted over much of the twentieth century. Individuals associated with the Columbus Water Works and the Columbus Consolidated Government are to be commended for providing the funds necessary to complete this study. Particular appreciation is directed to Water Works president, Mr. Billy Turner, and Mr. Cliff Arnett, vice-president of Systems Services and Water Resources Operations, who were very supportive throughout the project. David Ashley of Jordan, Jones, and Goulding, Inc. provided valuable coordination and logistical support.
This project was made possible, in great measure, through the diligence of past and present archeologists associated with the Columbus Museum, formerly the Columbus Museum of Arts and Sciences. That institution has been responsible for maintaining manuscripts, records, and collections from most work conducted on the site. The current archeologist for the museum, Mr. Frank Schnell, Jr., requires particular recognition. Frank is the authority on the Bull Creek site and the phases relating to the Mississippian Period of the fall line region of the Chattahoochee River Valley. He freely shared that knowledge as well as the resources of the museum.
A number of archeologists provided comments and assistance in preparing this report. Besides the contributing authors of this report, appreciation is expressed to Jack Walker, Richard Vernon, Dan Elliott, Sammy Smith, Mark Williams and David Hally. This report includes redrafted drawings from old manuscripts and some new graphics, most of which were prepared by Gisela Weis-Gresham. Michelle Walker transcribed the old manuscripts and assisted in editing the report. Tom Gresham assisted in editing the report and Johanna Minich was responsible for much of the report assembly.
v

TIUS PAGE LEn BLANK

Table of Contents

Preface

i

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. iii

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. v

List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ix

List of Tables

xv

Chapter 1 Introduction and Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 1

Setting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 3

Culture History

7

Chapter 2 Early Investigations Background to the WPA Excavations

19

Chapter 3 WPA Excavations of 1936 and 1937 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 33

Lester's Village Excavations .,

" 42

Cemetery Excavations

. . . . . . . . . . . . . . . . . . . . . .. 53

Chapter 4 Kelly's 1950 Investigations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 77

Kelly's Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 84

Results of Stratigraphic Test Pitting and Trenching of the Bull Creek Site. . . . . . . . .. 84

Recommendations for Treatment of Site and Development

87

Chapter 5 Schnell's 1959 Excavations

" 91

Chapter 6 Minor Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 103

1945 Survey by Frank and Barbara Fenenga ..... . . . . . . . . . . . . . . . . . . . . . .. 103

1950s Investigations of David Chase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 105

1960 Assessment by Huscher as Related to Construction of the Walter F. George

Reservoir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 109

1960 and 1969 Investigations at the Go-Kart and Dolly Madison locales.

110

1960 Investigations by Tyler and Jackson . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 117

1970s Developments

. . . . .. 118

1981 Excavations by Jacky Fuller. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 118

Chapter 7 1993-1994 Investigations

125

Chapter 8 Bull Creek Site Material Culture

143

Bull Creek Phase Ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 143

Primary Pottery Attributes

149

Bull Creek Rim Attributes

161

Vessel Form Attributes

169

Summary of Pottery Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 180

Other Ceramic Objects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 191

Shell and Bone Objects

192

Stone Objects

192

Modem Artifacts

194

vii

Chapter 9 Summary and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 195

The Bull Creek Village

198

The Bull Creek Cemetery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 200

The Bull Creek Phase from a Regional Perspective . . . . . . . . . . . . . . . . . . . . . . .. 207

References Cited

211

Appendix A Transcription of Lester's Field Notes, Bull Creek Site, 1936 Appendix B WPA Coding Sheets Appendix C Bull Creek Vessel Form Data

viii

List of Figures

Figure 1.
Figure 2.
Figure 3. Figure 4. Figure 5. Figure 6. Figure 7.
Figure 8. Figure 9.
Figure 10.
Figure 11. Figure 12.
Figure 13. Figure 14. Figure 15. Figure 16. Figure 17. Figure 18.
Figure 19. Figure 20. Figure 21. Figure 22. Figure 23. Figure 24. Figure 25. Figure 26.
Figure 27.
Figure 28.
Figure 29. Figure 30. Figure 31. Figure 32. Figure 33. Figure 34. Figure 35. Figure 36. Figure 37.

Location of Columbus riverwalk-related project areas surveyed by SAS from 1991

and 1994.

1

Location of known archeological sites between the South Commons of Columbus

and Fort Benning

2

Location map of Columbus and the Bull Creek site.

3

Portion of the 1922 Soil Survey Map of Muscogee County

5

Locations of sites referenced in text. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 8

Surveyor's plat of land lot containing the Bull Creek site. . . . . . . . . . . . . . . . .. 14

Portion of Sturges' 1818 map of Georgia showing the settlement of Marshalls on

south side of Bull Creek. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15

Portion of Patterson's (1950) map of the Bull Creek area . . . . . . . . . . . . . . . .. 16

1936 photograph of the Bull Creek site prior to excavation showing the abandoned

corn fields (NPS photograph 2368). . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 17

Portion of a 1938 Soil Conservation Service aerial photograph showing the area of

the Bull Creek site

18

Map of sites found by Mulvihill in 1924. The Bull Creek Site is site 1

19

Photograph of Margaret Ashley and Frank Schnell, Sr., taken during their work in

western Georgia in 1928. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 20

Points of reference described in Patterson letter dated March 9, 1936

24

Ca 1936 photograph of individuals associated with federal relief archeology

25

Close-up of previous photograph showing view of Frank Lester.

26

Drawing of fluted point from Bull Creek (NPS files).

29

1936 photograph of Isabel Patterson at the location of House Site 1 at Bull Creek.. 31

1936 photograph ofIsabel Patterson at House Site 1, view to south toward Bull

Creek Cemetery

30

1936 photograph of Isabel Patterson in borrow pit near House Site 1

30

Redrafted copy of Lester's 1936 excavation plan map of the Bull Creek.

34

Field photograph of excavations in the village area (NPS photograph 2367).

35

Two field photographs of village area excavations

36

Two field photographs of the cemetery area excavations

37

Field photograph of initial trench excavation in the cemetery area at Bull Creek . .. 38

Cemetery map of the Bull Creek site based on Frank Lester's field records. . . . . . 39

Circa 1940 photograph of laboratory workers at Macon cataloging artifacts from

WPA excavations

41

Redrafted sketch map of a portion of the Bull Creek site showing the area of "house

midden"

42

Field photograph of railroad cut in village area also showing road on borrow pit at

right and staked excavations on high ground

43

Field photograph of village excavation and pedestaled grid stakes

44

Photograph of "brickett" material from Structure 1

44

Field photograph showing fired clay covering floor of Structure 1

45

Field photograph of the raised clay "curb" within Structure 1

45

Floor plan of Lester's Structure 1

46

Photographs of two vessel fragments from trash pit adjacent to Structure 1

47

Plan of a portion of Structure 1 showing the location of urn and calcined bone

48

Field photograph of possible burial urn from Structure 1

48

Field photograph of three smashed pots on floor of Structure 2

49

ix

Figure 38. Field photograph showing workers clearing the floor of Structure 2 and the locations

of shattered vessels

50

Figure 39. Modern photograph of restored vessel from Structure 2.

51

Figure 40. Photograph of restored vessel from Structure 2

52

Figure 41. Reconstructed plan of Lester's village excavation based upon field note entries. . .. 53

Figure 42. Field photograph showing excavation of the cemetery, view to the north.

53

Figure 43. Four examples of burial field drawings from Bull Creek cemetery

54

Figure 44. Enlarged view of Lester's Bull Creek base map showing layout of the cemetery

grid.

55

Figure 45. Field photograph of the "more expert N.Y.A. workers" at the cemetery

56

Figure 46. Reconstructed map of the eastern part of the cemetery containing Burials 1-33 . . .. 57

Figure 47. Redrafted profile map of the cemetery area

58

Figure 48. Field photograph of the Burial 3 and first dog effigy vessel

59

Figure 49. Series of 1936 NPS photographs showing the removal of Burial 3 for exhibition. .. 60

Figure 50. Field photograph of Burial 5

61

Figure 51. Field photograph of Burials 6 in foreground and Burial 7 in background upper

right

62

Figure 52. Modern photograph of second dog pot.

62

Figure 53. Field photograph of Burial 8. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 63

Figure 54. Field photograph of Burial 12

. . . . . . . . . . . . . . . . . . . . . . . . .. 64

Figure 55. Field photograph of Burial 14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 65

Figure 56. Field photograph of Burial 16 and third dog pot

66

Figure 57. Photograph of tobacco pipe from Burial 17

66

Figure 58. Field map of Burial 18.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 67

Figure 59. Photograph of small bowl found with Burial 18

67

Figure 60. Field drawing of tobacco pipe from Burial 18

68

Figure 61. Field photograph of Burial 19

. . . . . . . . . . . . . . . . . . . . . . . . . .. 68

Figure 62. Field photograph of Burial 20

. . . . . . . . . . . . . . . . . . . . . . . . .. 69

Figure 63. Photograph of marine shell beads found with Burial 20

69

Figure 64. Field photograph of Burial 22

. . . . . . . . . . . . . . . . . . . . . . . .. 69

Figure 65. Plan map of western portion of Bull Creek cemetery.

71

Figure 66. Photograph of extended burial 38, a multiple burial

72

Figure 67. Field photograph of Burial 44 ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 74

Figure 68. Plan drawing showinginterpretedboundaries of cemetery excavations superimposed

upon Lester's base map.

75

Figure 69. Reconstructed plan of cemeteryshowing area thought by Lester to contain structural

evidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 76

Figure 70. Redrafted copy of a 1950 City Engineers map of the Bull Creek site.

81

Figure 71. Redrafted copy of the map of Kelly's excavations. . . . . . . . . . . . . . . . . . . . .. 82

Figure 72. Enlarged portion of 1950 aerial photograph of the Bull Creek site area

83

Figure 73. Field photograph of Unit 1 showing depth of railroad fill . . . . . . . . . . . . . . . .. 84

Figure 74. Photograph of ExcavationArea 3 of Unit 3, captioned: Dr. A.R. Kelly, Mrs. Wayne

Patterson and Doc discussing plans for work

85

Figure 75. Field photograph with notation: Dr. A.R. Kelly holding pottery vessel to show better

its size and shape. The dirt inside has not been removed

86

Figure 76. Photograph of vessel recovered by Kelly in 1950 . . . . . . . . . . . . . . . . . . . . .. 86

Figure 77. Field photograph captioned: Dr. A.R. Kelly indicating the width ofthe black midden

deposit; charred clay can be seen

87

Figure 78. Photograph dated 1957 showing a Bull Creek dog pot on display at the Smithsonian

Institution

90

x

Figure 79. Copy of a 1957 plat of the J.W. Woodruff property containing the Bull Creek

site

91

Figure 80. Plan map showings areas of Schnell's 1959 investigations

92

Figure 81. Series of photographs showing the 1959 excavations in progress

93

Figure 82. Profile 3 drawing of bank cut near Structure 1 originally excavated in 1936

94

Figure 83. Profile drawing showing the railroad fill at a location nearer the mouth of Bull

Creek and adjacent to XU 2

95

Figure 84. Plan map of pit found in XU-I.

96

Figure 85. Schematic representation of the soil stratigraphy adjacent to XU-2

97

Figure 86. Field photograph showing vessel fragments exposed in excavated Feature 3

98

Figure 87. Photograph of Lamar Complicated Stamped Bowl from Feature 3

98

Figure 88. Illustrations of two Fort Walton-style vessels from XU-2. . . . . . . . . . . . . . . . .. 99

Figure 89. Example of bold incised vessel fragment from XU-2.

100

Figure 90. Plan map showing the location of pits and postmolds in XU-2

101

Figure 91. Copy of a sketch map showing the locations of Fenenga Sites 2-5

104

Figure 92. Examples of ceramics from the Werocoba Creek Site

104

Figure 93. Portion of a 1955 aerial photograph of the Bull Creek site area

106

Figure 94. Copy of sketch map prepared by Chase showing location of excavation units. . . .. 107

Figure 95. Map of the Victory Drive site showing the Feature locations

110

Figure 96.. Facsimile of unattributed newspaper article describing discoveries at the kart racing

track

111

Figure 97. 1960 photograph showing Bettye Broyles excavating a large pit in the kart track

construction area.

112

Figure 98. Plan map of postmold pattern in the Go-Kart locale

113

Figure 99. Two views of pit excavation at the Dolly Madison plant

114

Figure 100. Selected aboriginal artifacts from the Creek pit in the Dolly Madison locale. . . . .. 115

Figure 101. Selected Euro-American artifacts from the Creek pit in the Dolly Madison locale. . 116

Figure 102. Redrafted sketch map prepared by Tyler and Jackson showing landmarks in the

vicinity of Pothole A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 117

Figure 103. Sketch map prepared by Jack Fuller showing the Bull Creek site in 1981.

118

Figure 104. Small complicated stamped vessel from the Bull Creek site.

119

Figure 105. Artist restoration of effigy vessel from trash pit excavated in 1981

. . . . . . .. 119

Figure 106. Examples of Bull Creek phase sherds from a pit excavated by Jack Fuller in 1981. . 120

Figure 107. Examples of incised and complicated stamped sherds from trash pit excavated in

1981.

121

Figure 108. Examples of Bull Creek phase pottery objects and Archaic bifaces found at Bull

Creek in 1981.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 122

Figure 109. Route of riverwalk through the Bull Creek and Go-Kart sites.

126

Figure 110. Photograph of the Bull Creek site as it appeared in 1993, view to the east.

. .. 127

Figure 111. Photograph of landfilling at the Bull Creek site in 1993, view to the southeast. ... 127

Figure 112. Photograph taken in 1993 of the approximately 50 ft thick landfill at the edge of Bull

Creek.

128

Figure 113. Plan map the Bull Creek site in 1993 showing locations of SAS shovel tests and

surface collections.

129

Figure 114. Photograph of 1993 shovel testing.

130

Figure 115. 1993 photograph showing the large erosional feature impacting the Bull Creek

site

131

Figure 116. 1993 photograph of gully wall following profile cleaning.

132

Figure 117. Profile drawing of a portion of the gully wall examined by SAS in 1993. . . . . . .. 133

Figure 118. Photograph of section of profile shown in previous profile drawing

134

xi

Figure 119. Tentative approximation of the boundaries of 9Me1 and 9Me50 based upon survey

results

135

Figure 120. Plan map of data recovery area of the Go-Kart site showing trenches and block

excavations

136

Figure 121. Plan map of block B at the Go-Kart site.

138

Figure 122. Two photographs of the excavation of the large Creek pit in Block B of the Go-Kart

site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 139

Figure 123. Examples of vessels and glass and metal beads from the Creek pit. . . . . . . . . . .. 140

Figure 124. Photograph of Bull Creek phase structure in Block D of the Go-Kart site . . . . . .. 141

Figure 125. Plan map of Block D at the Go-Kart site. . . . . . . . . . . . . . . . . . . . . . . . . . .. 142

Figure 126. Location of Bull Creek showing boundaries of Lamar and Fort Walton ceramic

traditions

143

Figure 127. Geographical extent of the Bull Creek phase . . . . . . . . . . . . . . . . . . . . . . . .. 144

Figure 128. Lamar Complicated Stamped Design motifs from Rood's Landing

145

Figure 129. Graph showing differences in pottery collections from Bull Creek phase and Stewart

phase sites

148

Figure 130. Drawing of typical Stewart phase incised motif

149

Figure 131. Copy of National Park Service summary sheet of 2240 sherds from the WPA

excavations at Bull Creek; various sherd attributes.

150

Figure 132. Copy of National Park Service summary sheet with counts for temper, paste and

texture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 151

Figure 133. Copy of National Park Service summary sheet of 2240 sherds from the 1936-1937

excavations which quantifies attributes of decoration, quality of ware and vessel portidti2

Figure 134. WPA drawings of Bull Creek phase plain and complicated stamped sherds. . . . . .. 153

Figure 135. WPA drawings of incised and punctate sherds from Bull Creek . . . . . . . . . . . .. 154

Figure 136. Enlarged photographs showing tempering agents found at Bull Creek

155

Figure 137. Examples of Lamar Complicated Stamped and Mercier Check Stamped sherds from

Bull Creek

157

Figure 138. Typical examples of incised and punctated sherds from Bull Creek

158

Figure 139. Comparison of pottery counts from four archeological investigations of the Bull

Creek site

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 160

Figure 140. Examples of appliqued rims from the Bull Creek site.

161

Figure 141. Examples of additional rim forms from the Bull Creek site.

162

Figure 142. Copy of National Park Service summary sheet of Bull Creek rim attributes.

163

Figure 143. Graph showing rim varieties identified from a collection of 278 sherds

164

Figure 144. Series of graphs comparing rim type and vessel treatment for sherds examined by

Schnell

166

Figure 145. Photograph of human effigy adorno

168

Figure 146. Animal effigy rim adornos from Bull Creek site collections

169

Figure 147. Examples of vessel forms identified from various Mississippian localities.

170

Figure 148. Partially reconstructed Lamar Complicated Stamped everted rim vessel with

appliqued pinched rim strip

. . . . . . . . . . . . . . . . . . . . . . .. 171

Figure 149. Graph showing the range of rim sizes for everted rim jars from Bull Creek.

.. 172

Figure 150. Examples of Fort Walton carinated bowls from Bull Creek.

173

Figure 151. Graph showing the range of rim sizes for carinated bowls from Bull Creek.

174

Figure 152. Examples of simple rounded bowls with slight to moderate rim constrictions.

175

Figure 153. Graph illustrating range of rim diameters for simple rounded bowls from Bull

Creek. .

176

Figure 154. Example of a simple open bowl from Bull Creek. . . . . . . . . . . . . . . . . . . . . .. 177

xii

Figure 155. Graph illustrating range of rim diameters for simple open bowls from Bull

Creek.

178

Figure 156. Illustrations of effigy bottles from the Bull Creek site with comparative examples

from other western Georgia sites.

179

Figure 157. Examples of motifs for Lamar Complicated Stamped var. Bull Creek and var.

Early. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 181

Figure 158. Example of coarse grit tempered, bold check stamped pottery from the Bull Creek

site

" 182

Figure 159. Comparison of common Lamar Incised designs from three Lamar phases . . . . . .. 184

Figure 160. Examples of Fort Walton incised pottery types and varieties typical of the Bull

Creek phase

186

Figure 161. Incised designs found at Bull Creek. .

" 188

Figure 162. Examples of other pottery types from Bull Creek

190

Figure 163. Examples of ceramic pipes and pottery discs from Bull Creek.

191

Figure 164. Examples of other ground stone objects from Bull Creek. . . . . . . . . . . . . . . . .. 192

Figure 165. Examples of projectile points found on the Bull Creek site which pre-date the Bull

Creek phase occupation.

193

Figure 166. Chert biface associated with the Bull Creek phase occupation. . . . . . . . . . . . . .. 194

Figure 167. Locations of major archeological excavations at Bull Creek.

196

Figure 168. Map showing estimated extent of the Bull Creek Village. . . . . . . . . . . . . . . . .. 199

Figure 169. Plan map of the Bull Creek Cemetery

" 201

Figure 170. Burial orientations in the Bull Creek Cemetery.

. . . . . . . . . . .. 205

Figure 171. Contrasting burial orientations of two burial clusters in the Bull Creek Cemetery. . 206

xiii

THIS PAGE LEFT BLAHK

List of Tables

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

Characteristics of Vessel from Fuller's collection . . . . . . . . . . . . . . . . . . . . .. 123

Tabulation of artifacts recovered from gully midden slump from 9Mel. . . . . . . .. 133

Averett phase radiocarbon dates from the Go-Kart site.

137

Ceramic counts from Schnell's 1959 excavations. . . . . . . . . . . . . . . . . . . . . .. 156

Analysis of sherds from Kelly's 1950 excavation

. . . . . . . . . . . . .. 159

Detailed rim analysis of 278 sherds from Kelly's 1950 excavations at Bull Creek. . 165

Comparison of rim diameter and width of folded pinched rim fold

168

Vessel forms and counts from Bull Creek.

173

Measurements of effigy bottles from Bull Creek.

178

Comparison of temper for zone punctate and incised pottery from Bull Creek

183

Presentation of Scarry's (1985) incised types and varieties of the Bull Creek

Phase. .

185

Incised design categories identified from the Bull Creek site collections.

189

Summary of Bull Creek Cemetery

202

Contrast of Burial Orientations from two burial clusters from Bull Creek

Cemetery

207

xv

TIUS PAGE LEn BLANK

Chapter 1 Introduction and Setting

During the early 1990s, the Columbus/Muscogee County Consolidated Government began designing and constructing several interrelated development projects along its riverfront. The various projects are interwoven geographically, administratively and financially through revenue bond issues, local option sales taxes, Intermodal Surface Transportation Enhancement Act funds and special federal appropriations. Several departments of the City are involved, but the Columbus Water Works is taking the lead on administering the set of projects. The City has been working with the consulting engineering firms of Jordan, Jones & Goulding, Inc. and Hecht, Burdeshaw & Johnson, P.C. in the conceptualization, design and permitting of the projects. The city projects consist of the continuation of sewer line improvements along the river, the continuation of converting the river edge to a landscaped riverfront walkway, a new civic center, an Olympic softball complex, and general landscaping and park/recreation development on a set of contiguous tracts of land that stretch from the south side of downtown Columbus southward to Fort Benning (Figure 1). Some of the planning for the recreation and interpretation aspects of the project, which will focus on tracts near Fort Benning, are still conceptual.

Four segments of the project area have been surveyed, tested and documented as archeological assessment reports (Wood 1991; Wood 1993; Ledbetter 1994a and 1994b).

One facet of the design process has been concern for potential construction impacts to archeological and historical resources in the project area. To allay the City's concerns in this matter, and to be in compliance with Section 106 of the National Historic Preservation Act of 1966, as amended, the City contracted with Southeastern Archeological Services, Inc. (SAS) to perform all necessary archeological surveys, historical research and site evaluations associated with the project.

o

2

5N5--5-~

km

,="",-"""":"""~-~~~~""",,:,,,,,,-~-~---:---,
Figure 1. Location of Columbus riverwalk-related project areas surveyed by SAS from 1991 and 1994.

1

The Bull Creek Site

The project segments consist of sections of Columbus riverfront where substantial improvements to a combined sewer line are to be made, after which the area is to be converted to a riverfront walkway/park. Other areas of proposed development, in large part related to the softball sports complex for the 1996 Olympics, were included in Ledbetter's 1994a survey report.

Cart site (9Me50), which is best known for its historic Creek occupation (ca A.D. 1800). 9Me1 and 9Me50 are actually contiguous sites on the north side of Bull Creek.

Figure 2 shows all archeological sites in the area of the riverwalk route between the South Commons and Fort Benning. All identified archeological sites within and immediately adjacent to the survey areas are shown. Most of these sites are prehistoric and thus relate most directly to the Bull Creek site. The riverwalk route and site locations are depicted on a copy of the 1908 USGS topographic map, which best depicts the landscape prior to modern development.

The archeological survey and assessment work revealed that two sites to be impacted by the proposed project were clearly significant (eligible for listing on the National Register of Historic Places).
The conclusions led to archeological mitigation on the two sites, Bull Creek (9Me1) which contains an important Late Mississippian Bull Creek phase component (ca A.D. 1500) and the Go-

9Me49.

9Me35

N

9Me24
n

9Me14

A
km

Figure 2. Location of known archeological sites between the South Commons of Columbus and Fort Benning (map source USGS Columbus 1908).

2

Introduction and Setting

Both site areas had been severely

included. Realistically, this is the only means

disturbed by past development and only

of examination, simply because there was

portions of each remained preserved. Both

never an adequate determination of the limits

sites have also been subjected to archeological

of the Bull Creek site prior to destruction of

excavation in the past, and the Bull Creek site

most of the site. By piecing together the

has been rather extensively excavated for

results of many surveys and excavations

several decades. The riverwalk was

conducted throughout the present century, we

constructed so as not to disturb intact portions

will attempt to define the extent and status of

of the Bull Creek site. However, state and

the Bull Creek site during aboriginal times.

federal officials agreed that the continued lack

of reporting and publication on various important excavations at Bull Creek was an

Setting

adverse impact that could be remedied by the current project. Therefore, a data recovery plan was developed for compiling the extensive body of hitherto unpublished excavation data for the site. The most significant data relates to the WPA excavations of the 1930s. A synthesis of these excavations and subsequent work on the site is the primary goal of this report.

The Columbus project area is located along the western edge of Muscogee County, Georgia (Figure 3). The most distant points of the overall project area extend a maximum distance of 2.7 km from the river. Bull Creek, which joins the Chattahoochee River approximately 2 km below the South Commons of Columbus, is the most significant tributary stream relating to the riverwalk survey.

Data recovery for the Go-Cart site included extensive field excavation conducted during the summer of 1994. The results of data recovery at that site will be presented as a

Tributaries of Upatoi Creek, another significant tributary of the Chattahoochee River, lie just to the south of the Bull Creek site.

separate report. However, as a result of that

investigation, we have confirmed that

there is substantial overlap between

occupation areas for certain

components, which means the site

boundaries as presently defined are not

entirely meaningful. Perhaps the most

important find of the 1994 excavations at the Go-Cart site was the remains of a

MUSCOGEE COUNTY

Bull Creek phase structure. We also

have evidence that the Go-Cart site

-e.
~

..

occupation areas for the Archaic and

historic Creek components extend onto the Bull Creek site. Because the two site areas are interconnected, the more important archeological discoveries

) Columbus
:@
\
BullCreek Site .... 9Mel

within the defined boundaries of the

Go-Cart site will be included in this report. Pertinent information from other nearby sites, which previous

CHAITAHOOCHEE COUNTY

o

5 Miles

o~~SkIn

researchers have suspected as "part" of

the Bull Creek site will also be

Figure 3. Location map of Columbus and the Bull Creek site.

3

The Bull Creek Site
The Bull Creek site is located in a geographical area known as the Fall Line, which is a zone of varying width that separates the lower Piedmont from the upper Coastal Plain. This zone, also referred to as the Sand Hills and Fall Line Hills, is widest along the Chattahoochee River and narrows to the east as it stretches across Georgia to the Savannah River and into South Carolina (Hodler and Schretter 1986:16-17).
One significant aspect of the Fall Line is that rivers drop dramatically in a short length. At the Fall Line the Chattahoochee River flows across a 4-km long series of bedrock ledges, descending 38 m in elevation before leveling off in Columbus. Chattahoochee Falls is the term given to this series of ledges north of the Dillingham Street Bridge. The backwaters of Walter F. George Reservoir reach this area: which historically has been considered the head of navigation for the Chattahoochee River. From this point south the river widens as it meanders toward the Gulf of Mexico.
The underlying bedrock at the Chattahoochee Falls consists of Precambrian metamorphic and igneous rocks, which form the base of the Piedmont Province. These rocks include granites, gneisses, schists, and granite-gneisses. Below the falls, in the Coastal Plain, are Cretaceous deposits resulting from ancient ocean sediment. Much of the sediment along the river edge below the falls are more recent alluvial deposits from Pleistocene and Holocene times.
Much of the riverwalk survey area follows a nearly level terrace overlooking the Chattahoochee River on the south side of Columbus. Although the Columbus survey area appears nearly level, there is actually about 10 m of elevation difference across the terrace. This relief is most apparent well to the south of the Bull Creek site, where the broad Chattahoochee River terrace abuts the uplands.
4

The main terrace, which includes the Bull Creek site, is about 15 m above average river level. The terrace top varies from about 70 to 73 m (230 to 240 ft) in elevation.
The project area is located upon alluvial soils that form the broad terrace system of the Chattahoochee River. The Bull Creek site vicinity is composed of two principal soil types, both of which provide insights into past land use (Johnson 1983). The area north of Bull Creek is classified as Dothan-Eunola series, 0-3 percent slope. This sandy loam soil of marine origin occurs on stream terraces and can be considered old alluvial deposits, as it is rarely flooded (Johnson 1983:21). The area south of Bull Creek is classified as ChewaclaDogue series. Chewacla-Dogue soils are modem stream deposits consisting of poorly drained to well drained loamy and clayey sands.
Figure 4 shows a portion of the 1922 soil survey map of Muscogee County. The soil associations shown on this early map are important because they represent the terrain immediately preceding the modem period of intensive urban development, clay mining, and landfilling. During the last 100 years tremendous deposits of alluvium have been deposited along the river edge as a result of severe erosion of upland soils in the Piedmont due to poor agricultural practices (Trimble 1974).
Within the Fall Line Sand Hills, upland soils are well-drained and support sparse vegetation represented by scrub oak, longleaf pine and low shrubs (Wharton 1978: 180). Lowland areas along the major streams originally supported a variety of deciduous species similar to that of the Piedmont. Much of the land in the project area was probably cleared for farming early in the settlement of Columbus. In light of the extensive Mississippian and Creek settlement of the area, some tracts of land were probably cleared well before whites settled the area.

Introduction and Setting

ot' Figure 4. PoHion the 1922 Soil SurVeY.Map of Muscogee County showing the project area.

The earliest known mention of Bull Creek itself was by Benjamin Hawkins, who was Principal Temporary Agent for Indian Affairs South of the Ohio. The first entry is dated January 9, 1797.
The banks of the river generally fifty feet high down to the town house (Coweta on the west bank). A creek empties in half a mile above, and it descends with a rapid current, going up this creek a small one {Weracoba} joins it on the left, about 200 yards from the river, continuing up one mile, the lands on both sides to the margin without any swamp; the bank steep enough anywhere for mills.
(Hawkins, 1916:60).

Two years later Hawkins provided another description.
In descending the river path from the falls in three miles you cross a creek running to the right, twenty feet wide; this creek joins the river a quarter of a mile above the Cowetuh town house: the land to this creek is good and level and extends back from the river from half to threequarters of a mile to the pine forest; the growth on the level is oak, hickory, and pine; there are some ponds and slashes back next to the pine forest, bordering on a branch which runs parallel with the river; in the pine forest there is some reedy branches.

5

The Bull Creek Site
The creek has its source nearly twenty miles from the river, and runs nearly parallel with it till within one mile of its junction; there is makes a short bend round north, thense west to the river; at the second bend, about two hundred yards from the river, a fine little spring creek on its right bank
(Hawkins 1916).
A rather poetic description of the setting for the Bull Creek site excavations is provided in an undated manuscript prepared by Isabel Patterson which was eventually edited for brevity and published in Early Georgia (Patterson 1950).
The geographic setting of the Bull Creek Site is significant in understanding the implications of the unusual finds during fourteen months of archaeological investigation. Bull Creek rises in rolling red foothills about twelve miles south of Pine Mountain Valley and approximately the same distance from the Chattahoochee River, which forms the western boundary of this section of the Empire State of the South. Bull Creek flows southward for nearly twenty miles almost parallel to the river until within one mile of their confluence, here the creek makes a short bend around to the north and there turns twice in a westerly direction to join the river three miles below the "Coweta Falls," and a quarter of a mile north of the historic site of the "Coweta Town House" on the west bank of the Chattahoochee. A fine little spring creek, now called "Weracoba," joins Bull Creek on its northern bank six hundred and sixty feet from the river. These streams form the eastern, southern and western boundaries of the Bull Creek Site and Victory Drive - (once the old Federal Road and later Lumpkin Boulevard) forms the northern -Briefly, this ancient home of aboriginal Americans covers three quarters of a square mile in the northwestern corner of section 66 of the Coweta Reserve on the east bank of the Chattahoochee, where in 1828, the "Trading Town of Columbus was established near the Coweta Falls" in Muscogee County, Georgia.
The Chattahoochee appears prominent as an avenue of glorious adventure associated with trails of legend and later history. Its head springs rise in the majestic Blue Ridge Mountains of northeastern Georgia, within a few hundred yards of the Hiwassee, which flows northwest to join the Tennessee and merge with the Mississippi. The course of the Chattahoochee is southwest from the "Valley of Nacoochee" to "Chattahoochee Old Town" in Troup County, where it forms the dividing line between the present State of Georgia
6

and Alabama. The river flows almost due south to the "Coweta Falls" where rough rocks and islands force the stream into two channels and here the river is fordable to the termination of the Falls. A creek once known as "Falls Creek" enters the river at this point just north of the site of the bridge that was built across the lower end of the Falls at Columbus (1832-33) to connect with trails centuries old.
South of the present city of Columbus, the river winds along forming great bends, and Coweta Town on the west bank occupied the first of these. As the river snakes along southward around "Jenny's Island" there are two famous aboriginal sites near the shoals "Woolfolks Mound" on the east and "Coweta Tallahassee" on the west - later known as "Abercrombie Mound," then "Broken Arrow Bend" just north of Upatoi Creek on the east and below this the great "Kasihta Town" of historic times - on the low flat of good land now occupied by Lawson Field - Fort Benning, Georgia.
The most important trail of this section was the branch of the Lower Creek Trading Path from the Savannah, near Augusta, to "Rock Landing" on the Oconee, below Milledgeville, through "Old Ocmulgee Fields" at Macon, and on across the Flint, near Benjamin Hawkins Creek Agency and westerly to Kasihta, where it branched - one going north through the Kasihta Fields to Coweta Falls and the other trail across the Chattahoochee to the west side where Fort Mitchell was later established. This trail was later followed by the Old Federal Road and a line of forts was established.
The "Point Towns" of the Chattahoochee lie in the Oswichee Bend south of Uchee Creek on the west bank and here too was the ancient Apalachicola Town - when the legendary Creeks came to this area. The river below this great bend flows again southward to be joined by the water of the Flint and forms the Apalachicola at the southwestern edge of the present State of Georgia and these continue to the Gulf of Mexico - through
northwestern Florida (Patterson n.d.).
Schnell (1963) provides a description of the Bull Creek site as it appeared around 1960.
A tributary, Weracoba Creek, joins Bull Creek on its north bank about 200 m from the river. These three waterways form the eastern, southern and western boundaries of the site.

The site is roughly bounded on the north by Victory Highway {U.S. 27 & 280}. The site may have extended slightly beyond Victory Highway but the remains, if any, were obliterated by the construction of the Columbus Municipal Airport and a Nehi Bottling Plant. The Bull Creek Site is entirely within the city limits of Columbus, Georgia.
The site itself has been impinged upon by several modern developments. Until a few years ago, most of the back section {away from the river} of the site was covered over by a race track, with stables in the fore section. Previous to that time, a railroad had been built across the site and a railroad bridge spanned Bull Creek at its confluence with the Chattahoochee. The bed for this track covered over or destroyed a large part of the peninsular portion of the site. The back part of the site is currently covered over by a golf driving range.
The soils of this area are derived from the unconsolidated sand and clay beds of the Coastal Plain but may be traced in part to the Piedmont Plateau from which they were transported by the streams and the river. The soils on the site are built up upon a base of fossiliferous marl. The flood waters of the Chattahoochee and its tributary streams such as Bull Creek have periodically covered this area, especially notable floods having occurred in 1919 and 1939. The topography of the land is nearly level above the bluffs of the river, with slight ridges and depressions. The bluffs of the river are about fifty feet high at the site, with numerous springs flowing from its side. Bull Creek joins the Chattahoochee River three miles below the Fall Line which divides the Piedmont region of northern Georgia from the southern Coastal Plain.
Construction of a Go-Kart race track just north of the Bull Creek Site uncovered in 19591960 evidences of an as yet, undocumented historic settlement. Two large pits yielded a large amount of Chattahoochee Brushed pottery, some Kasita Red Filmed, and numerous trade items, including glass, china ware, several iron mugs, and one musket barrel. At various other points on the site isolated finds of trade items have also been
found (Schnell 1963).
It is obvious from these observations that several conditions were highly favorable for the site's location. Perhaps of greatest importance were proximity to several water

Introduction and Setting
sources which included large springs, tributary streams and the river, sufficient elevation above the river to limit the disturbance from frequent flooding, and access to broad expanses of fertile farmland. Upland resources and particularly the rich resources of the falls were in close proximity.
Culture History
As a result of numerous archeological projects and overviews conducted within the region since the 1890s, the culture history of the Columbus/Lake Walter F. George area is relatively well documented. A survey report by Knight and Mistovich (1984) entitled Walter F. George Lake; Archaeological Survey of Fee Owned Lands was produced for the U.S. Army Corps of Engineers and is the most detailed reference for the region. A second detailed culture history of the region has recently been prepared for the Fort Benning area (Elliott et al. 1995). The known culture history of the region will be only briefly summarized here. Emphasis will be placed upon the periods most pertinent to the Bull Creek site, specifically the Mississippian and historic Indian periods. Figure 5 shows locations of important archeological sites referenced in the text.
The first people came to the Chattahoochee Valley during the Paleo indian period, approximately 10,000 to 12,000 years ago (Anderson et al. 1990), when the climate was quite different from today's and a canopy of oaks and other hardwoods covered much of the Southeast. Temperatures were cooler and possibly more equable as a result of the continental ice-sheet still present to the north. As the ice melted, the sea level began to rise and encroach on the Coastal Plain. It is believed that the earliest inhabitants were nomadic hunters and gatherers who occasionally killed mastodons, bison, and other Pleistocene megafauna. Through time these large animals became extinct due to human predation and climate change.
7

The Bull Creek Site PIEDMONT

Roods Landing

Kolomoki .~s~,

COASTAL PLAIN

N

I

o

SOMi

lI ::::====:::::"

Figure 5. Locations of sites referenced in text. 8

As prehistoric cultures adapted to the extinction of the large herbivores and climate changes, white-tailed deer became a prime component of the Indian's diet. The interval between approximately 8000-1000 B.C. is defined as the Archaic cultural period. Early in this period the Fall Line region of southwestern Georgia and southeastern Alabama was regularly exploited by Archaic hunters and gatherers and sites containing stone tools diagnostic of the Early Archaic period are fairly common. By about 6000 B.C. the climate became warmer and drier, and, while the forests still were characterized by oaks and other hardwoods, a more open vegetation pattern is indicated (Delcourt and Delcourt 1985). Relatively little is known of the Middle Archaic occupation of the Chattahoochee Valley. However, both Piedmont (Morrow Mountain) and Coastal Plain tradition (stemmed chert points) stone tools have been reported.
More reliable information is available concerning. the Late Archaic and Gulf Formational periods, which date from ca. 4000-800 B.C. Diagnostic stone tools include stemmed projectile points of various types such as Pickwick, Little Bear Creek, Abbey and Savannah River Stemmed. During this time the climate became moister and may have been similar to current conditions. Pines began to eclipse oaks as the predominant species in Coastal Plain forests. During the latter part of the Late Archaic period pottery was invented, marking the beginning of the Gulf Formational period in the region (Jenkins 1978). The earliest pottery is very distinctive because the clay was tempered with vegetable fiber, probably Spanish Moss. Fiber tempered pottery sites are particularly common along the Chattahoochee River below the Piedmont. It was during the Late Archaic and Gulf Formational periods that squash was introduced as the first domesticated garden vegetable. The addition of pottery and domesticated crops suggests that prehistoric lifeways were becoming more sedentary.

Introduction and Setting
The Early Woodland period dates from approximately 800-300 B'C, in the study area, but is poorly defined due to a genuine scarcity of identified sites of this period. Early Woodland pottery is tempered with sand and surfaces of vessels are often decorated with simple stamped and check stamped designs. During the succeeding Middle Woodland period (300 B.C.-A.D. 500), the Chattahoochee Valley was the scene of distinct changes.
Based on greater numbers of sites, there appear to have been both a population increase and the development of ceremonial activity echoing developments elsewhere in the eastern Woodlands. Corn was added to the other Woodland cultigens, but it did not playa vital role in the diet until the late prehistoric period.
The Mandeville Site (9Cy1), located in Clay County, Georgia is an important mound center during the Middle Woodland period. The site consists of a large village area with two mounds. One mound is an earthen platform which may have supported a ceremonial structure such as a charnel house or dwelling of the village leader, and the other is a conical burial mound. Ceramics from the site include check stamped wares and pottery decorated with new curvilinear designs known as Swift Creek Complicated Stamped.
There are clear indications that during the Middle Woodland period people in the Chattahoochee Valley were interacting with groups in Ohio, Tennessee, Alabama, Mississippi, and Florida. Late in the Middle Woodland period the foreign cultural influence waned and the settlement pattern seems to have become more nucleated, centering on the large sites such as Kolomoki to the south in Early County, Georgia (Sears 1956).
9

The Bull Creek Site
Between A.D. 500-750 the southern portion of the Chattahoochee Valley was influenced by people associated with the Weeden Island culture of Gulf Coast derivation. These groups shared similar pottery-making traits and ceremonial activities. A late variety of Swift Creek Complicated Stamped pottery was made, along with Gulf Tradition incised, punctated, and plain vessels. Kolomoki continued as a dominant settlement during this time. Again, few sites of this period have been identified in the surrounding area, probably because few locales have received more than basic reconnaissance surveys.
Even less is known about developments from ca. A.D. 750-850. Late Weeden Island cultures occupied the Apalachicola River valley in numerous settlements. According to Knight and Mistovich (1984) only a few late Weeden Island components are present in Walter F. George Reservoir and these are restricted to the southern part of the lake. During this interval the area around Columbus appears to have been occupied by cultures producing predominantly plain pottery. This may correspond to Chase's (1959) Upatoi complex, identified from excavations in the Fort Benning area. The Upatoi complex is characterized by a large percentage of plain pottery and small percentages of chevronincised and "ripple tooled" vessels (Chase 1959). If the complex persisted, it may have evolved into the early Mississippi Averett.
The Mississippi period dates from approximately A.D. 900 to 1550. The Averett phase is sandwiched between two emergent chiefdoms to the north and south. The "heartland" of Etowah culture was north of the Fall Line, centered in northwest Georgia and northeast Alabama. To the south the Fort Walton and Rood phase cultures were replacing Weeden Island cultural antecedents in the lower Chattahoochee and upper Apalachicola River systems. Differing from contemporaneous
10

Etowah pottery, characterized by rectilinear complicated stamped jars and bowls, Rood phase ceramics are typically plain, grit-tempered, globular, collared jars with strap handles. Stylistically, Averett vessels are virtually identical to those found in west-central Alabama (West Jefferson phase) and at the Macon Plateau site (Ocmulgee National Monument) on the Ocmulgee River in central Georgia. The sudden appearance of totally new vessel forms made with grit tempered clay suggests a radical cultural change and probably an influx of people from outside the valley. The Averett phase is well dated to between A.D. 900 and 1300 (Schnell and Wright 1993:28; Ledbetter 1995).
South of Columbus, a distinct culture known as the Rood phase developed from ca. A.D. 950 - 1400 (Schnell et al. 1981). In the area encompassed by the Rood phase, a hierarchy of site types is evident. Principal settlements contained multiple platform mounds surrounding a central plaza. Dwellings were scattered up and down the river from the main center. Smaller towns had only one to three mounds. Com agriculture became very important during the Rood phase. According to a recent examination of Mississippian settlement within the region, Rood phase sites do not occur in the Columbus area, although examples of Rood phase pottery do occur on Averett sites (Schnell and Wright 1993:28).
Sometime after A.D. 1300 the Rood phase was replaced by or evolved into the recently defined Singer phase (Schnell and Wright 1993:20). Ceramics of the Singer phase include early Lamar types such as Lamar Complicated Stamped and Lamar Bold Incised, as well as styles similar to Savannah Complicated Stamped. Fort Walton Zone Punctate sherds also occur. At present the territorial range of the Singer phase has not been extended into the immediate Columbus area (Schnell and Wright 1993).

By about A.D. 1400 sites containing Bull Creek phase material commonly appear in the Columbus area. According to Knight and Mistovich (1984) and Schnell and Wright (1993) Bull Creek ceramics can be considered as a blend of generic Lamar with generic Fort Walton of this period. Fort Walton Zone Punctate and Lamar Complicated Stamped pottery characterize the mixture of Gulf and Piedmont ceramic traditions.
Corn agriculture continued to be important in Bull Creek subsistence patterns and the settlement hierarchy was similar to the preceding Rood phase. Large villages appear to have functioned as political and ceremonial centers, while small farmsteads were located in outlying areas. Bull Creek phase sites are more widespread than Rood phase sites and are found throughout the middle Chattahoochee Valley up to the Fall Line at Columbus. Because Bull Creek sites occupy a more northerly territory, there are indications that Averett groups were replaced by, or absorbed into, a true chiefdom level society, one that was contemporaneous with chiefdoms in central Georgia (on the Oconee and Ocmulgee Rivers) and in the tri-state area of northeast Alabama, northwest Georgia, and south central Tennessee (on the Coosa, Etowah, Coosawatee, and Tennessee Rivers).
Recent work by Schnell indicates that the latter part of the Bull Creek period should now be considered a distinct phase (Schnell 1990; Schnell and Wright 1993:21). In this scenario the Bull Creek phase would date from approximately A.D. 1400 to 1475 and the Stewart phase would continue until approximately A.D. 1~50. The Stewart phase is distinguished by a decrease in the proportion of complicated stamped pottery (Schnell 1990:67-68; Schnell and Wright 1993:21).
The Hernando De Soto expedition penetrated the Southeast in 1540 and

Introduction and Setting
represents the first European exploration in the interior. However, it does not appear that De Soto visited the Chattahoochee Valley. Instead, after leaving his winter camp in what is now Tallahassee, Florida, he marched into central Georgia, followed the Flint River for a distance, then turned northeast, crossing the Ocmulgee River just south of Macon. While the expedition did not encounter Bull Creek phase people, there can be little doubt that news of the invaders resulted in disruptions.
It was not until the latter part of the seventeenth century that Europeans made sustained contact with the native societies of the lower Chattahoochee Valley. However, there appears to be a reduction in the number of sites dating to A.D. 1550-1650 as compared to the previous Bull Creek phase. If this corresponds to a population reduction, then European diseases may have had a severe impact. Population reduction must have been proportionately higher in a nucleated settlement where disease could be easily spread from household to household.
Pottery of the Abercrombie phase, the period from A.D. 1550-1650, shows continuity with the preceding Bull Creek/Stewart phases, with Lamar Complicated Stamped still present in low percentages, with bold incised, and Fort Walton zone punctate wares as majority ceramic types. A distinctive attribute of Abercrombie is the occurrence of shelltempering in some vessels. There appear to have been strong cultural affiliations with the Dallas culture of northwestern Georgia and eastern Tennessee (Jenkins 1978) and the Alabama River phase in central Alabama (Knight and Mistovich 1984).
During the first half of the seventeenth century, a chain of Catholic missions was established across north-central Florida and along the Sea Islands of the Georgia coast.
11

The Bull Creek Site
San Luis de Talimali, located in Tallahassee, was the principal mission town. It was protected by a bastioned fort and a Spanish garrison. From this mission in 1679, Father Juan Ocon and two other missionaries were sent to administer to the population in the Chattahoochee Valley, but were ordered to leave by the great chief of Coweta. In 1681, two more Franciscans attempted to establish a mission, reinforced with an escort of soldiers, but again the natives resisted and the foreigners were forced to withdraw.
The settlement of Charles Towne by the English in 1670 proved that the Spanish had only weak control of the Southeast. By 1680 many of the missions along the Georgia coast had been abandoned due to raids by Indians allied with the English, and the friars and their charges retreated southward to the protection of St. Augustine. The English aggressively pursued trade with the Indians, and soon traders were making their way west into the Chattahoochee Valley. From their main supply depot at Charles Towne, the traders exchanged firearms, cloth, rum, and other material goods for deerskins and Indian slaves. The desire for European goods prompted intertribal warfare and slave raiding. The 1685 visit to Coweta by Dr. Henry Woodward, an English trader, brought a quick response from the Spanish garrison at San Luis de Talimali. Commander Antonio Matheos marched into the valley with 250 Apalachee Indians and Spanish soldiers in an unsuccessful attempt to capture Dr. Woodward. They returned in the winter of 1685 and again failed to capture Dr. Woodward, but in the meantime they burned four Indian towns: Coweta, Kasita, Tasquiqui, and Colone. The Spaniards then built a bastioned fort, Fort Apalachicola, at the town of the same name on the west bank of the Chattahoochee River. The Spanish coercion was intended to force the Indians into an alliance, but ultimately it had the opposite effect.
12

Shortly after the completion of Fort Apalachicola, many nearby Indian towns were abandoned, and there was a large-scale migration eastward to the Ocmulgee River, near present-day Macon. There, the Carolinians intensified trade and continued to supply the Indians, referred to by the English as Creeks, with firearms and other goods. In 1704, the Creeks, under the command of Col. James Moore of South Carolina, attacked the mission villages of north Florida and whole villages were massacred or carried off into slavery. The fortified mission of San Luis de Talimali was one of the few to withstand the onslaught. Spanish influence over the Indians withered as time passed, leaving much of the Georgia coast, the Chattahoochee Valley, and north Florida virtually depopulated.
Indian ceramics of this period, the Blackmon phase (A.D. 1650-1715), include the types Chattahoochee Brushed, Ocmulgee Fields Incised, Kasita Red Filmed, and coarse and fine plainwares. Ocmulgee Fields Incised represents the late variety of Lamar Incised. Through time, the incised motifs got more complicated and the width of the incised lines became very narrow.
After years of English exploitation the Indians rebelled in 1715 in what is known as the Yamassee War. A pan-Indian movement, the attacks were well coordinated and focused on white settlements along the east coast; more than 400 colonists were killed. English traders were also systematically executed in Creek and Choctaw towns as far west as Alabama. The revolt was quickly suppressed and the Indians fled to the west, re-occupying the Chattahoochee Valley. At this time many diverse groups settled in the valley including the Oconee, Savannah, Yuchi, Apalachee, Hitchitee, Sawokli, Eufaula and the Apalachicola peoples. It is informative to note that there were language divisions in the Valley after the influx of refugees.

The northern valley towns, including Coweta, Kasita, and Apalachicola spoke Muscogean, while the southern towns, Hitchitee, Oconee, Sawokli and Sawokliutchi, spoke the "Stincard" tongue, or Hitchitee. Sandwiched between these two groups were the Yuchi who spoke a third distinct language. Collectively, these settlements, inhabited by the remnants of once powerful chiefdoms, formed the Lower Creek towns. The Upper Creek towns were located in Alabama on the Coosa and Tallapoosa Rivers.
Aboriginal occupation near Columbus from A.D. 1715-1836 is termed the Lawson Field phase. The ceramics show a continuation from the Blackmon phase and include Chattahoochee Brushed, Kasita Red Filmed, and Ocmulgee Fields Incised pottery types, with Ocmulgee Fields Incised out of production by the mid-eighteenth century. European artifacts are fairly common on Lawson Field phase sites, consisting primarily of gun parts, rum bottle fragments, sheet brass fragments (recycled brass kettles), clay pipe fragments, and glass beads. Apparently, the Creeks did not develop a preference for European ceramics as did many of the Cherokees who lived in north Georgia.
The Indians in the Lower Creek towns became more acculturated than their counterparts in the Upper Creek towns. In part this was due to the efforts of Benjamin Hawkins, the Indian agent appointed by the federal government. He encouraged the Lower Creeks to adopt modern farming techniques, such as intensifying and diversifying their crops, and using the plow. These efforts paid off in 1813 when the Upper Creeks declared war on American settlers in south Alabama and the Lower Creeks joined American forces to defeat the Upper Creeks in the Red Stick War. Tensions were running high at this time. Although Spain claimed Florida until 1821, there was a strong British presence in the town

Introduction and Setting
of Pensacola. British agents were willing to pay Creek Indians a bounty for every American scalp brought in. In addition, the Shawnee prophet, Tecumseh, had visited the Upper Creeks and exhorted them to attack the Americans. There was a skirmish between the two forces at Burnt Corn Creek, but the major event occurred in 1814 when the warriors from the Upper Creek towns, known as the Red Sticks, attacked and massacred nearly 500 men, women, and children at Fort Mims, located north of Mobile. The Red Sticks were defeated by American and Lower Creek forces under the command of General Andrew Jackson at the Battle of Horseshoe Bend in central Alabama. Jackson then marched into Spanish West Florida, where he captured Pensacola after a brief siege.
In 1825, Creek territory in Georgia and Alabama was ceded to the United States in exchange for an equal amount of land further west. The spokesman for the Creeks was William McIntosh, a mixed-blood chief. Acting without consulting other leaders, he signed the fraudulent Treaty of Indian Springs, which led to his assassination. By 1827, Georgia claimed control of the former Indian property and a state lottery for Muscogee County was held. Land lots were 202.5 ac in size (Figure 6). In 1830 the federal Indian Removal Bill was passed. Fort Mitchell, located on the west bank of the Chattahoochee River, became the main collection point for Indians living in the valley. Between 1830 and 1836 the fort served as the point of embarkation for the long journey to Oklahoma. Due to the fraudulent treaty, a new Creek "war" flared up along the Chattahoochee in 1836. The small settlement of Roanoke, located a few miles north of present day Georgetown, was attacked and burned and twelve of the inhabitants were killed by Indians. Similar raids up and down the Chattahoochee River resulted in regular patrols by the Georgia militia.
13

The Bull Creek Site

The area

around the Bull Creek

site has received

relatively little atten-

tion with respect to

detailed mapping or

historical accounting

over the past two

centuries. Following

the establishment of

Columbus in the latter

1820s, the area con-

tinued to be used pri-

marily for farming.

Much of the area lay

within extensive land

holdings of large

plantations. Wool-

folk's plantation,

which lies to the south

of the Bull Creek site,

is perhaps the best

documented (Ashley

and Schnell 1928).

Later, much of the

land was gradually

Figure 6. Surveyor's plat of land lot containing the Bull Creek site.

mined away for the manufacture of brick.

The area was grad-

The location at the head of navigation

ually developed with the expanded population

on the Chattahoochee River and the presence

growth of Columbus and following the est-

of tremendous water power potential were

ablishment of Camp Benning in 1918. Within

important economic factors for the establis-

the project area south of Bull Creek, dev-

hment of Columbus in 1828. The town was

elopment has not been as extensive.

laid out in a grid system with a long axis

paralleling the river to the west. Commons

From both archeological and historical

were established to surround the town on all

perspectives, an important period of time

four sides. The history of Columbus has been

associated with the project area relates to the

detailed in a number of reports and papers.

frontier community of the late eighteenth and

From an archeological perspective, this history

early nineteenth centuries. During this period

is detailed by Lupold and Schnell 1991; Wood

the area was the point of direct contact be-

1993; Ledbetter 1994. The reader is directed

tween Creeks and white traders and settlers.

to these documents for historical accounts of

During this time the Creek town of Coweta lay

the city north of the Bull Creek site

directly across the river in present-day Russell

boundaries.

County, Alabama.

14

Introduction and Setting

On the Georgia side, settlements

mouth of Bull Creek (Sturges 1818). Maps

composed of Creeks, whites and half-breeds

prepared by DeJarnette (1975:9) and Swanton

developed around Kennard's Ferry located in

(1922: plate 9) show the location of Marshalls

the South Commons of Columbus and

slightly below the confluence. A map

Marshall's Trading Post, which lay south of

prepared by Isabel Patterson of the Bull Creek

Bull Creek. The settlement at Kennard's

area shows the location of Marshall's Trading

Ferry was located at the point where an old

Post as just south of Bull Creek near the

trail crossed the Chattahoochee River in the

confluence with the Chattahoochee River

South Commons area (Lupold and Schnell

(Patterson 1950:34).

1991:13). The settlement known as Wewoka,

which developed around the ferry, contained

Information concerning the trading

several log houses, an inn, and four stores.

post is presented in an early archeological

Possibly, as many as three hundred people

report of the Bull Creek area (Schnell 1979).

may have lived in Wewoka prior to the

In that report, Schnell notes that archeological

founding of Columbus in 1828 (Martin

site 9Me30 is the apparent location of

1874:10). Kennard's Ferry and the old trail

Marshall's trading post. He also makes

appear on the earliest survey maps of

reference to the July 7, 1797 records of

Columbus. The location of Kennard's Ferry

Richard Thomas, secretary to Benjamin

lies just south of present-day Oglethorpe

Hawkins, concerning an Indian trader by the

Bridge. Following the formal creation of

name of Thomas Marshall, "a trader in the

Columbus in 1828 the Wewoka settlement dis-

Coweta Town, very well off in negroes, cattle,

persed (LupoId and Schnell 1991:13).

etc." (Hawkins 1916:467). This reference

infers that Thomas Marshall was the proprietor

The history, as well as the location, of

of Marshall's trading post.

Marshall's Trading Post

is less well defined. A

map published in the

early nineteenth century

(Sturges 1818) shows the

settlement of Marshalls

on the south side of a

large creek opposite the town of Coweta (Figure 7). This stream has been interpreted by many as

--- ---.._- --...

Bull Creek (Brannon

1930; DeJarnette

1975:19). The larger

stream to the south called

Hatchee-thluc-co on the

1818 map is present-day

Upatoi Creek (Brannon

1930). The Sturges map

,

r-'
(

shows Marshalls as a

\)

small group of structures Figure 7. Portion of Sturges' 1818 map of Georgia (published by Eleazer

clustered around the

Early) showing the settlement of Marshalls on south side of Bull Creek.

15

The Bull Creek Site

An unpublished manuscript by Margaret Ashley and Frank Schnell, Sr., provides a somewhat different interpretation of the Sturges map. These researchers spent several days in 1928 digging on the Kyle's Landing site (9Me2), located about 4 krn south of Bull Creek, and recovered European trade material that postdated the mound constructed on the site (Ashley and Schnell 1928:13). Ashley and Schnell concluded that the village site at Kyle's landing was Benjamin Marshall's trading post (Ashley and Schnell 1928:14).
The authors note: In looking over various
maps in an endeavor to identify this site with some historically known village, I noticed that there is a place marked "Marshall's" on plate 9 of Swanton's "Early History of the Creek Indians and Their Neighbors." If the Kyle's Bend site is the remains of this trade village it would be dated at about 1818. This would not mean that the mound was built at that time, but rather that an old village to which the mound belonged had been made into a trading post by Benjamin Marshall. This Indian, a halfbreed, was a contemporary of William McGilivary and Opotheoholo. His name is annexed to treaties of 1832 and from the history of Columbus one can well judge that he played an important part in the early days of the city, as he was one time granted land in Alabama opposite Columbus, on the land
upon which Phoenix now stands (Ashley and Schnell 1928:14).

Patterson's published map does not show the ferry at the end of the old path (Patterson 1950:34), notations on Patterson's 1907 field map do identify the ferry as such.
Aside from the initial date of operation, little is known of the ferry. The owners of the ferry were well-known and influential people. Seaborn Jones was a highly respected businessman and politician who initiated manufacturing at the falls of Columbus in 1828 with construction of a grist mill. Jones also served as a member of Congress (Martin 1874:43).

Another site of some historical importance is the ferry established in 1831 by Seaborn Jones and Stephen M. Ingersoll (Martin 1874:29). According to Martin, a new ferry, located about one mile below the town of Columbus, was established in June of that year (Martin 1874:29). The approximate location of the ferry is shown in Figure 8 on the map of the Bull Creek excavations of 1936. Although

MOliN'
MoVNC
Figure 8. Portion of Patterson's (1950) map of the Bull Creek area. The location of the 1831 ferry is shown at the end of the old path (label is our notation).

16

Introduction and Setting

Dr. Stephen M. Ingersoll was an able businessman who moved to Russell County, Alabama soon after the founding of Columbus. Ingersoll maintained friendly relationships with the Creeks and served as a mediator between the Indians and white settlers (Martin 1873:42). Apparently, Ingersoll died soon after the ferry was established (Martin 1873:42).

The Bull Creek site was a large corn field at the time of the first excavations (Figure 9). However, the Bull Creek site had already been affected by a number of landscape modifications associated with the mining and transportation of clay used for the production of bricks. The remains of a 1920s railroad bed and bridge are still evident across Bull Creek.

The elevated railroad

bed and trestle were built

across the mouth of Bull Creek

in 1926 by the Columbus Brick

and Tile Company to transport

clay from pits south of the

creek to their plant. The

bridge washed away during a

flood in the following year and was rebuilt at a new location at the confluence of Werocoba

Figure9.1936 photograph of the Bull Creek site prior to excavation
showing the abandoned com fields (NPS photograph 2368).

Creek and Bull Creek

(Patterson 1950:37). Borrowing of fill dirt for

survived modern destruction. That area lies

construction of the first bridge apparently

beneath the fill dirt of that narrow railroad

destroyed substantial portions of the prehistoric

spur embankment. Figure 10 shows the site

site (9Me1) at Bull Creek and the Confluence

area as it appeared in the late 1930s.

site (9Me32). There is an ironic twist to this

damaging period of construction. Today, only

one small part of the Bull Creek site has

17

The Bull Creek Site
Figure 10. Portion of a 1938 Soil Conservation SerVice aerial photograph showing the area of the Bull Creek site (area near the mouth of Bull Creek is distorted by photo overlap). 18

Chapter 2 Early Investigations Background to the WPA Excavations

The Bull Creek Site first appeared in the literature in 1925 (Mulvihill 1925:36-41). In the previous year, Frank J. Mulvihill of Cambridge, Massachusetts made a survey of planned recreational areas for the city of Columbus. While studying on the ground for the possible creation of parkways and parks in the Bull Creek drainage, Mulvihill noted and made collections from several Indian sites. Figure 11 shows the locations of sites found by Mulvihill. The Bull Creek site was the first located and was designated Site 1. Mulvihill's short article on his findings gives the following description of the site.
Site 1. At the south end of Tenth Avenue in Columbus is the city cemetery. Alongside the cemetery between it and the river, is a road or wagon path which follows the top of the river bank in a southeasterly direction. The road passes a rifle range and continues on to a pasture where, upon opening and closing a gate, a much frequented path to the right leads to springs at the bottom of the river bluff. But, continuing on the road, a house and farm buildings are reached. To the north of the house the surface of the ground was turned up. In such places were found fragments of pottery. Of the one hundred and twenty-five pieces selected no two seemed to match or come from the same pot. Some few are reddish, most of them are gray, and some are bluish black on one side. All are unglazed, and the degrees of hardness varies. The mixture used is sand and clay. Sharp quartz is evident in some fragments, while a few had shell particles for a binder. There are plain surfaces and ornamented surfaces. Various traces of basket weaving or grass mat patterns distinctly show on some, while others had bands of incised lines and dots, or both, five pointed stars in relief and vertical imprints giving an effect of beading. No fragment is large enough to reconstruct or work out the original design or convey the size and shape of
the original vessel. (Mulvihill 1925:37).

found a possible Indian Mound near the Bull Creek site. While the mound was later determined to be a modern feature (see following), Mulvihill's published account of this mound (Site 5), so close to the Bull Creek site, has been a matter of much confusion through the years.
In 1928 Margaret Ashley, a student of W.K. Moorhead of the Peabody Museum, conducted survey and excavations in the Chattahoochee/Flint river valleys. She was engaged by Mr. Tom Houston of Columbus to make a survey of the Columbus area to find potentially productive sites for excavation. During this survey Ashley and her assistant Frank T. Schnell, Sf. (Figure 12) revisited the sites described by Mulvihill. The photograph showing the two archeologists was taken that year at the Neisler Mound site on the Flint River.

Key TO Lo eAT/oNS
Sihs t wAien /~diOJl
fOlJ."d o~ the $u.rPo.ee
CoLCJ!'1l3CJ.5, GoR.GtA
SPT. 6-q If:;",!

Relies

were

err;
C,..oL.1JMf>\JS
a e.R

Mulvihill's ceramic descriptions match well the types found by later researchers at Bull Creek. Mulvihill also

Figure 11. Map of sites found by Mulvihill in 1924. The Bull Creek Site is site 1 (redrafted map by Mulvihill published in 1925).

19

The Bull Creek Site

specimens were found, though Mulvihill (p. 37), states that pottery fragments were plentiful. He lists this place as Site # 1. Across the creek from this site is still another. It covers a rather large area and offers much pottery and flint material. The dirt that has been thrown up on either side of the railroad bed contains fragments as well. I do not doubt but that this was part of site # 2, though I have
given it the number of site # 6." (Ashley and Schnell 1928: 15).

Ashley's Site 6, later named the Confluence Site (Fenenga and Fenenga 1945), was described in further detail.

It covers a rather large area and offers much pottery and flint material. The site is just at the junction of Bufl Creek and the Chattahoochee River and it was found that the village was heaviest on the creek side rather than the river. This was
also true of Site # 2. (Ashley and Schnell 1928: 17).

Ashley and Schnell also examined Mulvihill's Indian Mound located north of the Bull Creek site. Mulvihill had described the mound in the following manner:

A conical mound used at the

present time as the butt of a rifle range.

Figure 12. Photograph of Margaret Ashley and FraDkSchnell,
Sr., taken during their work in western Georgia in 1928 (Columbus Ledger Magazine, October 21, 1928).

was unable to get an explanation of its origin or why it is so situated. A shape not natural on the plain caught my eye. If it were constructed for use as a rifle range

butt, the handling of the earth to make

Ashley expended relatively little time in examining Mulvihill's Site 1. Apparently

the mound is a better piece of work than I can credit the average worker of today. The soil or marl was scooped from an area of uniform distance

the land had lain fallow since the time of

from the base of the cone. Where successive

Mulvihill's visit and there was little surface exposure. Ashley's report describes the Bull Creek site, which she numbered as Site 2, as follows:

peckings of shot have exposed the earth of the mound back of and above the target pits, an examination shows the marl in laminations. This condition might be expected if the soil were basketed and deposited by dumping successive

baskets during its construction.

A camp site was located in a field above

the right bank of Bull Creek a short distance from

its confluence with the Chattahoochee River.

Because of the dense growth of grass few

20

In relation to the river the mound occupies a site on axis lines of the river so that a view can be had from it south southeast about two miles, and west northwest for about the same distance. It is possible that the Indian was never concerned with it, but the location with relation to the river and the unnatural shape of the plain has not been
explained to me (Mulvihill 1925:38).
The results of the investigations by Ashley and Schnell follow.
There is a small mound on the right bank of the Chattahoochee south of Columbus on the rifle range. This is listed by Mulvihill as Site 5, p. 38, and is described as a "conical mound" possibly of Indian origin. We made two trips, August 24th and 27th, out to the range and on the second visit tested the mound. This was to verify the report that it had been constructed as a rifle butt during the Spanish-American War, the earth coming from the target pit that is at its base. A test pit was dug on the summit and several test holes made at different points on the side. The dirt was uniform red clay with some broken stones. No pottery or flints were found or any other signs to indicate Indian occupation. Because of the sites on Bull Creek being so near, it is quite natural that one would take the mound to be of Indian origin. I am
satisfied that it is not (Ashley and Schnell 1928:15).
During the same period, Ashley and Schnell examined Kyle's Mound, located approximately 4 Ian south of the Bull Creek site. This large site complex, which includes Kyles's Mound and the Cooper site (9Me2 and 9Me3) dates to the Mississippian and historic Creek periods and has produced substantial amounts of pottery contemporaneous with the Bull Creek site occupation (UGA and Columbus Museum site collections). Ashley found the mound entirely washed away by the river and the village site was being seriously eroded. Several days were spent digging, resulting in the conclusion that the village area contained European trade material and postdated the mound (Ashley and Schnell 1928:13). The researchers concluded that the village site at Kyle's landing was Benjamin Marshall's trading post (Ashley and Schnell 1928:14).

Early Investigations
It wasn't until 1936 that the site was again brought to the attention of archaeologists, through the efforts of Isabel Patterson as part of a survey of the Chattahoochee valley. As a setting to the following chapter, the events leading to the site's rediscovery will be examined.
Archeology became intimately associated with the federal relief programs of the early 1930s as a means of employment for large numbers of individuals. Dr. A.R. Kelly, the federally-funded archeologist stationed in Macon, Georgia, coordinated these programs throughout the state through various sponsors. Isabel Garrard Patterson became the sponsor for the Columbus area. Patterson's involvement is detailed in a series of letters written by Kelly and Patterson, which have been compiled as a separate volume (Ledbetter 1995b). Passages that relate directly to Bull Creek are excerpted here. The first identified correspondence from Kelly to Patterson, dated August 26th, 1934, discusses the means by which an archeological project might be instituted in Columbus:
Finally, I am aware that Columbus is the locus of an area in which a number of important documented historic sites belonging to the historic removal of the lower Creeks are found. Particularly do I wish to see the site of the very large Coweta town located near the present Columbus on many of the early maps. On the Chattahoochee, somewhat further removed from Columbus, should be identified the site of a Kashita village, possibly two of them.
Within the week I shall present to Miss Shepparson in Atlanta a rather expansive program of archaeological and historic research. It does not seem wise to deluge her with too many separate projects. What is granted by way of specific local research will largely depend upon the interest taken by the communities involved, their willingness to contribute a small sum of money to defray additional expenses, the labor supply and work relief needs of the area, and the changing set up and purposes to be achieved under the new program of work relief planned for this fall and winter.
21

The Bull Creek Site
At Milledgeville, for instance, the local people are much interested in having a scientific field party explore mounds and historic village sites {Old Oconee Town} near that city. When they have organized local sentiment, provided a sinking fund of two or three hundred dollars to take care of incidental expenses not provided for by government disbursements, and have presented their requests vigorously to Atlanta through their divisional engineer under the usual GERA procedure, it is probable that Miss Shepparson will ask me to investigate, determine the scientific value of the projected work and make recommendations to her office. If they have something of real interest and give substantial indications that they want archaeological work done, it is very probable that a project will be initiated.
That was the way things were handled at Dublin. The Dublin people are much interested and have been for some months. The local D.A.R. and business men are combining to set up a sinking fund to take care of expenses not provided for under the present GERA set up.
I would suggest that you appraise the situation at Columbus, do some discreet "politicing", work with your district and divisional engineers for a specific project to explore the historic Coweta town near Columbus. This seems to me to be the procedure most calculated to get more immediate results. Also it is a plan which will fit in with a program already initiated for the whole state.
I have read most of your articles in the Constitution. They are admirable. I am hoping that another year of expanded research and dignified publicity may result in some permanent organization to carryon such work in this area
(Ledbetter 1995b:1).
A month later, a second letter dated September 28, 1934 was sent to Mrs. Patterson.
The Columbus situation will probably rest until you get back from your trip. Dean Smith of Wesleyan told me yesterday that he has an older brother living on the Chattahoochee who had been interested in a scholarly way in historic Creek towns on that river. He wants to do a bit of reconnaissance. Something may come out of this
(Ledbetter 1995b:4).
22

The steps toward implementation of a Chattahoochee valley survey were spelled out in a letter from Kelly to Patterson dated Christmas day, 1935.
The Chattahoochee Survey is practicable. I would suggest a short range project involving the whole Georgia span of the river and its immediate tributary streams, concentrating on the towns mentioned in history for special reference and exploration. Time, four months; from February to June---possibly to July, depending on the life of the WPA. Supervision: James A. Ford. The project should be written for labor only, to be supplied locally as shovel men are needed on various sites in the different counties. Expenses to be funded locally interested parties need not amount to more than $1250, as follows: 1) Ford's salary underwritten for four months, $150 a month---$600; an assistant---he need not be an archeologist---$250 for the period, about $62.40 per month, oil and gas allowance per month, $75, making $300 altogether; miscellaneous items of expense, including photograph and necessary materials, paper, etc. $100. Grand total, $1250.
If it proves desirable, and the time limit of WPA projects allow, the survey might be carried to July for an additional month of survey for $250, making a total for 5 month survey of $1500. All collections should remain in Georgia, with the proviso that they be properly housed; preferably at the University of Georgia; with a possible proviso that representative samples of materials be sent as a duplicate collection to the National Museum. Since you will be drawing heavily upon Swanton and Setzler, I would suggest this proviso as a matter of policy.
Mr. Woodruff and Mr. Peabody are both interested in the University of Georgia and in the Chattahoochee. They might be willing to underwrite the project for the amount named if you put the matter to them along the lines suggested.
Ford would employ the survey methods used to advantage by him in the lower Mississippi. He would sink small test pits in the village sites; might do some exploratory work in sinking pits on sites where village sites were known or expected to be on basis of historical descriptions, but where river silting had buried the remains two feet or more. Pottery and flint, thus obtained, would be of incalculable value, in working out a tentative archaeological chronology for the Chattahoochee. Later, as time, funds, and proper institutional

guidance permitted, large scale work might be done on key sites as that at Kolomoki. Ford's work would give the archaeologist of the immediate future something to shoot at, a peg on which to set out details gleaned from more intensive work on single sites.
No large number of workmen would be required, perhaps 15-20, in some cases as many as 50; depending on the size of the site, its assumed importance, and the amount of overburden to be removed. Ford will be moving rapidly from site to site, going downstream. He would have to have some flexible arrangement by which the required number of men could be supplied him from WPA rolls in the different counties. I would say, however, that you are not entirely dependent upon WPA for the success of your program. For the same amount of money, $1250, Ford, with one assistant, could carry out a four months survey which would yield valuable results. His energy and drive are tremendous when his enthusiasm is aroused.
Please advise me what you think of this prospectus. Ford is young, his enthusiasm is infectious; he is likeable. I think he would do a great deal to develop interest in archaeology in your section if the plan outlined did materialize. And I think you realize, as I do, that Georgia archaeology will flourish only when interest becomes more general and is not confined to a single locality. That is meant in no way to belittle what has been accomplished at Macon by the inspiration of a few individuals. Put that way, I believe the General and the Doctor and Lint would
agree with me (Ledbetter 1995b:9).
As the time approached for the Chattahoochee work, Kelly was forced to scramble to find a field director as Ford was transferred elsewhere. The problem is mentioned in a letter to Patterson dated February 19, 1936.
I am not forgetting that the primary person to satisfy in the Chattahoochee is you as sponsor. You were willing to have either Ford or Hanna. For the same amount of money we can probably get an experienced field man in archaeology. Possibly Frank Setzler can help us out of our difficulties with a suggestion.
I have only two other men in my organization whom I would trust with the responsibility of helping you in your Chattahoochee

Early Investigations
work, both engineers whom I have trained in field methods during the past two years, Frank Lester and Joe Tamplin. They are thoroughly acquainted with all the techniques of excavation, profile reading and recording, draughting, mapping, cataloguing required in the projected work at or near Columbus. Moreover, they know the ropes in handling NYA and other relief workers, and in making contact with WPA officers. I would sorely miss either of these men but am ready to send you either one you might want if you would rather have a Georgia man or someone local than to bring in someone from the outside (which latter procedure might mean more delay).
It would not be necessary to pay them, either of them, as much as $1 50 a month. If acceptable as assistants to you, and I am ready to recommend them without any doubts, $30 a week would be adequate. Transportation was an item of extra expense to be considered in the field budget whether Ford, Hanna, or anyone else was assigned to the job. I rather hoped that you would be able to find support for your plans in handling that factor.
Frank Lester went to the University of Tennessee. Joe Tamplin is a graduate of Georgia Tech P.S. You probably remember Lester, engineer in charge of Mound D a nice clean-cut
youngster (Ledbetter 1995b: 12).
For nearly two years, Patterson communicated with Kelly and others in an attempt to bring a major archeological project to Columbus. During that time, Patterson backed some level of local site survey, thus providing both Kelly and John Swanton of the Smithsonian Institution with sherd collections. Because several large historic Indian villages existed in the Columbus area, it logically followed that one of these towns be selected for excavation. This all changed in early March 1936, with Patterson's discoveries at the Bull Creek site. Patterson's letter describing the discovery is dated March 9, 1936. Figure 13 shows a portion of a contemporary map with points of reference described in the following letter. A map originally prepared by Patterson has not survived.
23

The Bull Creek Site
My Dear Dr. Kelly: Yesterday afternoon I found the road by the Columbus Air Port was dry enough to get over in my car. It was too muddy and really impassable when Frank was here, so I could not show him one of my mainest treasure finds -
A house in a mound! There is about a foot of top soil then a layer of about 2 inches of black dirt. Just below this are wood timbers, which seem to slant a bit; then at intervals of some several feet are upright wood pieces. In absolute conjunction with this wood are fragments of pottery sticking out of the bank. It has been dug into some and I found a bit of wood loose and about to fall, also a piece of pottery right by it just hanging in loose dirt, so I am sending it for you to see. Please then send it to Frank Setzler with this letter.
This mound seems to have been a house site of some kind - and there is a good portion of it undisturbed - but I fear people are slipping down there and doing some private digging. It is on a corporations property - on the bank of the Chattahoochee - 411 feet north of the mouth of Bull Creek - I stepped it off - (There were 274 of my paces, which measure 18 inches). Height of mound about 10 or 12 feet. I drove to the edge of Weracoba Creek which joins Bull Creek just before it reaches the river.

Weracoba is exactly one mile south of the Air Port Hanger (car mileage 817.5 creek - 818.5 hanger).
There is a large field - full of broken pottery - where plows have thrown it up - which extends from a flat just north of Weracoba, along to its confluence with Bull Creek and on to the river. There are several large bumps of dirt, full of pottery, that may be mounds in this field; - the mound by the river with timbers exposed - is just opposite this field and about 1/1 0 of a mile from the creek - Weracoba. (817.5 creek Weracoba along the new road 817.6 point opposite mounds).
Just opposite the Hangers, where a city golf course is being constructed a mound was leveled. It looked high to me in past years, now it is about even with the Roadway by the Hanger. Some smart city officials we have. Yesterday, I collected some more sherds along here on a site covering 2 tenths of a mile - mileage as measured from Weracoba CreekWeracoba Creek . . . . . . . . . . . . . . . . .. 81 7-5 Point opposite mounds and field of pottery 817-6 Rough farm road to mounds on side toward Chattahoochee River . . . . . . . . . . . . . .. 81 8-0 Old mound (now leveled) but field of pottery on a high flat - opposite the Columbus Air Port Hangers . . . . . . . . . . . . . . . . . . 81 8-3 to 81 8-5

STREET MAP
OF
CCOJLUMIB3U
GEORGIA
o"",-==-::-===,;;;3000 Feet
OFFICE OF CITY ENGINEER ADOPTED BY THE CITY COMMISSION. JAN.l0.1928

1, WeracobaCreek 2. Point OpoositeMoundsand Pottery Field 3. Road To Mounds
4. Pottery on High Flat

Figure 13. Points of reference described in Patterson letter dated March 9, 1936. The reference points are shown on a redrafted portion of a city map of the era.
24

Jim Woodruff got hurt yesterday sprained his knee. He had expected to go to see Dr. Peabody today, but that has to be delayed - so I am arranging to get the money myself. How soon can Mr. Frank Lester come over? I am all set now. Can't you and Mrs. Kelly, Dr. and Mrs. Griffin come over this Saturday? I would like Lester even before that if possible. I have just got permission to go into this Mound that I am sending the wood and the pottery from - and that winds up all the land on each side of the River from Columbus through Fort Benning and from Girard Ala. to the Shagee Creek below Oswickee Blvd. in Alabama. May I hear from you as soon as possible
(Ledbetter 1995b: 14).

Early Investigations
According to Patterson's letter, Frank Lester (Figures 14 and 15) had been in Columbus just prior to the Bull Creek site discoveries. Lester returned to Columbus within a few days and began excavations. Lester's report, which will be covered in the following chapter, describes the excavations at Bull Creek, but little else, particularly with respect to the ongoing "Chattahoochee Survey." Two newspaper articles published in April 1936, provide some details.

Figure 14. Ca 1936 photograph of individuals associated with federal relief';;~cheology. Photograph at Macon
earth lodge shows Kelly, center, and Lester, far right (photograph courtesy Columbus Museum).
25

The Bull Creek Site
Close-up otprevious photograpllSb.6wing view of Frank Lester.
26

Early Investigations

Valuable Finds Spur Interest In Archaeological Explorations In Famed Chattahoochee Valley

Indian Village Site Being Explored and Remains of House Located As Noted Anthropologist Points to Wonderful Possibilities in Area by Isabel Garrard Patterson

The recent visit of Dr. Frank M. Setzler to Columbus has aroused a keen interest in the wonderful possibilities for
scientific excavation in this section of the state.
Dr. Setzler is the acting head curator of the Department of Anthropology of the National Museum at Washington. His opinion is most valuable for he has had years of experience in "field work" in several sections of the United
States and has contributed much information to his chosen profession. He has been promoted
rapidly to his present position on account of his ability and his splendid work which has national recognition.
Dr. Setzler inspected a number of sites in the Chattahoochee Valley during the latter part of February. He has been greatly interested in this locality
for it is significant that any attempt to develop a
systematic procedure for archaeology in the southeastern United States must begin with the knowledge of the locations and movements of the Indian tribes in historic times.
"The present and future archaeological investigations in Georgia must depend to a large extent upon the results obtained from the excavation of selected sites in the Chattahoochee valley. The historically docu-
mented sites in and around Columbus provided they contain sufficient archaeological evidence will enable the anthropologist to reconstruct the aboriginal history of Geor-
gia." This quotation is
made from a letter received from Dr. Setzler, dated March 7, 1936.

Very Favorable Report A very favorable
report coupled with the recommendation started immediately has resulted in the actual beginning of this very important work.
The first site selected for study is just north of Bull Creek on the banks of the Chattahoochee River.
The middle of March, Mr. Frank E. Lester came to Columbus from Macon to conduct scientific exploration and make a survey according to approved methods.
Mr. Lester has been at work at Macon since the beginnings of the project which was inaugurated in December 1933. He is a native Georgian and feels unquestionable pride that all have in this particular section of the United States.
He has been in charge of the very important excavation of the now famous "Council House" of "Old Ocmulgee Fields" which is also recorded as "Mound D" of the Macon group.
Although there are some 800 known mounds in the state of Georgia work has been concentrated at Macon because of paramount interest of persons in this locality.
A National park has been secured for the region termed Old Ocmulgee fields. It is the largest archaeological expedition ever conducted in America.
Equipped for Survey Mr. Lester is
thoroughly acquainted with all the techniques of excavation, profile reading, recording, draughting, mapping and cataloguing required for an accredited survey and proper exploration of ancient village sites and

mounds. Mr. Lester has
been the engineer in charge of the work in Macon and Columbus is extremely fortunate in having been able to secure his services. He is highly recommended by Dr. Setzler and by Dr. Arthur R. Kelly who is the head of the archaeological excavation around Macon.
Last Saturday Dr. Kelly, Mr. Linton Solomon, Mr. Hugh Hanna and Mr. Antonio Waring motored over from Macon to see the work that is being done in the Chattahoochee valley; concentrating on a house site that has been discovered in a village site on Bull Creek.
These men were surprised and greatly impressed by the wealth of material which has been located.
Value of Survey The approved
methods of archaeological survey might seem tediously slow to those who are unaware of the necessity for comprehensive reports of preliminary work. It is necessary however that such information be obtained in order to facilitate the expediency of more intensive exploration on spots which prove to be the key sites by information gathered on the survey.
The site selected is laid off in squares of convenient size so that definite measurements can be made of all materials found in such squares and properly recorded according to their depth and placement with exact relationship to other such material definitely associated with them.
As work progresses, photographs and exact cataloguing of materials and engineering

records are taken along with archaeological interpretations which are placed on such material. Therefore such work properly done can be of untold value for comparative studies by scientists in other sections of the country whether they have visited the particular site in question or not.
Exploration in Valley Exploration of
this nature has been started in the Chattahoochee valley. A village site is being explored and the remains of a house has been located. Excavated portions of this house reveal a hard-packed clay floor with clay plastered timbers for walls and roof. The briquettes show use of grass or other vegetable fibers mixed with the clay to give better binding qualities. Some of these briquettes have been found mounded around sections of charred timber supports and others show lines caused by the lashings which held these timbers in place. Fragments of pottery remain on the floor showing definite signs of human occupancy at the time that the house was destroyed. Beyond the immediate confines of the house there is a very rich midden deposit indicating comparative extensive occupation.
Such materials as bits of bone flint, pottery and those botanical specimens as have been able to withstand the ravages of time all have their proper place in reconstructing the story of this aboriginal occupation.
(Columbus Ledger Enquirer April 5, 1936.

27

The Bull Creek Site

survey included the territory adjacent to

Indian Relics to be Sought in this Area

the Chattahoochee from the famous Coweta falls to the mouth of the Flint River

at the southwestern corner of the State of

Jaycees Will Sponsor Archaeological Survey in Bull Creek Section

Georgia. Each scene of aboriginal encampment was mapped, collections of potsherds and flints obtained and

The board of directors of the Junior chamber of commerce, preceding a meet-

Many think relics are beneath the ground in this vicinity where a burial ground

catalogued for study. It immediately became apparent that a number of the sites visited could not be identified as a

ing of the membership of the and an Indian village are

location of the tribes who occupied this

body last night, voted to thought to have been located.

territory within historic times, so the sites

sponsor an archaeological Officials of the Jaycees hope

of unknown age have been designated by

survey for indian relics along Bull Creek between Columbus

to uncover many valuable relics and at the same time

their present geographical boundaries.

and Fort Benning.
Mr. Lester, archaeological engineer, will conduct the survey for the Jaycees. If the survey shows evidence of man relics in the section, an effort will be made to have the Smithsonian Institution conduct excavations.

learn a great deal about the early life of the people.
Excavation Planned Mr. Lester will make
a number of test excavations. He has already done some preliminary work along the banks of the creek.
(Columbus Ledger April 24, 1936)

The most remarkable of these incidental sites were found in section sixtysix of the historic Coweta Reserve, on a narrow bluff immediately north of the confluence of Bull Creek and the Chattahoochee River, and three miles South of the Court House in Columbus, Georgia. This particular area, so rich in tradition and history is comparatively unexplored and unknown to students of American prehistory, to supplement, therefore, other observations these previously unrecorded

sites are now described as "The Bull Creek

Patterson's column provided the

Village" and the "Bull Creek Cemetery."

citizens of Columbus a "text book" explanation of how scientific archeology was to be conducted at the Bull Creek site. The second

Careful examination of the many local collections obtained from the field adjoining this narrow point of land provided additional infor-

article shows that public support was generated, although the first order of importance was the recovery of valuable relics.

mation for the survey record. The surface material of this field area, now included in "The Bull Creek Village," covers approximately three quarters of a square mile. The northern boundary of this area is

Two Patterson manuscripts provide

the present city limits of Columbus, and the eastern boundary is formed by Weracoba Creek

important background information for the

{see Figure 13 city map}.

WPA excavations. The first is entitled Archeological Survey of the Chattahoochee Valley (patterson 1936), the second Bull Creek Village (Patterson 1937). Excerpts from Patterson's manuscripts endow a quixotic

The collection made by the survey party suggested such a wealth of material that further investigation was warranted and the owners of the land, the Muscogee Trading Company of Columbus, gave permission for this work.

prologue to offset, to some degree, Frank Lester's rather succinct presentation of excavation data which will follow.

A spur railroad was constructed across the western part of the field and extended along the narrow bluff between Bull Creek and the river

An exceptionally interesting Indian site in Muscogee County, Georgia was discovered recently in the course of an archaeological survey of the Chattahoochee Valley. Attention had been increasingly drawn to the significance of the historically documented habitations of the native tribes found in this region and an investigation was made in an attempt to determine the present

by means of a cut through the northern portion and a fill toward the southern. The trestle crossed over the mouth of the creek. The remains of a railroad cut had left exposed an occupational level and evidences of a house site. Pieces of pottery and timbers were found protruding from a dark layer of earth about 12 inches below the present ground level, on March 9, 1936.

surface indications of these sites. The preliminary

28

Early Investigations

The former inhabitants had a fine view of the Chattahoochee and the low terraces along its western bank where the historically famous Coweta Town was located.
Mr. Frank E. Lester, who had assisted Dr. A.R. Kelly in the exploration of Old Ocmulgee Fields near Macon, came to Columbus on the eleventh of March to direct the excavations at Bull Creek. Work started the following day with a small crew.
As the work progressed, local people became more interested in the site, and, fortunately, a number reported finds in this area. Probably the most amazing specimen produced is a spear head, found by Bentley Chappell, Jr., that has been identified by Dr. Frank H.H. Roberts, Jr. {Figure 16}. This generalized southeastern type of Folsomoid material was picked up in the field about four hundred feet northeast of House Site #1 of the Bull Creek Village
(Patterson 1936).

The approximately 11,000

year old Paleoindian point shown in

Figure 16 was the subject of several

letters between Patterson and Kelly in 1935 (Ledbetter 1995b:5) although there is no indication that Bull Creek

Figure 16. Drawing of fluted point from Bull Creek (by Bettye J. Broyles).

was ever viewed as a potentially important Paleoindian site. Patterson viewed the find as highly significant and indicated that a

Seaborn Jones and S.M. Ingersoll established "a new ferry" in June 1831, one mile below the town beyond the South Common. The distant scene

search for similar material led to her discovery of the burned structure at Bull Creek (patterson 1950:37). Patterson's Bull CreekVillage manuscript, which is actually a photographic journal, details the discovery.

beyond the roadway on the right shows "Shepherd's Plantation Gardens" now partly occupied by the landing field of the Municipal Air Port, and Guyner's hill against the horizon, where a residential section began in the late thirties. About 1890, the family of Edward Thompson Shepherd

decided to erect a brick plant in their gardens lying

The first evidences of an Early American house, approximately three miles below the "Falls of the Chattahoochee" and four hundred and fifty feet north of the mouth of Bull Creek were found by Isabel Garrard Patterson, March 8, 1936 (Figure 17). She is indicating these signs of habitation in the accompanying photograph to Myra Bullard Hart and calling it the "Bull Creek Village." Trees in the left background of this photograph mark the remains of an old trail which is still discernable, and where according to written documents

along Weracoba and Bull Creeks, and in 1920 they sold this property to the Columbus Brick and Tile Company. In 1925, this company employed engineers to survey the western part of this area with the purpose of constructing a spur railroad and a trestle across the mouth of Bull Creek to transport materials from the clay deposits south of this creek to their plant. Robert Wiley Smith was in charge of these operations and, in 1926, the rails occupied the site of the present road shown in the photographic record.

29

The Bull Creek Site
':;.,
Figure 17. 1936 photograph of Isabel Patterson at the location of House Site 1 at Bull Creek, view to the NW (Columbus Museum files).
Figure 18. 1936 photograph of IsabelPatterson at House Site 1, view to south towardBull' Creek Cemetery (Columbus Museum files). 30

The pine forest in the left background of this {second} picture {Figure 18} is on the site of Ben Marshall's Trading Post on the south side of Bull Creek and exactly opposite the principal structures of the historic "Coweta Towne" of 1739 on the right bank of the Chattahoochee. The Bull Creek Cemetery was found May 3, 1936 by Alfred Owen Blackman II about four hundred feet south of House Site No. 1 {in the right foreground of the photograph}. The site of the cemetery is not clearly visible, but can be located by the sycamore tree without its leaves left of the roadway, outlining the edge of this picture. The spur railroad constructed in 1926 was abandoned in 1929 as it became unsafe due to flood waters and soil erosion on this high peninsular between the last bend of Bull Creek and the Chattahoochee. The cut through the high bluff in the foreground is approximately fifty feet above the normal water line of the river, and here the road runs straight but across the recently filled land the path extends in undulating curves for over two hundred feet, then proceeds in a straight line to the supports of the former trestle across the creek. The earth borrowed from the high bluff on right and some immediately left of the roadway was used to make the fill between this point and a lower bluff at the mouth of Bull Creek. The two bluffs were called "Indian Mounds," but the Brick and Tile Company

Early Investigations
of the Twentieth Century were not impressed by the fragments of ceramic art which they disturbed. Scattered bits of pottery almost pave the roadway today, but pieces of other vessels found protruding from a dark layer of earth in the embankment twelve inches below the present level of the bluff were the first indication that a part of this house remained unharmed for study. Charred timber and hardened clay were seen during a closer inspection and these offered substantial evidence of Indian occupation.
This {final} scene shows the site of the great "Coweta Town" across the Chattahoochee from the prehistoric sites at Bull Creek {Figure 19}. The trees in the background are on the Alabama side of the river and those with bare branches closer to high bluff as well as the one at the left of this picture are on Georgia's banks. The elevation of bluff above existing roadway is approximately seven feet, and a section of the "borrow pit" composes the foreground.
With Patterson's manuscripts providing a historical setting, Lester's report will now be presented in the following chapter.

Figure 19. 1936 photograph of Isabel Patterson in borrow pit near House Site 1, view to the west toward Alabama (Columbus Museum files).
31

THIS PAGE LEFT BLANK

Chapter 3 WPA Excavations of 1936 and 1937

As a 1930s public works project, a substantial amount of time and manpower was expended on the field work at Bull Creek and subsequent laboratory procedures, such as cataloging of the artifacts and transcribing field notes and records. As a result of this painstaking work, a substantial body of information has been preserved from these six-decade-old excavations. Of course, some of the paperwork has deteriorated over the years and there are problems involved in interpretation of the field and laboratory techniques of the period, but overall, the surviving documentation provides an excellent record of the excavations at Bull Creek. Fortunately Frank Lester, prepared a plan map that shows the areas of site excavation. This map provides the basis for interpretation of the WPA-era excavations at Bull Creek {Figure 20}.
Isabel Patterson's letter of March 9, 1936 to A.R. Kelly, which described her discovery of an aboriginal structure within a presumed mound at the Bull Creek site, brought almost immediate results. In less than a week Frank Lester had arrived and field work commenced. Lester's excavations began in mid-March 1936 and were to continue until early April 1937. During that year Lester supervised groups of children and a number of Muscogee County men in extensive excavations.
Following field work, Lester prepared a manuscript entitled a Preliminary Report of Excavation at Bull Creek Village Site, Muscogee County, Georgia (Lester 1938). The report is one part of a compilation of documents transcribed by National Park Service (NPS) personnel. The documents are sequentially numbered from page 27 to 93. Accord-

ing to a handwritten note attached to the manuscript, the first 26 pages have been misplaced and the original contents are unknown. The main sections of the manuscript include the preliminary report (pages 27-44), transcribed field notes (pages 45-74) and various lists and compilations of artifact and burial data (pages 74-93). The exact date of Lester's manuscript is undetermined but it probably was written in the late 1930s. Copies of the manuscript, curated today in the archeological files of the Columbus Museum, have been made available for this report. The manuscript will be referenced as dating to the year 1938, which is our best approximation of the date of completion. A copy of Lester's field notes is present in Appendix A.
Lester's plan map (see Figure 20) shows two main areas of excavation at Bull Creek. The northern area, which lies just west of a large borrow pit, represents the village excavations where two Mississippian period structures were excavated. Examples of photographs taken in 1936 of general excavation in the village area are illustrated in Figures 21 and 22. Relatively few photographs of the village area excavations exist. Other photographs, which detail the structure excavations, will accompany the text of Lester's report.
The second area of excavation was designated on Lester's map as the burial ground. This is the famous Bull Creek cemetery from which three dog effigy vessels were excavated. Figures 23 and 24 present photographs of the cemetery area excavations. A reconstructed cemetery map showing the distribution of burials and excavation boundaries is shown in Figure 25. The cemetery map has been reconstructed from Lester's field records.

33

The Bull Creek Site
\,
\

\
\
,\
Ii \
\\

1)

SKETCH MAP

Ii
II I !

BULL CREEK EXCAVATION

I

Figure 20. Redrafted copy of Lester's 1936 excavation plan map of the Bull Creek.

34

WPA Excavations
Figure 21. Field photograph of excavations in the village area (NPS photograph 2367). 35

The Bull Creek Site Figure 22. Two field photographs of village area excavations (NPS photographs 2381 and 2370).

WPA Collections Figure 23. Two field photographs of the cemetery area excavations (NPS photograph 2491 and 2414).

The Bull Creek Site
Figure 24. Field photograph of initial trench exca.vation in the
. cemetery area at Bull Creek (NPS photograph 2488). 38

fl .F .

,./.
~
IJ iii

WPA Collections
4-27 4-28
. . 4-25 4-26
I( ., ~"
"0
. . 4-21 4-22

(?)
"$:r '(j~..:::.1 ~
4-1 I 4-12

/

BULL CREEK CEMETERY
,.

ON
A0 0 / /

2m 5 Feel
/

L33 LJI L29 L27 L25 L23 L21 LI9 LI7 LI5 LI3 LII

L9

L7

L5

L3

LI

4-1 4-2 RI

R3

Figure 25. Cemetery map of the Bull Creek site based on Frank Lester's field records.

The Bull Creek Site
Lester excavated Bull Creek using 5 x 5 ft squares, which was a simple and straightforward means of site excavation. However, the stake and grid designations utilized by Lester differ substantially from the standard bicoordinate system of today and requires some explanation. Lester did not utilize a comprehensive site-wide grid. A series of 5 ft-wide trenches, designated Controls 1 through 4, were excavated and subsequent expansions were tied to these trenches. At Bull Creek, Control Trenches 1 and 2 were excavated in the village area and Control Trench 4 was excavated in the cemetery. The location of Control 3, if it ever existed, has not been determined.
Trenches were placed in a generally northerly orientation (grid north) but were not aligned with magnetic north. According to Lester's field notes, the control trenches initially excavated in the northern area were parallel and lay on a line 15 degrees west of magnetic north. In the cemetery area, grid north appears to be approximately 45 degrees west of magnetic north. The extreme divergence from magnetic north in the cemetery area causes problems when trying to interpret Lester's text with respect to burial positions and orientations.
Each control trench was staked at 5-ft intervals, forming a series of individual 5-ft squares. Stake locations were designated first by a Control Trench prefix number (1,2,3 or 4), followed by an individual stake number. Number 1 was assigned to the stake in the extreme southwest corner of each trench. Odd numbers were assigned to stakes along the western line and even numbers were assigned to stakes along the eastern line. As an example, the first square in Control Trench 1 (southwest corner stake) would contain stake number 1-1. The designations for the remaining stakes would be 1-2 in the southeast corner, 1-3 in the northwest corner and 1-4 in the northeast corner.
40

Lester's system became more cumbersome as the trenches were expanded to the east or west. Moving to the east, each 5 ft stake interval was designated by the letter R followed by an odd number (Rl for 5 ft east, R3 for 10 ft east, R5 for 15 ft east, etc.). Moving west, stakes were designated Ll for 5 ft west, L3 for 10 ft west, L5 for 15 ft west, etc.). A fully designated stake number to the east of a control trench would encompass the coordinates of the eastern line of that trench plus the R number (for example 1-6R3). To the west, the control trench stake from the western line would be used with an L number (for example 1-5L3). In this example, the actual distance between the two points represents the distance of R3 to the trench (15 ft), the width of the trench (5 ft), and the distance to L3 (15 ft), for a total of 35 ft. In actual practice, the locations of objects or features found during excavation were identified by measuring from the nearest stake. For example, an object might be assigned a location of 20 inches south and 15 inches west of stake 1-5L3.
Essentially, each 5 x 5 ft square was an individual excavation unit and individual Control Trenches existed independently from the rest of the site. This would raise serious problems for interpretation unless a field map was prepared which shows the exact locations. In the case of Bull Creek, the less than complete map shown in Figure 1 does exist. Most of the squares which were actually excavated can be determined through examination of field notes. Certain relationships, such as the distance between the village excavations and cemetery excavations must be taken at face value as depicted on the field map. One point is clear, Lester's map {see Figure 20} is not a layout of the completed excavations. The map appears to be a plan for excavation which shows the grid orientations of the two excavation areas. By following Lester's field notes most of the actual area of excavation can be recreated. This is one goal to be addressed in this chapter.

Artifacts and notes from Bull Creek were treated in the same manner as those from other sites excavated during the WPA period. Material was transported to the regional laboratory in Macon where a staff of technicians carefully washed and individually numbered each specimen (Figure 26). Artifacts were cataloged on individual provenience cards. All notes were transcribed in triplicate and all photographs individually numbered. Artifacts, such as aboriginal pottery, were subjected to detailed attribute analysis (Fairbanks 1940:27).
Lester's preliminary report, which concentrates on the cemetery excavations and the two structures excavated in the village area, is a condensed version of his field notes. Greater detail is found in his field notes, which have been transcribed and included in the appendices volume of this report. Further insight into Lester's excavations is gained through examination of a series of manuscripts

WPA Excavations
prepared by Isabel Patterson. Patterson's record of the excavations include a report in Early Georgia, the publication of the Society for Georgia Archaeology (Patterson 1950), and several scrapbooks, photographic journals and manuscripts which detail various aspects of the WPA-era excavations. In addition to the published report, sections of three Patterson manuscripts will be cited. These manuscripts are titled Archeological Survey of the Chattahoochee Valley (Patterson 1936), Bull Creek Village (Patterson 1937), and Notes on the Exploration of the Bull Creek Site, Columbus, Georgia (Patterson n.d.). The last manuscript is an earlier version of the 1950 paper.
Lester's record of the 1936 excavations at Bull Creek will be presented under two headings, a report of the village area and the cemetery excavations. The bulk of the information is derived from Lester's preliminary report, supplemented by references to field notes and observations by Patterson.

Figure 26. Circa 1940 photograph of laboratory workers at Macon cataloging artifacts from WPA excavations (from Fairbanks 1940:30).
41

The Bull Creek Site
Lester's report will be presented as originally written with minor editorial changes. These changes include standardization of burial designations and deletion of a few specific references to photographs, maps, or drawings that have been removed from the original manuscript. In certain instances, coordinates have been omitted. Photographs and maps that illustrate various aspects of the project have been added to the original text.
Lester's Village Excavations
WPA excavations in the village portion of Bull Creek commenced in mid-March with Mrs. Patterson accompanying Lester and a small crew of workers. The beginnings of the

excavation are detailed in a letter from Lester to Dr. A.R. Kelly dated March 15, 1936. Figure 28 is a sketch map prepared by Lester to illustrate his initial findings.
Dear Dr. Kelly:
Arrived safely but was unable to reach Mrs. Patterson until Thursday morning. We visited the site on Bull Creek first, made a simple sketch map and that afternoon went over in Alabama to a rather large village site just across from Bull Creek.
Mrs. Patterson was mistaken about it {the Bull Creek site} being a mound, as it is just a rather heavy midden deposit about two hundred feet long, covered with a yellow sandy clay silt deposit.

culflVQfed field

~fJO$ed cuf shows occuootion from (f) fo(l). /-louse debris from@ to@.
Bull Cree/( - Muscogee Co., Cia. FE Lesfer - Mar 19.36
Figure 27. Redrafted sketch map of a portion of the Bull Creek site showing the area of "house midden" (copy of original 1936 map by Frank Lester).
42

WPA Excavations

The house debris is almost in the center of the deposit. All this is exposed in the side of the cut made for a railroad a few years ago. External appearances, very similar to Mossy Oak.
Started excavations Friday (the 13th) with a crew of three negro boys and myself. So far we have made excellent headway, finding quite a bit of material in the very rich midden deposit. The majority of the sherds are extremely small fragments, plain, and very hard tempered. Yesterday, I started hitting baked clay and quite a number of briquettes showing reed imprints. Also the top of what is possibly a fire bowl, however, future troweling will prove more enlightening. It is doubtful if enough of the house remains to get a very complete floor plan for the river has made rather deep inroads from the west and the railroad cut on the east so that only a small portion from eight to twelve or fourteen feet wide is left. However, Mrs. Patterson was interested in getting it out and also it would not seem such a bad idea to start this group off with something that they can actually see.
Everything is progressing very nicely at present, of course, it seems slow after the large crews at Macon but we are making headway. I hope that it will be possible for you to come over
with Hanna when he comes (Lester 1938).

cultivation for quite a number of years and the occupational level is so close to the surface that all signs have been destroyed except in the section marked on the map {see Figure 27}. The underlying soil in the immediate region is a yellow clay silt with a very high percentage of mica in it (Yellow Congaree Silt). Top soil is ordinary grayish tan sand.
The remains of an old railroad cut has left exposed an occupational level and evidences of a house site, briquettes and charred timbers showing in the side of the cut. This area was staked off in five foot squares and overburden removed by
horizontal method {Figures 28 and 29}.

The first portion of Lester's preliminary report dealing further with the village area is presented in the following paragraphs. The area of initial excavation and the "house midden" deposits exposed along the railroad cut are shown in Figures 20 and 27. The cemetery area had not been discovered when this map was produced.

In March 1936 work was begun on an archaeological survey of the Columbus area of the Chattahoochee Valley through the efforts of Mrs. Wayne Patterson and other locally interested people with the co-operation of the National Youth Administration.

The first site selected for investigation was located on the banks of the Chattahoochee River at the point of confluence of Bull Creek. The site is situated on a high bluff, approximately fifty feet above the normal water line of the river, and the surface material of the village covers approximately three-quarters of a square mile. However, the entire section has been under

Figure 28. Field photograph of iiilroad cut in village area also showing road on borrow pit at right and staked excavations on high ground (NPS photograph 2369).

43

The Bull Creek Site

The briquettes themselves are rather massive in size and show the use of grass, straw and other vegetable fibers as a binding agent. It appears that the interior posts of the house were also stuccoed with briquette materials, a large section of one of the interior posts was removed intact from the floor, briquette materials still in place around it.
Working through briquette material {we} uncovered a floor of hard-packed clay with baked areas, presumably patches in spots. Post moulds are from sixteen to twenty-two inches apart and indicate a square or rectangular house with slightly rounded corners. Interior posts suggest cross beams for holding roof supports. Mould impressions on briquette materials show the use of small poles, one to two inches in diameter, bound together with vines or thongs as cross members {attached drawings are referenced in text but have not been located}.
Lester's field notes reference a prepared raised clay area (curb) which was first thought to be a prepared clay hearth but later thought to define the edge of the structure. Figure 32 shows the feature. An excerpt from Lester's field notes for March 14, 1936 describes the feature.

Figure 29. Field photograph of vil1l:l.g~ eXdlvation and pedestaled grid stakes (NPS photograph 2378).

The upper soil level was of the yellow subsoil thrown up by the excavation of the cut, next was plowed soil from the late historic cultivation with a slight trace of sod development just under it. Very little material was found in any of those levels but immediately below this at a depth of approximately eleven inches below surface is a reddish sandy clay level just above the roof debris.
Subsequent troweling uncovered a mass of moulded briquette material belonging to the wall and roof of the house {Figures 30 and 31}. This material takes a decided rectangular form and indicates the remains of a house site twenty to twenty-five feet across and seven to ten feet wide, other portions having been destroyed by railroad c o n s t r u c t i on.
44

Figure 30. Photograph of "brickett" material (impressed daub) from Structure 1 (Columbus Museum files),

WPA Excavations

Figure 31. Field photograph showing fired clay covering floor of Structure 1 (NPS photograph 2372).

Figure 32. Field photograph of the raised clay "curb" within Structure 1 (NPS photograph 2374).

The Bull Creek Site

A baked clay ridge was exposed, 1" deep in red level, this ridge appears to be floor material, and is shaped something on the order of a fire bowl, except that it has a very small degree of curvature .... Immediately adjacent to the baked ridge mentioned above, are other baked areas appearing at irregular intervals and slightly different levels. It is probable that all of these belong to the
wall and roof structure (Lester 1938:46).

of large sherds (Lester 1938:51, Figure 34). There are also additional references in both Lester's field notes and Patterson's manuscripts relating to the burial urn that raise questions concerning the accuracy of Lester's interpretation presented in the preliminary report.

The text of Lester's preliminary report continues with reference to an unusual vessel.

Field note entries for March 19 and May 22, 1936 state:

A small portion of a burial urn was found on the floor of the house covered with fragments of calcined human bones. This and a few miscellaneous sherds were all that remained on the floor but immediately beyond the confines of the house proper is found a comparatively rich midden accumulation about ten inches thick, also several refuse pits were found close by containing quite a number of sherds, charred animal bones, and broken bits of stone. At the present time the floor proper has not been cut through as there are no indications of floor pits
(Lester 1938:2728).

Started troweling floor section and uncovered a piece of broken pottery, laying flat on the floor, with a few fragments of human bones in it. This is located just on the edge of the bank {road side} 19": S - 16" E of sta. 1-50 (Lester 1938:47).... Also found some small calcined bone on floor of house site between Sta. 1-47 and 1-48 to 1-49 and 1-50. Bones are very tiny and have been burned completely through (Lester 1938:51).
Chattahoochee River
Edgeof BOA/( 50'titrp

This concludes Lester's report on Structure 1. Figure 32 shows Lester's map of the postmolds, fired clay features and one large pit associated with the structure. An examination of Lester's field notes provides further insight into the structure. The large pit adjacent to the structure produced a number

~5dia.
cr:"...

+r-ss

+'-51

+'-ss

,LfOUSE SITENO. t

BULL CREEKEXCAVATION

'*- S'Q'~ 2.5'

July I!JJ6

Borrow Pit

Figure 33. Floor plan of Lester's Structure 1 (redrafted copy of Lester's original drawing).

46

WPA Excavations

~
,.:/.;.
~;:~:// -> ": .

. '.7-:'
. ~: .. :::~
.. ~.:
... -.~
. . ' "

'<;,..<., .".:.

o

3

e-.

i

em

Figure 34. Photographs of two vessel fragments (complicated stamped and plain bowls) from trash pit adjacent to Structure 1 (NPS files).

47

The Bull Creek Site
Interestingly, Lester's reference to the urn does not indicate the bones were calcined while the second entry concerning burned bones does not indicate that they were human. Apparently the bone was not saved. Figure 35 represents a composite drawing of the two areas of bone based upon a sketch from Lester's field notes.
Patterson referenced the burial urn in one manuscript and mistakenly inferred the presence of a cremation (Patterson 1937). The photograph is reproduced in Figure 36 with her caption.
According to field notes, additional work took place around Structure 1 during May 25 and May 26, 1936: Continued work on
#1 house site; more evidence of post mould impressions. Removed moulded briquette from floor of house ... Very little pottery showing up on house floor. Floor still shows hard-packed clay variety, with layer of charcoal just over floor. What posts remain in holes are partially charred, but what fiber remains uncharred is totally rotted.

,/-4r-7-------------------/-i50 I

I

I

I

I

I

I

I

I

I

I

I I I

Calcined Bone

I

IL

~

/-47

/-48

Figure 35. Plan of a portion of Structure 1 showing the location of urn and calcined bone (adapted from field sketch maps).

It appears that the posts charred, all around the outside and the portion still remaining in natural state is the center of the charred portions (Lester 1938:51 ).

Along the eastern edge of the embankment ... a small portion of a burial urn was found on the floor of House Site No. of Bull Creek Village. The broken urn of hand molded clay pottery with pinched rim was covered with fragments of calcined human bones. The larger piece of pottery on the right is the identical one found protruding most prominently from the side of the high bluff above the roadway in the preceding photographs ...The knife at left in this picture shows the approximate size of the broken pieces of the urn (Patterson n.d._:photograph 12)
Figure 36. Field photograph of possible burial urn from Structure 1 (NPS photograph B.C. 12).
48

WPA Excavations

The remains of a second structure were also found at the northern end of Lester's excavation. Lester's preliminary report deals with this second structure only sparingly. Significantly, several smashed vessels were discovered on the floor, providing important data not recovered from the first structure. One factor probably limited the amount of time allotted to the excavation of the second structure. The Bull Creek cemetery was discovered at nearly the same time and apparently diverted attention away from structure excavation. It is unclear just how much of the structure was actually excavated. Lester's conclusion that postmolds were absent probably reduced the interpretive importance of the find and may have been one reason for the limitation of effort expended. Excavations conducted at the site more recently indicated that postmold patterns survive on the site even when other structural remains have been plowed away (see Chapter 7). Fortunately Mrs. Patterson provides additional details. Lester's account of the excavation of Structure 2 follows.
About one hundred and sixty feet north of house site number one the ridge widens out considerably and test trenches were dug in this area. The soil levels remain essentially the same except that this section has been under cultivation for the past few years. The plowed soil contains quantities of sherds and there is a trace of an occupation level immediately below the plowed soil; occupation level runs out between stations 1115L5 and 1-119L5.

evidences of post mould impressions are gone {Figure 37-38}. From the remaining briquette material it appears that wall construction was very similar to the first house but lack of post moulds make any architectural floor plan impossible.
Further exploration of the immediate area failed to yield any information and Control #2 was excavated just across the railroad cut from the above houses. Intensive plowing had destroyed all evidences of an occupation level. Pottery fragments are found mixed with late historic material and one refuse pit was found containing quite a bit of material but otherwise there is no definite sign
of occupation (Lester 1938:29).
Based upon a list of the special finds recorded in the field notes, portions of several restorable vessels were found on the floor of Structure 2. The list included the following vessels:
Find 37. Broken pieces of a pot, found on floor level of Structure 2. Bowl and rim found 42" east and 7" north of station 195Ll. Bottom found 32" east and 14" south of 1-95Ll. Pot approximately 9" in diameter, plain design, with fluted rim.

In this area was found the remains of another house but constant plowing had damaged it so much that it is impossible to get any comprehensible floor pian. Briquette material, several large pots, and fragments of charcoal, resting on what was evidently the floor and small baked areas give some clue as to approximate size but all

~;i~~f~~ili!~[~~~i~~~!~~;\!:j:l\~-:::".=:..:,::::-'.'
Figure 37. Field photograph of three smashed pots on floor of Structure 2 (NPS photograph B.C. 22).

49

The Bull Creek Site
Figure 38. Field photograph showing workerS Clearinglhe floor of Structure 2 andfhelocatioDS
of shattered vessels (NPS photograph 2382). 50

Find 38. Pieces of a large pot, notched rim, incised and punctated design on body of pot. Pieces found grouped together on floor level of Structure 2 between stations 1-97L1/L3 and 1-99L1/L3.
Find 39. Pieces of a small pot (approximately 9" diameter) resting on floor level of Structure 2, 7" west and 31" south of station 1-99L3. Pot has notched rim and plain body. Body notched due to firing.
Find 40. Several pieces of a large pot found on floor of Structure 2. Found 24" south and 18" west of station 1-99L3, 13" below surface.
Find 41. Portion of broken rim of large pot (13" dia.). Found resting on floor level of Structure 2, 13" below surface, 8" west of station 1-99L3. Pot shows painted or beaded rim and circular stamped design.

WPA Excavations
Find 42. Portion of rim of large pot (13 " dia.). Found resting on floor level of Structure 2, 13" below surface, 13" south and 18" west of station 1-99LE.
According to National Park Service records, only one of the vessels (Find 38) was restored, although a second restored vessel was found in the collections. The second smaller vessel (Park Service No. 39-22151) apparently was not assigned a Find Number, but is catalogued as found on the floor of Structure 2. Figure 39 shows the smaller restored vessel. Figure 40 shows the partially rereconstructed Find 38, a large bowl which had been restored but has since deteriorated.
A search of collections to relocate other restored Structure 2 vessels was unsuccessful. However, the descriptions suggest these vessels are large bowls similar to Find 38.

cm Figure 39. Modern photograph of restored vessel (ace. no. 39-22181/22151) from Structure 2.
51

The Bull Creek Site

o3
~
em

Figure 40. Photograph of restored vessel (Find 38) from Structure,2 (photograph from files of Columbus Museum).

Lester's field notes provide some additional insight into Structure 2. The entries for May 7-18 include:
Continued work on House Number 2 with very small group of workers. Removed pottery from floor level and continued on down. Ground so hard that necessary to keep wet all the time. Floor appears to be very uneven, sloping toward the southwest. Found baked area in S.W. corner at least 3" below level of that in N.E. corner. Still finding some pottery in floor level and traces of briquette material. Very few post moulds have turned up at the present time. Floor area is very
52

uneven, showing patches of baked material in spots. All indications point to use of hard packed clay floor, with patched places baked in place (Lester 1938:50).
Relatively little information may be gained from examination of Lester's field notes with respect to the results of excavation elsewhere in the village area. Figure 41 shows the total area of Lester's excavations based upon field note entries. These entries are not complete and do not, for instance, define the extent of excavation of Control Trench 2 or the location of Trench Control 3.

WPA Excavations

--
Figure 41. Reconstructed plan of Lester's village excavation based upon field note entries.

Cemetery Excavations
The discovery of an aboriginal cemetery at Bull Creek in 1936, rich in exotic burial offerings rapidly stimulated the interest of Mrs. Patterson, the citizens of Columbus, and, for a brief while, the archeological community. The existence of the cemetery was unknown when Lester's excavation began. The cemetery was buried beneath the railroad fill near the mouth of Bull Creek. Ironically, the construction of the railroad levee which elsewhere destroyed portions of the village had actually helped preserve the cemetery area by capping the deposit and reducing the impacts of erosion and looting. Figure 42 shows the depth of excavated soil in the cemetery area.

Figure 42. Field photograph showing excavation of the cemetery, view to the north. Note the location of road in upper left of photograph in relationship to the excavation line (NPS photograph 2485).
53

The Bull Creek Site Lester's preliminary report provides
some information on each of the burials excavated at Bull Creek. Unfortunately, most burial descriptions are minimal. The field notes contain more information, especially in the form of field sketches of many, but not all, of the burials. Figure 43 illustrates examples of field drawings. Photographs exist for a
1-/3

large percentage of the burials but many photographs are of poor quality. As one facet of the present study, the burial remains, curated at Columbus Museum, were examined by Chad Braley but an exhaustive osteological analysis was not conducted.

I, .. -.
Loca.f;~n aur;'~( :# S i

~, 1fJ ~\

._.~ litt,

J

~

b ~\'
" ............&r

~

"C"

~

~

~~_I,!II~

\Q
~

,,
!Y
t

Figure 43. Four examples of burial field drawings from Bull Creek cemetery (redrafted from field notes). 54

WPA Excavations

The composite map of the Bull Creek cemetery (see Figure 25) is not based upon an original field drawing. The cemetery map represents our redrafting of a map in the Park Service files which appears to have been produced using the individual burial sketch maps in Lester's field notes. There is no indication that a complete cemetery map was ever produced by Lester. As previously noted, there are problems concerning the burial orientations furnished in Lester's report. Based upon the Bull Creek base map, supposedly produced by Lester (see Figure 20), grid north for the cemetery actually lies nearly 45 degrees west of magnetic north. This alignment appears to be confirmed by the photograph shown in Figure 42. In that photograph, the line forming the boundary of the cemetery excavations intersects the roadway of the old railroad corridor at a noticeable angle. If this point is correct, all references to orientation should be adjusted by approximately 45 degrees (Figure 44). The text of Lester's report uses grid north bearings for burial descriptions.

bones sticking out of the bank and in great excitement reported their find to Mr. Lester. Naturally the search for other evidence of skeletal remains commenced on May 6, 1936. Pits were carefully excavated by the more expert of the N.Y.A. group of workers in this area (Figure 45) although some were left to complete the work on the House Sites of "Bull Creek Village" ... The displaced skull was listed as Burial #1 even though it was removed from position by the person who
found it (Patterson n.d.).
Lester's preliminary report provides fairly detailed descriptions of many of the burials. Contemporary photographs and sketches, sections of the cemetery plan map, artifact illustrations, and some details from Patterson's manuscripts are inserted into the text of Lester's report.

The events leading to the discovery of the cemetery are not mentioned in Lester's report but are recounted in detail in Patterson's writings. The following excerpts are from Patterson's original version of her 1950 paper.

About four hundred feet south of House Site #1 on the bank of Bull Creek two burials were found about 100 yards from the creek's confluence with the Chattahoochee. Mr. Alfred O. Blackmar III and some friends were having target practice on what they thought was a gourd, but when they crossed the creek to examine their marksmanship, they discovered they had been shooting at a human skull. This had been washed out of the bank by the floodwaters during the winter of 1936 and only the skull remained in the yellow clay muck. They examined the bank above and saw a portion of the lower leg bones and thigh

/
Figure 44. Enlarged view of Lester's Bull Creek base map showing layout of the cemetery grid. Note difference in grid north and magnetic north.

55

The Bull Creek Site

Lester to compass directions are referenced to grid north. The quotation includes the complete text of Lester's cemetery excavations.

About four hundred feet south of house site number one a burial was discovered by a visitor. The body had been washed from position by the high waters of Bull Creek and only the skull remained. Further investigation yielded the tibia and femur of another burial protruding from the bank. Control Trench 4 was staked and excavation was begun in this section.

The profile at this point shows about three inches of water-laid sand, four to five inches of yellowish clay fill from the railroad bed, historic sod development seven to ten inches, ten to twelve inches of accumulated tan sand, Indian occupation level, four to six inches thick, tan sand about twenty to twenty-six inches and underlying yellow clay loam with heavy silt deposit. The burial pit first showed about six inches above the clay loam and the burial was located at a depth of sixty-seven inches or about two inches in yellow clay loam.

Figure 47 shows a profile drawing of the cemetery originally prepared by Lester. The drawing may represent the western edge of the original trench excavation.

Figure 45. Field photograph of the "more expert N.Y.A. workers"

Burial 2 is of the contracted

at the cemetery (NPS photograph 2395).

flexed type, the body is laying on the

right side, oriented east and west with

As previously noted, the locations of all burials are not precisely known. Figure 25

the head to the west. The bones are in a rather poor state of preservation, all of the left side of the skull being gone. None of the cervical vertebrae

represents a revised version of a NPS file composite map of the cemetery upon which we have added approximate locations of missing burials using field notes. Locations of a few burials could not be approximated. For the following descriptions, the cemetery map has

and only the lower five joints of the thoracic vertebrae in place. The knees are drawn up almost to the thorax and the arms are folded with the left hand going between the legs about midway between the knees and foot. The skull is rather long in proportion to the width but the fact that it is partially crushed may cause this to look more

been cut into two halves to show greater

pronounced than it really is.

detail. Figure 46 shows the eastern half.

Again as previously noted, any reference by

56

WPA Excavations

4-25 4-26

Bl8
u~ .~ 0 <J

~ B19 !)

B5 11l\t'~""

4-21 4-22











~ ,= 1) iY~'=B37

B35



.

B29!

B8~~

Q/ B7
~ ~'i,.





t ~


B27
1,?

j
~
Bll

~



~-~

7J~o
O'\~)B6

Bl5 &f1. e

~.~'@ Bl4

B13 ff.~ B22

;;:. .@~
B28
.". ~
B26 (?)

@)
BlO

Bl6

~
""'"

:..


~:~.I

~



e

.B3
~
~k

..




.

.

B25

.

.

4-11


4-..12

c;::::.:~ ..

.z B30"\'tl',~0ri ~~

.;~::'

::.:~~:> B17 ll'~~

"t
ei Bl2


<i"~B2
.

B31







~o.~;:'

.

If e1



> /'.

/

Q

1
~q

B32

e

B24

,/

./



4-5
:,/

.

/

/

'4-. 6

.

B34
It, "'"

./

e

. ~ B33

~

.

..

/'

/' ./ /'

4.-3

4-. 4

.

/'

/'

./

/'

L13 Lll L9 L7 L5 L3 LI 4-1 4-2 RI R3

Figure 46. Reconstructed plan map of the eastern part of the cemetery containing Burials 1-33 (dotted outlines are approximations).

57

The Bull Creek Site

4-7L5

" -9LS

'1-15L5

q -I7LS

Q-/9/S

q'21L5

.-23L.5

q-2SLS

Q'27LS

@ water -Iaia sand
yellow clay @ histone sad
@ tan sand
@ occopation level
(j) tan sand @ yellow loam

... ,.... _ -.:.\:
Typical Profile Bull Creek Cemetery

Figure 47. Redrafted profile map of the cemetery area (based upon original 1936 drawing by Lester).

Extension of the trench brings forth another burial about six feet north of Burial 2 (see Figure 46). The body is in an excellent state of preservation, laying on the left side, oriented north and south with the head to the south. It is the same type of contracted flexed burial as number two with the knees drawn up to the face. Both hands go between the legs, right hand about four inches below the knee and left about five inches above the ankle.

There are some indications of post mortem movement of the body, one of the lumbar vertebrae being about six inches below the pelvis, and some of the bones of the upper body being slightly displaced. Burial furniture consists of a few beads of pierced seed {shell} around the neck and a painted dog-effigy type of water bottle {Figure 48}.

58

WPA Excavations

Mrs. Patterson considered the Burial 3 find so remarkable that the entire burial was removed intact (Figure 49). The burial was carefully removed to the lobby of the Merchants and Mechanics Bank of Columbus for exhibition during June and July. Following the exhibition the burial was moved to the U.S. Post Office until it could be placed in Columbus' new museum (Patterson 1950:39).

Lester's report continues:

In extending the excava-

tion northward to remove Burial 3

another burial (Burial 4) was

discovered about four feet to the

north at a depth of forty-four

inches below the surface and only

twenty inches below the bottom

of the occupation level. A portion

of the skull and pelvis were de-

stroyed by the laborer, but such

bones as remain are in a very

good state of preservation. The

body is laying on the left side

oriented northwest and southeast,

of with the head to the southeast.
Figure 48. Field photogra.ph the Burial 3 and first dog effigy vessel The position is a slightly different

(NPS photograph 2396).

type of flexed burial in that the arms are straight at the sides,

The bottle is in a perfect state of preservation and is placed just between the skull and the knees. It is in the shape of a small dog with a circular body, short legs, stubby tail, and the facial characteristics are accented by the use of the paint. The design on the body of the vessel is a

vertebrae straight, legs folded under the body, pelvis to knee at right angles with the vertebrae and feet only a few inches below the pelvis. To all appearances this is the burial of a female. No burial furniture associated with this burial at all.

double scroll or spiral with annular bands around the dog's tail, the neck of the vessel is decorated with spiral vertical lines coming down onto the body of the vase.

Excavated a 15ft wide trench north 25 ft from Burial 3 to a depth of seventy-eight inches at stations 4-23L1. Pottery occurs in quantities in all levels down to occupation level and a few sherds

This burial was removed in situ by pouring

have worked into tan sand below.

a concrete moulding around the block of soil on

which it was resting and by slipping a steel plate

under the form. The approximate weight of the

exhibit is between seven hundred and fifty and one

thousand pounds (Lester 1938:31).

59

The Bull Creek Site

SKELETON OF INDIAN IS ON DISPLAY HERE
The skeleton of an Indian, who apparently lived in a settlement in this vicinity some 500 years ago, recently discovered on the banks of the Chattahoochee east of Columbus, is now on display, just as he was buried, in the lobby of the Merchants and Mechanics bank.
Excavation work has been carried on in this section for several weeks under the direction of Frank E. Lester who has had the cooperation of the National Youth Administration and the Smithsonian Institute, as well as that of several interested Columbus citizens. The present find constitutes an important link in the reconstruction of burial and cultural customs of the people who inhabited this part of the Southeast hundreds
of years ago (The Columbus Ledger 13, 1936).

upper left: burial supported by concrete slab upper right: burial is boxed and hoisted away lower left: burial is transported to Columbus lower right: display is complete
Figure 49. Series of 1936 NPS photographs (accompanied by newspaper article) showing the removal of Burial 3 for exhibition.
60

Excavated Burial 5 which is of the contracted flexed type, laying on the left side, oriented south southeast, north northwest, head to south southeast {Figure 50}. The remains are in a very poor state of preservation, all of the skull gone except upper and lower teeth - no trace of cervical vertebrae or scapular. Right arm folded across the body, left arm straight, going between the femora with hands at heels. Burial furniture consists of a small greenstone celt about seven inches long. This was stolen by a visitor on the job. Burial 5 was found directly under station 4-21 L3, thirty-six inches in tan sand, seventy-two inches below surface.
Four burials (6-9) were initially found in this section. Burial 6 {Figure 51} was found directly under station 4-17L 1 seventy-eight inches below surface and forty-two inches in tan sand, no trace of a burial pit. This is the flexed type of burial, body laying on right side, oriented northwest and southeast with head to the northwest. The femora are at right angles to the vertebrae and the feet are folded back almost touching the pelvis. On the whole the bones are in a rather poor state of preservation and were considerably damaged after discovery by heavy rains. Portions of the pelvis, both femora, patellas, and tibiae; both hu-

WPA Excavations
rneri, one of the ulnar, both clavicles and scapulars, and four points of the cervical vertebrae remain. The skull has been rather badly crushed and only a few fragments were salvaged. Burial furniture consists of a small greenstone celt about four inches long, two small greenstone chisels, a black flint knife of very excellent workmanship, several pieces of worked flint of no definite shape, and a mass of red pigment, apparently some ferrous oxide used as paint. From the position of the pigment it appears that the paint was probably in a small pouch of some kind.
Burial 7 was found forty-two inches in tan sand, seventy-eight inches below surface {see Figure 51}. The remains are in a very poor state of preservation, the body is laying on the left side, oriented northwest and southeast with the head to the southeast. The remaining bones consists of portions of the skull, right fibula, both femora and tibiae. It appears that at least one of the arms went between the legs as fragments of the right ulnar and radius were found in such position. Another of the dog-effigy type of painted vessels and a small greenstone discoidal stone are found associated with the burial {Figure 52}.

Figure 50. Field photograph of Burial 5 (NPS photograph 2406). 61

The Bull Creek Site

Figure 51.

photograph of Burials 6 in foreground and Burial 7 inbackground upper right (NPS

photograph B.C. 49).

The vessel is identical with the one previously described, except that it was rather badly broken by ground pressure. All of the parts were in place and the vessel has since been
restored (Lester 1938:33).
The second dog pot also received some public attention. A short newspaper article in the Columbus paper described the find and following restoration the vessel was put on display alongside Burial 3 and the first dog pot.

Figure 52. Modern photograph of second dog pot.
62

Another Piece of Indian Pottery Found Along Banks of Chattahoochee

A strange looking effigy vase in the form of a dog with a painted design encircling the vase, apparently representing the horizon, was found recently near the Bull Creek site on the Chattahoochee river.
It was discovered about twenty feet from the skeleton now on display at the Merchants and Mechanics bank, and was also with an Indian skeleton.
Judging from the design and shape of the vase, or water jug, it was buried about 500 years ago, and was perhaps the favorite water bottle of the Indian. It is also the opinion of the workers that the design on the vase must have been first put on with some chemical known to them and then burned in the clay. At any rate, the design

is still plain, and the paint is black.
This is the second vase that has been found near the Bull Creek site on the banks of the Chattahoochee river, and is one of the four now known to be in existence. The other two vases were discovered in North Georgia and near the Flint river.
The first vase unearthed here was also buried with an Indian, the body having been buried in a flexed position, with the jug between his knees and his head. The design on this bottle is also very distinct and is the same as that on the newly discovered jug.
Both water bottles are on display at the Merchants and Mechanics bank, together with the skeleton.

Columbus Ledger July 21, 1936

WPA Excavations
lower body consists of parts of the pelvis, both femora and tibiae, and small fragments of the fibulae. Traces of the metacarpal bones between the femora. Apparently this is of the same contracted type of flexed burial as number eight. The body is laying on the left side oriented north and south with the head to the south. And is located sixty-six inches below surface, thirty-two inches in tan sand. Burial furniture consists of a small greenstone chisel three and a half inches long.
Burial lOis located south of Burial 9, fifty-four inches below surface, twentyseven inches below the occupation level. The burial consists of a few crushed fragments of a skull, no other traces of the burial being evident. There is no sign of a burial pit or outline of any of the decayed bones.
In working the section thus far investigated down to natural formation a few more burials have been located {Figure 54}. These were made from the same occupation level as those above, but at slightly different depths. On the whole, there is nothing to indicate any difference in the time interval or in the methods of burial.

Burial 8 is of the contracted flexed type, the body laying on the left side, oriented east and west, with the head to the east. The body is directly under station 4-19L3 at a depth of sixtyeight inches below surface, thirty six inches in tan sand. The face is exceptionally long in proportion to the size of the skull and the skull appears rather long for its width. The cervical vertebrae, portions of the scapulars and both clavicles remaining intact. The left arm is folded across the body with the hand resting at the pelvis, the right arm is straight, going between the femora with the right hand below the right tibia. Pelvis, tibiae, femora, fibulae, three sections of the lumbar vertebrae, and the metatarsus bones in excellent state of preservation. No trace of burial furniture associated with the burial.
Burial 9 is in a very poor state of preservation, nothing remaining of the upper body except the teeth, the cervical vertebrae, and a small section of the right humerus about four inches long,

Figure 53. Field photograph of Burial 8 (NP8 photograph 2415) .
63

The Bull Creek Site

;J,~"

.....................

/

:~:( ..

,;~

1:::::.<, ~

FigUre 54. FieldphOtograp

1)f:f)1~ph B.C.Ci2), band-tintoo for Patterson.

Contrary to Lester's description, the burial lies on its right side.

64

WPA Excavations

In working the section west of Burial 6 another burial (number eleven) was found at a depth of seventy-three inches below the surface, thirty-eight inches below the occupation level in tan sand. This is the flexed type of burial, oriented south southeast and north-northwest. The arms are straight at the sides extending almost halfway down the femora. The femora point straight downward with the feet drawn up almost touching the pelvis. The bones are in a fairly good state of preservation, a few of the ribs and the metatarsal bones remaining in place. The skull is rather small and the individual was of comparatively small stature. This is a female burial.
Burial 12 {see Figure 54} consists of another of the contracted type of flexed burials, the body is laying on the left side oriented northwest and southeast with the head to the northwest. It was found south of Burial 4 at a depth of sixty-nine inches below the surface, thirty-five inches below the occupation level. The remains consist of fragments of the skull and portions of the femora and tibiae. The body is in a very poor state of preservation, but enough remains to indicate that the legs were drawn up almost to the thorax, and represents the typical type of contract flexed burial.

an interesting pathological enlargement. A small depression is present at the lambdoid intersection. It is the burial of an old adult, male individual {Figure 55}.
Burial 15 found just a few feet west of Burial 6, was found seventy-eight inches below the surface, fifty-three inches in tan sand. The body is in a very poor state of preservation, oriented north northwest and south southeast with the head to the south southeast. It is the contracted flexed type of burial, arms appearing to go between the legs, but it is impossible to determine any exact relation, body laying on the left side. The remaining bones consist of the skull which is rather badly crushed, short sections of the humeri, traces of ulna and radius of right arm, fragments of the pelvis, and only traces of the femora and tibiae. No trace of the fibulae or any of the appendages
(Lester 1938:36).

Burial 13 was found immediately below a large tree between stations 4-17-418 and 4-15-4-16. Nothing is left except portions of the skull, a few of the teeth, and a few sections of the cervical vertebrae. From the remaining fragments it appears that the body was laying on the left side, oriented north and south with the head to the south. There is no trace of a burial pit or burial furniture. This burial was found forty inches below the surface, twenty-six inches in tan sand.

Another burial (Number 14) was located about ten feet to the north in the same section as number thirteen but at a slightly different level, being fifty-one inches below the surface, thirty-four inches in tan sand. It is of the contracted type, laying on the left side, oriented east and west with the head to the east. The knees are drawn up toward the body with the arms passing between the femora and the hands at the tibiae. On the whole, the body is in a good state of preservation and the skull is in exceptionally good shape. The skull is slightly misshapen, the right parietal shows

Figure 55. Field photograph of Burial 14 (NPS photograph 2420).

65

The Bull Creek Site

Lester's preliminary report mentions Burials 16 and 17 but omits descriptions of these two important burials. The report does mention burial artifacts, a third dog pot and an effigy pipe following the discussion of Burial 20 (page 37) but the artifacts are not referenced to burial number. For the sake of convenience those paragraphs are inserted at this point along with the field note descriptions of the two burials.

Burial 16 consists of a few fragments of tibia and femora, bones are scattered over a large area and this is really the remains of two bodies; general orientation is south southwest and north northeast. Burial furniture consists of the 3rd effigy dog vessel {Figure 56}. Found 36" below
occupation level (Lester 1938:60).

The vessel is slightly different from the

other two in that the neck is higher and smaller,

the designs being painted black on a red back-

ground, design slightly different, and the legs are a

little longer and at a slightly dif-

ferent angle (Lester 1938: .37).

Figure 56. Field photograph of Burial 16 and

third dog pot (NPS photograph 2431).

Burial 17 is of the con-

tracted flexed type, oriented

NW-SE head to NW. Body

laying on right side with the

knees at right angles to the

vertebrae and both arms going

between the legs. Skeletal

remains in rather poor state of

preservation, one side of skull

crushed and all of metacarpal

and metatarsal bones gone.

Only a few joints of the lumbar

vertebrae remain but 5 of the

thoracic ribs. Portion of skull

was all that could be removed.

Found 70" below surface, 40"

below occupation level in tan

sand. Burial furniture includes a

very remarkable human effigy

pipe {Figure 57}. This pipe is a

remarkable piece of work, facial

characteristics are typically

Indian, and the whole expression

has been captured by the artist.

A tobacco "cake" almost a

sixteenth of an inch thick still

remains inside the pipe.

Figure 57. Photograph of tobacco pipe from Burial 17 (Columbus

Museum files).

66

WPA Excavations

Burial 18 is located directly under station 4-24, sixty-one inches below surface, thirty-nine inches below the occupation level, and twelve inches in yellow clay loam. The body is laying on the left side oriented southeast and northwest with head to the southeast. The knees are drawn up. We have here a typical Bull Creek type of contracted burial, the remains consisting of a few fragments of the skull and sections of the femora {Figure 58}. Burial furniture associated with the burial is a small bowl, approximately three and a half inches in diameter with a design of incised scrolls combined with punctate markings, several large beads of shell, a small piece of conch shell core, and an unbroken clay pipe of the human effigy type {Figures 59 and 60}.

Bones

Pipe.
BOW/~
+
Sta 4 -21f

c:Sk.t.///

Figure 58. Field map of Burial 18.

3
i
em
Figure 59. Photograph of small bowl found with Burial 18 (Columbus Museum file photograph).
67

The Bull Creek Site

Although the Bull Creek cemetery is by no means completed we may briefly summarize the facts known about the village from what has been accomplished thus far. Flexed burials of two slightly different types represent the burial custom of the inhabitants. However, there is no consistent orientation. It appears that the contracted burials must have been lashed into position, but there are no traces of coverings of any sort whether of hides, bark, or blankets.

------ ........

From the special finds, pottery finds, and lack of any European culture, it is sufficient to say that this site represents a prehistoric village very similar to the Lamar village below Macon. The principal difference is a slight influx of Gulf Coast material and the use of the painted effigy ware as burial offerings.

Figure 60. Field drawing of tobacco pipe from Burial 18 (Lester 1938).

Burial 19 is only a few feet to the east of Burial 18 at a depth of sixty-seven inches below the surface, forty-five inches below the occupation level, fourteen inches in yellow clay loam. It is a female burial, laying on the right side, oriented northwest and southeast with the head to the southeast. This body is contracted more than any found to date, the knees resting against the chest. The skull is rather badly crushed and none of the cervical vertebrae remain, but the rest of the body is in a very good state of preservation {Figure 61}.

Burial 20 is an example of the other type of flexed burial found at Bull Creek {Figure 62}. The body is laying on the left side, oriented east southeast and west northwest, with the vertebrae straight and femora at right angles to the body, feet drawn up almost touching the pelvis. Male skeleton in a fair state of preservation, remaining bones consisting of skull, cervical vertebrae, both femora, portions of both tibiae, clavicles, scapulae, and small sections of the pelvis. A string of seed beads was found in place around the neck but nothing else with the burial {Figure 63}.

Figure 61. Field photograph of Burial 19 (NPS photograph 2427).

68

Figure 62. Field photograph of Burial 20 (NPS photograph 2428).

WPA Excavations
Burial 22 was lying on the right side, oriented NW and SE with the head to the NW {Figure 64}. The skull has been crushed but otherwise the body was in an excellent state of preservation. The pelvis was slightly higher than the head and had been slightly misplaced by root growth but was in an excellent state of preservation. The body was found 54" below surface, 40" in tan sand. Remaining bones consisted of skull, clavicles, scapular, both humeri, ulna, radii, a few of the metacarpal bones, practically all of the vertebrae, tibiae, femora, fibulae, metatarsal bones, and phalanges.
Burial 23 consisted of a few small fragments of scattered bones that had reached such a state of decay that it was impossible to determine anything about them other than that they were human remains, apparently they were part of the long bones of the body but were beyond all definite identification. Found 57" below the surface, 33" in tan sand.
Burial 24 consists of broken fragments of the skull and portions of the humeri, femora, and tibiae. The body is lying on the left side, oriented northwest and southeast with the head to the southeast. It appears that this is a burial of the contracted flexed type. Found 28" in tan sand 59" below surface.

Figure 63. Photograph of marine shell beads found with Burial 20 (Columbus Museum files).
Burial 21 was located 36" below surface, 21" in tan sand. The remains were in a very poor state of preservation, consisting of portions of the skull, tibiae, femora, ulna, and radii. The body was lying on the right side, oriented NNW and SSE with the head to the NNW. The legs were flexed back under the body with the arms straight at the sides.

.;:::.;
Figure 64. Field photograph of Burial 22 (NPS photograph 2433).
69

The Bull Creek Site
Burial 25 is in a very poor state of preservation. The body is lying on the right side oriented NW and SE with the head to the SE. Remaining bones consist of portions of the skull and fragments of the femora, tibiae, and fibulae. From the position of the femora and tibiae it appears that the legs were doubled back under the body with the knees in front, flexed burial of the 2nd Bull Creek type. 52" below surface, 21" in tan sand.
Burial 26 is an extended flexed burial, lying on the left side oriented north and south with the head to the south. Skeletal remains in a very poor state of preservation consisting of fragments of the skull and portions of the tibiae and femora, 79" below surface, 41" in tan sand.
Burial 27 was found 75" below the surface, 41" below the occupation level in tan sand. The remains are in a very poor state of preservation but it appears that the burial is the typical type of Bull Creek flexed burial with the knees drawn up toward the face. Remaining bones consist of fragments of the skull and tibiae and femora. Oriented NNW and SSE with the head to the SSE. No burial furniture.
Burial 28 is located only a foot away from burial #27,80" below the surface, 35" below the occupation level, 5" in reddish brown loam. The body was lying on the back flexed to the left. Oriented ESE and WNW with the head to ESE. Remaining bones consist of portions of the left tibiae and femur and the skull with all of the facial bones missing. Burial furniture associated with the burial included a small amount of red pigment, a piece of mussel shell, a fragment of worked bone presumably an awl, several other pieces of animal bone, and a small stone of the same material as the pigment.
Burial 29 was found 88" below surface, 47" below the occupation level 6" in reddish brown loam. Nothing is left except sections of the tibiae, femora and a short section of the right humerus. Body oriented NW and SE with the head to the SE. There is no trace of the skull at all. Appears to be typical contracted flexed type. Burial furniture consists of a small ball of the red "paint" pigment.
Burial 30 is located 49" below the surface, 11 " below the occupation level. Oriented NNW and SSE with the head to the NNW. Lying on the right side. Remaining bones consist of portions of the skull, tibiae, femora, and a small section of the pelvis. No burial furniture.

Burial 31 was a triple burial the first of this sort found at Bull Creek. The remains were in a very poor state of preservation and it was impossible to determine the exact relationship of the individual bones. It appears however that at least two of the bodies were flexed in the typical Bull Creek manner. The other burial consisted of the skull only. (A) consisted of the skull only, was slightly higher than the other two and all of the skull was gone except a portion of the back occipital and part of the lower jaw. All that remains of (B) was the skull, frontal bones and facial bones missing, and a short section of the right humerus. (C) consisted of portions of the upper and lower jaw, traces of both humeri and radii, and sections of both tibiae, fibulae, and femora. Found 61" below surface, 23" below occupation level, oriented NW and SE, heads NW.
Burial 32 was the contracted type of flexed burial, lying on the right side, oriented Nand S with head to the N. The remains were in a rather poor state of preservation, remaining bones portions of the skull, tibiae, femora, humeri, and ulna. Burial furniture consists of several rocks, flint chips, and fragments of animal bones - impossible to determine whether or not they had been worked, all of these articles were grouped around the neck. 52" below the surface, 34" in tan sand.
Burial 33 was located 36" below the surface, 7" in yellow clay loam, body lying on the right side, oriented Nand S with the head to the S. Nothing remains except portions of the skull and teeth, tibiae, femora, fibulae. From the position of the bones it appears that it was the typical type of contracted flexed burial.
Burial 34 was found to the west of Burials 32 and 33 at a depth of 37" below the surface, 16" in tan sand. The body was lying on the left side, oriented east and west with the head to the west. The remains were in a very poor state of preservation, consisting of skull, which was almost entirely decayed, and sections of both tibiae, which were slightly higher than the skull. This burial was evidently of the contracted, flexed type with the arms between the legs. No burial furniture {Figure 65}.
Burial 35 was located to the west of Burials 27 and 28, 67" below the surface, 27" below occupation level in tan sand. The body was lying on the left side, oriented east and west, with the head to the east.

70

WPA Excavations

/

/

/

/

/
/

/

/

/ / /.

Find 72


/

/

/

./

/
/

B44

. /
/~/o.~



B46

;'1

,~



if


B45

D B42
()

















. t~
B43

. B36
.. ~ ~



,J


B40

L33 L31 L29 L27 L25 L23 L21 L19 LI7 LI5 L13 Figure 65. Plan map of western portion of Bull Creek cemetery.
71

The Bull Creek Site
Burial 35 was of the contracted, flexed type, with the knees drawn up toward the chest and the right arm crooked at the elbow - not going between the legs as is the usual case. The skeletal remains were in a rather poor state of preservation. The skull was very badly decayed, nothing left of the upper body except a short section of the humerus and ulna of the right arm. The lower body consisted of a few fragments of the lumbar vertebrae, practically all of the pelvis, and portions of both tibiae, femora and fibulae. No burial furniture is associated with the burial.
Burial 36 was found to the southwest of Burial 35 at a depth of 60" below the surface, 35" below occupation level in tan sand. The body appeared to have been placed on the back with the head turned to the left, oriented east and west with the head to the east. The knees were evidently against the chest and were found about 6" inches higher than the skull. Only the skull, both tibiae, and traces of the right humerus remained.
Burial 37 was found immediately east of Burial 35, 86" below the surface, 36" below occupation level. The body is oriented east and west, head to the west. Though of the contracted, flexed type, it differs from previous burials in that the body was face downward. It appeared that

the legs were lashed up with the knees touching the chin, with the left arm flexed downward on top of the right thigh, the right arm flexed up with the hand at the face. Remains were in a rather exceptional state of preservation. No burial furniture.
Burial 38 {Figure 66} was found approximately 15ft northwest of Burial 35 at a depth of 36" below the occupation level. This is a double, extended burial. Both bodies were on the right side, oriented northwest-southeast, with the head to the northwest.
A. These remains were in rather poor state of preservation. The skull was crushed and broken parts have been scattered. Remaining fragments consist of portions of skull, humeri, ulna, radii, portions of tibiae, femora and fibula.
B. These remains were in much better state of preservation. Remains consisted of skull; one side of which is rather badly decomposed, sections of ulna, humeri, radii, clavicles, scapulas, all of the vertebrae, with part of the ribs attached, part of the pelvis tibiae, femora, fibula and most of the bones of the feet.

..,-,-.

;::"::;:::::::'"

Photograph of extended burial 38, a multiple burial (NPS photograph 2453).

72

From the comparative size of the bodies it appeared that A was female; B was male. No burial furniture was associated with this interment.
Burial 39 was located to the southwest of Burial 38 at a depth of 42" below the occupation level. The body was lying on the left side oriented southwest-northeast with the head to the southwest. This burial was of the contracted flexed type with the legs drawn up toward the body and the head tilted downward toward the knees. It was interesting to note that instead of the hands going between the legs as is the usual case in this type of Bull Creek burial, the arms were flexed at the elbow and both hands were at the chin. Skeletal remains were in excellent state of preservation, consisting of: skull in excellent condition, all of the vertebrae, humeri, ulna, radii, all the metacarpal bones, portions of the pelvis, tibiae, femora, fibula, metatarsal bones.
Burial 40 consisted of a few fragments of the skull found to the south of Burial 36, at a depth of 61" below surface, 42" below occupation level in tan sand. Careful troweling failed to reveal any other traces of the body.
Burial 41 was found approximately 10ft west of Burial 40 at a depth of 39" below the occupation level. The body was lying on the right side oriented east and west with the head to the west. It was of the contracted flexed type but it was impossible to tell the exact position of the arms as only a small portion of the left humerus remained. Remaining bones consisted of fragments of the skull, small parts of both clavicles, pelvis, tibiae and femora. There was a small piece of animal bone close to the upper part of the humerus, probably accidental placement as part of pit fill.
Burial 42 was found south of Burial 39, 26" below occupation level in tan sand. Burial was of contracted flexed type oriented north and south with the head to the south. The body was lying on the left side. The feet were gone but there must have been a contact with the pelvis with the knees drawn up to the chest. The left hand went between the thighs about 2" from the pelvis and the right arm was across the chest making the ulna and radius vertical. The head was turned to the left thus making it face downward. On the whole the body is in fairly good state of preservation except that the back of the skull is rather badly crushed.
Burial 43 was of the extended flexed type. The burial was located west of Burial 41 and was

WPA Excavations
found 50" below occupation level. Body was lying on the left side oriented east and west with the head to the east. The feet were flexed back almost touching the pelvis and the knees were straight down away from the body. It was impossible to tell the exact position of the arms as there was no trace of them. Only the skull has been rather badly crushed, sections of the clavicles, vertebrae, both femora, tibiae and parts of the pelvis remained.
Burial 44 {Figure 67} was found to the northwest of Burial 42 at a depth of 26" below the occupation level in tan sand, oriented east and west with the head to the west. This was a rather unusual type of burial in that it was placed in the pit face downward. The arms were straight at the side and the legs were slightly flexed to the left. It appears that the body must have been lashed in this position before interment. On the whole this is the most complete burial yet to come from the Bull Creek village. All of the body remaining with the possible exception of a few of the ribs.
Burial 45 was found south of Burial 44 at a depth of 33" below the occupation level. The remains were in such a poor state of preservation that it was impossible to get any exact information about this burial but it appeared that the body was lying on the left side oriented east and west with the head to the east and from the short sections of the tibiae, femora, it seems that it was of the contracted flexed type.
Lester's field notes describe at this time the recovery of a small vessel at a depth that should only be associated with a burial.
Found a small unbroken vessel, Find #72, at a depth of 27" below occupation level; 33" north of Station 4-17L21 on line of stakes. There is no evidence of burial pit in this location and no trace of any bones, but this vessel must have been an article of burial furniture associated with a burial
(Lester 1938:73).
Lester's observation is of critical importance because it demonstrates the difficulty of defining some of the burial pits. Coupled with other field note entries that comment on this difficulty, it is quite probable that some burials were missed entirely. Apparently, only burial pits that intruded the richest occupation midden zones were well defined. However, it seems that most burial pits did intrude into dark Bull Creek middens of earlier occupations.
73

The Bull Creek Site

Figure 67. Field photograph of Burial 44 (NPS photograph 2460).

Burial 46 was found to the west of Burial 45, 28" below occupation level. It was impossible to get any accurate picture of the burial as nothing remained except the crown of a few teeth and short sections of both femora and tibiae. From the position of these it seemed to be of the contracted flexed type lying on the left side oriented north and
south with the head to the south (Lester 1938:44).
Lester's preliminary report abruptly ends at this point without further discussion. Perhaps a portion of the report has been lost, but, just as likely, the report ends with the entry for the last burial {Number 46}. Several field notes concerning the excavation and observations are not presented in the preliminary report. Lester's notes include a running account of excavated squares. While this record may not be absolutely correct due to some omissions and transcription errors from the original field notes, it does allow for an approximate recreation of the boundaries of

the cemetery excavations. Importantly, these notes also provide a few specific location references to important landmarks, such as the western edge of the railroad levee and the upper edge of the river bluff. Lester notes that the western edge of the railroad fill lies 20 feet west of station 4-27L15 (Lester 1938:70). He also notes that the western edge of the excavation was extended to the bluff edge on a line from station 4-13L33 to 4-21L27 (Lester 1938:74). These locations correlate very well with Lester's burial ground grid plan shown on the site base map (see Figure 20). Figure 68 illustrates our approximation of the cemetery limits as interpreted from Lester's field notes. Importantly, this reconstruction seems to support the conclusion that all of Lester's burial orientations are based upon a grid north and not a magnetic north.

74

WPA Excavations

II

II

II

II

II

I
I

II . .'31~?~3e~"6'3~ Bu(.f'a/ Ground

I ..... ' . '3..\\/':;>~~9..A

/. . . I I ..' .:.:~iy ~}.6i"\

if!'li. \ I I ~.;.:,. .

I I ,..<-. . . I.... ,., ~.,., .

...)/' . -~'9-1-";0. \
. "\?.p...~..\.\1.\c\;'~06\.t...\

I
I

I
I

. .....
....

.,.\..')"03\',."/\1\~~.~

\

I

",~

.:../

../>
/"""

"

""
,,

I I I ./ \"''''''',.,''

I " I I _ ...."- ..,/ 10"""

. /..... ./.\.,v. . . / "

>

I
I
I

I
I 1./

/

// /

/

/

{/

", >

I

/

J

I

I

/

II

I

I

II

I
I I

:o
t\:l
~

II I

/,'.,,"./"/ ,.:/I/ /I

/ / /

/

/

(

I~ I i

I ::!'1 I I

I ::::: I \

I

\

I II

I I\

II

Figure 68. Plan drawing showing interpreted boundaries of cemetery excavations superimposed upon Lester's

base map.

Lester's field notes also contain observations concerning the possible existence of structures in the cemetery area. By omitting these observations from the text of his preliminary report, questions concerning the nature and origin of the burials remain unanswered. Based upon the information presented in Lester's report we cannot determine the relationship of the cemetery area to the rest of the site. Among the possibilities are 1) the burials lay within a cemetery segregated from the rest of the village; 2) the burials lay beneath

mound fill which might once have existed but had been leveled by railroad construction; 3) the burials lay beneath and surrounding the floors of domestic structures; or 4) the burials lay beneath the floor of a large ceremonial structure such as a council house.
Lester's field notes indicate that evidence of structures was found during the cemetery excavation but that this evidence was dismissed as too insubstantial. Two examples of Lester's tentative evidence follow.
75

The Bull Creek Site
September 22. The profile at this point shows a band of clay from 2" to 6" thick between station 4-17L9 and 4-23L9, suggesting a house floor. There is also some trace of a sod development below the general occupation level, though it shows no definite form at this particular point (Lester 1938:65).
December 29 " .. began work on cut from 4-11 L13 -L 15 to 4-27L 13 -L15. Working down to occupational level to examine what appears to be a house site.
Lowered section between 4-11 L13 -L15 and 4-27L13 -L15, to the Indian occupation level cleared whole area 4-11 L9 -L15 to 4-27L9-L15. While the area contains a very rich midden accumulation and several buried and charred areas and a few miscellaneous post mould impressions, there is no definite sign of a house site. There is quite a profusion of pottery and a few briquettes (Lester 1938:70).
The area of the cemetery which concerned Lester is shown in Figure 69. The area is intriguing in that it represents somewhat of a void of bodies.

An important consideration is that the cemetery lay within sandy deposits while the two structures excavated in the village area lay within a well-consolidated silty sand. These differences in soil would have meant different preservation states for any structures constructed in the two areas. The soils containing the village area structures would have provided nearly ideal conditions for preservation whereas the sandy soils in the cemetery area would have been greatly affected by leaching and various forms of bioturbation. Also, the village structures had been burned, resulting in brick hard surfaces which would have been less affected by various forms of erosion.
Lester's work left a number of unanswered questions concerning the nature of the cemetery. Determining the answers to these questions eventually led Dr. A.R. Kelly to return to the site to conduct further excavations. These and other investigations that were conducted over the next four decades will be the subject of following chapters.

4-27 4-28

Clay Floor in Profile ~(:-.

4-25 4-26

t(

~.

.",

4-21 4-22

I'

v
4-11 4-12

AGN

~2m

o

s Feet

/

/

w3 wI L29 L27

L9 L7

L5 ts

LI 4-1 4-2 RI

R3

Figure 69. Reconstructed plan of cemetery showing area thought by Lester to contain structural evidence.

76

Chapter 4 Kelly's 1950 Investigations

A.R. Kelly, by 1950 head of the Department of Archaeology and Anthropology of the University of Georgia, returned to the Bull Creek site to conduct further excavation in the Spring of 1950. The intent of this work was to expand upon the work of Lester and Patterson in 1936 and 1937. As previously noted, Lester's preliminary report was simply an abbreviated rewrite of his field notes. The manuscript chronicled the field project but there was little interpretation and virtually no artifact examination. The inability to produce a detailed site report clearly frustrated both Kelly and Patterson during the following years. Unfortunately, Bull Creek was like most WPA excavation projects in that adequate funding was available for excavation but not for write-up. The closest Kelly came to producing a meaningful presentation was a paper prepared immediately following Lester's field work which examined Bull Creek within the context of other important Lamar sites.
In 1937, Kelly prepared a paper for the annual meeting of the Society of American Archaeology in New Haven, Connecticut entitled Lamar and Related Sites Excavations (Kelly 1937). In that paper, the Bull Creek site was accorded its highest level of archeological significance, being included with such sites as Lamar, Mossy Oak, and Irene. This brief moment of fame departed quickly and the Bull Creek site slipped into obscurity.
Kelly's New Haven paper was never published and subsequent publications by Kelly included only brief references to the site. For example, in his Bureau of American Ethnology report on Macon Plateau, Kelly simply noted the contributions of Patterson and Lester and stated the importance of Bull Creek "lies in the fmding of a specialized funerary ware of painted effigy dog pots found with burials in the village midden. The domestic ware from the midden

represented typical examples found in the Lamar pottery complex. The same conjunction of Lamar-like pottery traits with painted dog pots had previously been noted for the Neisler site, on the Flint" (Kelly 1938:51). On the same page, Kelly mentioned that a separate report for Bull Creek was to be prepared. Thereafter, Bull Creek was relegated to one example of the Lamar phenomena.
The New Haven paper presents a rare glimpse into Kelly's basic concepts of interpretation of the Bull Creek site. Selections from correspondence as well as pertinent portions of the paper are presented here to explore these concepts. The basic ideas for the New Haven paper were discussed in several letters between Kelly and Patterson. Besides detailing Kelly's views, these letters provide insight into the academic isolation of the south at the time and the fundamental fmancial and political realities of Depression-era archeology.
In a letter dated October 22, 1937, Kelly conveyed the major concepts of the paper to Mrs. Patterson.
You will appreciate the importance of Irene in connection with previous research carried out at Lamar Mounds in Macon and at Bull Creek, at Columbus. These three sites together give an idea of the cultural characteristics of tribes who inhabited Georgia at about the same time over wide areas, as indicated by the Chattahoochee, middle Georgia, and coastal sites mentioned. These sites show types of pottery, mound characters, burial traits, game pieces, ornaments, pipes, and other cultural features implying a mixture of cultural habits between two groups. One of these, represented by the Lamar complicated stamped pottery indices, native to the southeast; the other indicated by the incised ware, almost certainly were immigrants from the lower Mississippi east, crossing many streams, and finally settling on the Georgian rivers.

77

The Bull Creek Site
The Kasita and Kawita were in the vanguard of these early tribal movements. Bull Creek, Lamar (Ocmulgee Old Town), and Irene represent stations on which settlements and remains of these first contacts of the ancestral Creeks with "strange Indians" already inhabiting the area were made.
Time indicated for these settlements would be pre-De Soto according to my conceptions. And I believe that Dr. Swanton would concur in this judgment as he feels that the restless, unsettled condition of the country at the time of De Soto was an aftermath of tribal dislocation incident to these original migrations of the Creeks.
The above summary gives about the gist of the remarks which I would like to make at New Haven. I hope that this will serve to cover the information you advised me you would like to have.
Please let me have your reactions (Ledbetter 1995b:26).
The New Haven paper is mentioned in a letter from Kelly to Patterson dated November 4, 1937.
I am sending on a copy of manuscript "Lamar and Related Site Exploration", to be offered for the meeting at New Haven. It strikes me that this topic would be of more general interest to Georgians in general as well as to my scientific colleagues. It touches on crucial exploration in three areas of Georgia, the Chattahoochee, the Ocmulgee, and Savannah. It brings into comparison other site exploration familiar to students in the Southeast who know
existing archaeological literature (Ledbetter 1995b:28).
Mrs. Patterson replied to Kelly in a letter dated November 16, 1937.
Thank you a million for your letter of November 4th containing a copy of your manuscript "Lamar and Related Site Exploration," to be offered for the meeting at New Haven .... It would be difficult to describe my appreciation of your going and telling about Georgia. I am very grateful to you for this information for I had only fragmentary inklings of much that you now make plain ....
I am sure the trip would do you a lot of good and do much for Georgia. I am glad that you had such a nice talk with Mrs. Woodward and I am delighted over the new light in regard to emphasis between state universities, Park Service, and the
78

Smithsonian. I hope Georgia will not be a laggard in
taking advantage of this service (Ledbetter 1995b:30).
In a letter from Kelly to Mrs. Patterson dated December 18, 1937, the realities of WPA era archeology are evident. Kelly's letter stresses the benefits of involving the larger academic community in the work in Georgia but the main topic of the letter is simply pulling together enough money to make the trip.
I have less than a week in which to make preparation for the New Haven jaunt. That is my fault alone because I have not pressed this matter on the one or two occasions when we had occasion to discuss it. The situation has developed in an unanticipated manner and I think I should be frank in dealing with the embarrassing features of the case.
My impression over the telephone when the New Haven matter was presented by you was that the Colonial Dames and others felt strongly that Georgia should be represented at the meetings and that papers dealing with Georgia archaeology, which has attracted such widespread attention, should be given. That there was a feeling that I should have been present at the last meeting and must certainly attend this one . . ..
Now that my name and that of G.Willey are down for papers on the morning of Dec. 30th at New Haven, I must go. And I will manage in some fashion. But still I am going to be stubborn about the finances unless it turns out that you have got the support you expected.
I have no objection to letting your organization help me to make the trip because I feel that the subject of my paper has great significance for work in which they have been particularly interested for some time; i.e., the Lamar sites description pretty well outlines the archaeological picture of what Georgia was like at or just before the time of De Soto's journey ....
I don't know what I'll do about clothes--and in writing that I'm not talking in the same sense as the debutante worrying about a coming out party. Gosh, it may be alright to go without a hat in the amenable Georgia Climate; that can be set down as eccentricity, maybe I'm just collegiate---but a 1200 mile trip to the less clement North is another matter. However, that's a purely personal element so we'll
skip it (Ledbetter 1995b:35).

A letter to Pattersondated January 12, 1938 gave Kelly's impressions of the meetings.
The New Haven Conference was from most points of view successful and propitious .... As to the reading of the papers, I should advise you that I found the fifteen minutes allotted as expected altogether a handicap to permit reading the entire paper. It was necessary to talk more or less extemporaneously .... I am writing H.C. Collins of the National Museum to send my papers so that I can add notes bibliographically with an eye to publication in American Antiquity. This will give fuller publicity to the interesting connections between Lamar and other related field work in the Georgia area, notably the
work at Bull Creek and at Savannah (Ledbetter 1995b:37).
The actual details allotted to the Bull Creek site in Kelly's paper are minimal. Because analysis had not been conducted on the excavated material, Kelly focused upon the high points of the site and specifically the cemetery. Kelly was obviously faced with the problem of being required to include some information on all of the sites sponsored by the individuals and communities of the region. Excerpts of Kelly's paper which deal specifically with the Bull Creek site follow.
The Bull Creek site on the Chattahoochee, within the outskirts of Columbus, Georgia, was explored under the field direction of Frank Lester, engineering student trained in archaeological field methods at Macon. Recognition of the importance of the Bull Creek site grew out of historical documentary studies made by Mrs. H. Wayne Patterson of Columbus, interested in checking the location of historic Creek settlements on the Chattahoochee, following leads provided in publications of Dr. J.R. Swanton. The investiqations employed National Youth Administration workers during a period extending from March, 1936 to April, 1937.
The work at Bull Creek uncovered remains of a village site exposed in the river bank and in an old road cut; also, a burial ground in conjunction with the village revealed features of special interest. Concisely, Bull Creek shows a typical Lamar-like site, minus pyramidal mounds, with Lamar complicated stamped and broad-lined incised ware contrasting with a very ornate painted effigy ware found with burials in the cemetery belonging to the village. No less than three dog effigy vessels were found with

Kelly's 1950 Investigations
burials. In the same midden in which the burials occurred and in association with other burials in the same cemetery were catalogued numerous sherds and some whole pots characteristic of typical Lamar complex. There was no evidence of later intrusion of the burials with the specialized effigy funerary ware nor could the profile features be interpreted to mean that the burials with effigy pottery were preexisting.
At the Neisler mound and village site on the Flint river, near Roberta, Georgia, thirty miles south of Macon, river wash has exposed burials associated with painted dog effigy pottery vessels identical with those found at Bull Creek. So far as the study of surface collections from the Neisler site can indicate, it is evident that the same combination of characteristic complicated stamped pottery and
incised ware occur here as at Bull Creek (Kelly 1937:5-7).
Kelly's paper continues to summarize other sites and examine Lamar traits. Burial traits were singled out for Bull Creek. Kelly noted that while many Lamar traits showed great consistency across a broad geographic area, burial customs observed at several key sites showed wide divergence.
The general condition is to find primary flexed, semi-flexed or extended burials in large graves or pits; group burials are frequent; burial furniture tends to be abundant, consisting of pottery, polished stone artifacts, and pipes; burials occur in house floors, outside the walls of the houses, on the ramps of house mounds, or in the midden layers.
At Bull Creek on the Chattahoochee and at the Neisler mound on the Flint, we have noted the presence of a specialized burial ware of painted dog effigy pots found with flexed primary burials. Ordinary domestic ware of the Lamar type occurs with other burials on the same sites .... Other Lamar-like sites, as judged from surface features and pottery in
Alabama, show urn burial also (Kelly 1937:12-13).
Kelly's work with the National Park Service took him away from Georgia for several years but he continued to conununicate with Mrs. Patterson about Bull Creek. Upon his return to Macon, Mrs. Patterson attempted to jog his memory. This is evident in a letter from Kelly to Patterson dated May 17, 1945.
79

The Bull Creek Site
You want to know about Bull Creek - I don't know and will have to check. Better write me in detail on what you have in mind. Remember, I've been away a long time and its a slow torturous business trying to pick up all the loose ends. And there are plenty of them - things have been going to rock and ruin these years - deterioration is accelerating now and I'm hard pressed to get funds
out of NPS, or to get adequate help (Ledbetter 1995b:43).
Archeological interest in Bull Creek revitalized with threatened commercial development. TIlls threat is mentioned in a letter from Kelly to Patterson dated November 8, 1949.
Dear Isabel, I haven't heard from you since the flurry last summer in the matter of Bull Creek site and successful efforts to stay off threats of construction there. I have since had further correspondence with the Secretary of the Chamber of Commerce who seems to be a good friend of your museum project and in helping to do something with
archaeology around Columbus Falls (Ledbetter 1995b:44).
In a letter dated November 29, 1949 Kelly discussed with Mrs. Patterson his plan to write a Bull Creek report.
Yes, I'll write it up for publication if we can assemble sufficient notes and data. And we'll plan to get it out in our new series on Anthropology and Archaeology at the University Press, if I can rake up about $300 to get it out. It will probably cost about that much on the basis of what it takes to get out Bill Sears' Kolomoki Season I, now in press. Citizens of
Blakely gave us the money for that (Ledbetter 1995b:45).
Ironically, Kelly's plans for writing up Bull Creek were postponed by the opportunity for further excavation on the site. Apparently, the project was supported in large measure by the Columbus Chamber of Commerce. Virtually nothing is known about the excavations except the material included in a Chamber of Commerce newsletter article and a report written, at least in part, by Kelly.
The field project was obviously meant to provide data for additional large scale excavation
80

of the site. The city supplied the labor and a photographer. The photographer was responsible for some of the best, albeit excessively posed, documentation of any of the early excavations at Bull Creek. The photographs have been preserved in an annotated album compiled by Mrs. Patterson (Columbus Museum files).
The city also prepared a detailed map of the Bull Creek site (Figure 70). A second version of the map, prepared by Frank Schnell, Jr., is shown in Figure 71. Figure 72 shows site area as it appeared in a 1950 aerial photograph.
The text of the newsletter is presented as an introduction to Kelly's report. The text of Kelly's report that deals specifically with the field investigations follows. The introductory culture history portion of Kelly's report has been omitted.
The text of the chamber of commerce
article titled Local Archaeological Possibilities Studied May 1950 follows.
Definite steps toward developing archaeological possibilities in Columbus were taken this week, when the Board of Directors requested President J.W. Woodruff, Jr., to appoint a Committee to study recommendations made to the Board by Dr. A.R. Kelly, Head of the Department of Archaeology and Anthropology at the University of Georgia.
Dr. Kelly met with the Board of Directors on Monday, May 22, at the request of Chamber officials. Mr. Mahan, also of the University, and Mrs. Wayne Patterson, local archaeological enthusiast, were also guests of the Board.
Chamber officials had requested Dr. Kelly to visit Columbus several months ago to investigate local possibilities and to submit his recommendations regarding the archaeological possibilities existing in and around Columbus. Dr. Kelly's report and recommendations were based on his subsequent visits and studies. Dr. Kelly pointed out that the Chattahoochee River area, near Columbus falls, has long been recognized in the annals of American archaeology and ethno-history as one of the most populous areas in aboriginal times. He stated that more evidences of this occupation were exhibited in the form of sites.

Dr. Kelly went on to state that "Dr. John R. Swanton of the Smithsonian Institute probably showed more Indian village site locations on the Chattahoochee, within a 30-mile radius of the Columbus falls, than at any other point on the route to the Gulf." Dr. Swanton's showing of the historic and proto-historic tribes was based on studies of 17th and 18th century maps of the region, bolstered by researches on the ground. According to Dr. Kelly's report, archaeological surveys undertaken thus far have tended to substantiate the indications of the ethno-history.
Dr. Kelly pointed out that the Bull Creek site is suffering from erosion and that commercial development in that area, which will follow the

Kelly's 1950 Investigations
completion of the Chattahoochee-Apalachicola-Flint project, will destroy valuable relics existing in that area.
Dr. Kelly recommended to the Board that the area be excavated immediately, and that relics salvaged be removed from the site and stored in a place for safe-keeping until such time as a local museum is available to house the materials. He recommended against preserving the relics "on the site," but that they be removed before they are destroyed by commercial development and erosion. President Woodruff announced that he would
appoint a Committee in the near future (Progress, Volume III, No. 25, May 1950:3).

.. --r ~,)":i~;,))
~ 1936 FINDINGS
c:::2Zl 1950 FINDINGS

LAY~UT
SHOWj~~G
CITY OWNED PROPERTY ALONG CHATTAHOOCHEE RIVER AND BULL CREEK WHERE FINDINGS OF INDIAN INHABITATION WERE MADE
DRAWNBY ~e ;;If-/
OFFICE OFCITY ENGINEER CITY OF COLUMBUS, GA.
1950
SCALE 1;100' APFlOXIMATELY

Figure 70. Redrafted copy of a 1950 City Engineers map of the Bull Creek site. 81

The Bull Creek Site

<>;
.... :.'J":::.

. .'.'

\

.....

N
!

9MEI
Schematic Map of
Excavation Units Kelly: 1950

o

30m

o

-100ft

. ," , . : ...:: .".: . '.

Figure 71. Redrafted copy of the map of Kelly's excavations (adapted from Schnell 1963)

82

Kelly's 1950 Investigations
Figure 72. Enlarged portion>of1950aenalphotograph of the Bull Creek site area. USDA, Soil Conservation Service photograph taken approximately two months prior to Kelly's field work.
83

The Bull Creek Site

Kelly's Report

importance of the site for commercial development poses a threat to the preservation of any

The first three pages of Kelly's 1950

prehistoric cultural remains still remaining in the ground. The area between the river and the creek

report, entitled Report on the Archaeological

has been cut back by both streams and is

Possibilities of the Bull Creek Site, Columbus, Georgia, with Recommendations for Development presents a background to the prehistory and early history of the region and an introductory paragraph for Kelly's work in

undergoing rapid erosion. It has also been severely hacked by the construction of the industrial railway embankment constructed in 1926, so that, practically, one needs to know just how much of the deepest and richest deposits, which yielded 46 Indian burials, several house sites, and much in the

early May 1950. The background pages have been omitted. The quoted portion of the report begins with the introductory paragraph on page 3 and continues through to the end of

way of relics possessing high artistic qualities, still remains to be salvaged. A review of findings in five units of stratigraphic trenching and test pitting of the area follows.

the narrative account of the report (page 7). Figure 72 is an enlarged portion of an aerial photograph taken of the Bull Creek area in

Results of Stratigraphic Test Pitting and Trenching of the Bull Creek Site

March 1950, two months prior to Kelly's excavations. Several of the landmarks referenced by Kelly, such as the fisherman's

Unit #1 {see Figure 71}. Beginning on the summit or parapet of the embankment fill directly overlooking the confluence of Bull Creek and the

path, are visible in the photograph.

Chattahoochee river, a 5 foot trench was put down

through the fill and extended down slope toward

This brings us to the present survey reconnaissance by A.R. Kelly of the Department of Anthropology and Archaeology at the University. Investigation was undertaken with a view to determining what changes had taken place in the condition of the Bull Creek site in the last 12 years,

Bull Creek for a distance of twenty five feet or more {Figure 73}. The purpose was to determine depth of fill at this point, and to ascertain whether slumping and erosion sinks in the embankment summit could possibly indicate subterranean pits or Indian burial areas.

how much of the site

remained in relatively

undisturbed condition,

and how profitable

further exploration on

the site might be

calculated to be.

Most of the Bull Creek

site proper, located

between Bull Creek

and the Chat-

tahoochee, at the

junction, is on pro-

perty of the City of

Columbus. (see map,

Figure 70). This pro-

perty fronts on the

river and has strategic

importance in future

industrial develop-

ment, perhaps as a

site for a municipal

dock when Columbus

eventually has a nine-

foot channel to the Gulf. The critical

Figure 73. Field photograph of Unit 1 showing depth of railroad nil. Dr. Kelly is
shown standing at upper end of trench (Columbus Museum files),

84

Kelly's 1950 Investigations

The excavation revealed clearly that the

might not have been found or excavated, and the

embankment at this point had been filled to a

indicated job was to complete the exploration of

depth of twenty feet or more and that no aboriginal

this mound. If house structures, floors, wattle and

occupation level could intervene to account for the

daub walls, refuse pits, and other features were

observed slumping of fill dirt. Further exploration

abundant, with no lenses of basket-filled dirt in

at this point is therefore not recommended to

place indicating a mound, then the 1936-1937

uncover archaeological features in the original

burial discoveries would probably relate to burials

occupation level.

made in house-floors. In this case, any large thick

black occupation level still remaining, with

Unit #2. {See Figure 71 map for reference

indications of house structures, would suggest that

to unit location}. This unit was a profile clearing of

further burials and pottery and other objects

a portion of the eroded terrace overlooking Bull

associated would continue to be found in further

Creek at a point where considerable broken pot-

exploration. Excavation #1 of unit #3 revealed

tery, animal bones, and scrap flint indicated mid-

scattered pottery under about 4 feet of recent

den spill from an occupation level, exposed by the

water-laid sands and gravels deposited in flood

erosion of the area in recent floods. No house

stage by Bull Creek or the back swirl of the Chat-

sites, burials, or other archaeological features were

tahoochee. No house floors, pits, burials or spe-

uncovered, but several bags of pottery and other

cific structures were uncovered in place. Ex-

remains were catalogued, showing that the original

cavation #2, made farther upslope from Excavation

village level at this point had a midden thickness of

#1 , revealed the buried occupation level extending

about 12 inches. The profile cleared is forty feet

upward, with some pottery in place but no features

or more from the embankment fill.

uncovered, and no indications of mound structure.

Excavation #3 was put down on the north side,

Unit #3 {See Figure 70 for approximate

facing Bull Creek, of the backfill to the 1936

locations}. This unit consists of three separate

excavations. Here a deep, crescent-like, eroded

excavations made in the marginal slopes around

bank of Bull Creek, was cut back and profiled

the site of the 1936-1937 excavations made by

along two converging shelves {Figure 74}.

Engineer Frank Lester and Mrs. Wayne Patterson.

That work {1936} uncovered the remains of 46

prehistoric burials,

accompanied with rich

grave furniture,

including no less than

3 effigy dog pots of

unusual artistic merit.

The problem here

was: had the 1936

exploration unwittingly

cut back from the

original exposed Indian

burials in the bank of

Bull Creek into what

was really an Indian

Burial mound, partially

mantled and con-

cealed by the 1926

railway embankment

fill, or were these

burials nested in deep

village midden, the in-

terments having been

made in the floors or

immediately outside

the walls of houses?

If a burial mound was

indicated, then other burials and structures

Figure 74. Photograph of Excavation Area 3 of Unit 3, captioned: Dr. A.R. Kelly, Mrs. Wayne Patterson and Doc discussing plans for work (Columbus Museum files).

85

The Bull Creek Site

In both pro-

files, rich black dirt,

full of pottery, animal

bones, flint indicated a

thick occupation level

14 inches or more in

thickness, under about

3 feet of recent soil

mantle, most of which

was detritus from the

railway embankment

some thirty feet away.

What was significant

here was that in one

place we found baked

clay daub, showing

the impressions of

wattles, extending for

3 feet in the profile.

This almost certainly

indicates a house

structure with wattle

"

and daub walls, whose floor section will be found to extend back into the unexcavated ground for

Figure 75. Field photograph with notation: Dr. A.R. Xiiiy holding;~n;;" vessel to
show better its size and shape. The dirt inside has not been removed (Columbus
Museum files).

twenty feet or more.

Close by in Excavation #3, Unit #3, was a

aeological materials. Also, that house sites, and

complete pottery vessel, of the plain ware asso-

very likely additional burials, will be encountered in

ciated with Bull Creek people, exposed about half-

further exploration at this point.

way down in the block

soil occupation zone

{Figures 75 and 76}.

No definite human

burials were un-

covered in this initial

cleaning of the face of

the cut. But the con-

ditions imply that

burials originally found

in 1936 in this area

did occur in house

floors or in the village

midden around house

walls. It is also evi-

dent that 14 inches or

more of thick rich mid-

den still remain in the

area between Bull

Creek and the bluff

overlooking the Chat-

tahoochee, at varying depths beneath the clay embankment, and that this occupation is very rich in arch-

o
t-

em

f,'

Figure 76. Photograph of vessel recovered by Kelly in 1950 (Columbus Museum files).

86

Kelly's 1950 Investigations

Unit #4 {Figure 71}. On the eroded terrace overlooking Buff Creek, at a point down slope from the bare road area frequently used for car parking, and just behind the site of a former shanty (one outbuilding in ruins remaining), this profile cut was made and cleared horizontally for a space of ten feet or more, to check black topsoil and subsoil in which considerable potsherds and other midden was showing on surface, and in the eroded bank. Ten or twelve bags of pottery all broken, and other artifacts were recovered here. The situation suggests from the amount of occupational debris, that there is a prehistoric Indian house structure to be found within a few feet of this point.

can be no doubt but that it extends back underneath the embankment to join up with the rich black Indian village midden revealed in the excavation to Unit 4.
In addition to the above, profiles were cut to determine if all of a house structure, uncovered by Mr. Lester 1936, had been removed. This portion of the site is on a path entering the embankment path, alongside the scar of the old borrow pit from which soil was removed in 1936 to make the embankment fill. Apparently Mr. Lester did remove enough of the overburden here to uncover as much of his structures as then remained.

Exploration was not continued to bring this out as we did not wish to expose valuable structures that might excite visitors to the area to

Recommendations for Treatment of Site and Development

dig into the site during the absence of the archaeologists. Enough was found to indicate the probable location of a house structure at this point. Also, that the original village occupation zone was still strongly indicated here, supporting conclusions drawn from Excavations #2 and #3.

Analysis of the results of trenching and test pitting, from the beginning of the old railway embankment and the borrow pit to the junction of Bull Creek with the river, shows that there is a long narrow, fingering section of rich, deep midden, covered with varying depths of recent fill, 6-8

Unit #5 {see Figure 71}. Almost directly opposite the railway embankment from Unit #4, on the exposed face of the high bluff overlooking the Chattahoochee, and where a path had been worn down to the river by fishermen, we found strong outcrops of black midden in the path and in

feet nearer to the borrow pit, 12-20 feet closer to the junction. Down slope, toward Bull Creek side, varying patches of this village midden are relatively unmantled by detritus from the embankment fill, and can be uncovered under 3-4 feet of overburden.

slumped dirt. Cutting

a new, clean profile

here revealed again

14-18 inches of black

midden dirt, original

village midden, pre-

served under 3 Y2 to 4

feet of the em-

bankment fill. Pottery,

flint, animal bones

were abundant. Some

charred clay running

about half way

through the midden

suggests burning was

extensive at one stage

of the occupation;

possibly some of

these represent dis-

integrated "bri-

quettes," clay-daubs

from Indian houses.

The occupation here

has an imposing depth

and is very rich {Figure 77}. There

Figure 77. Field photograph captioned: Dr. A.R. Kelly indicating the width of the black midden deposit; charred clay can be seen (Columbus Museum files).

87

The Bull Creek Site
Unit #3 excavations suggest a cluster of house sites, in which complex of structures, apparently in 1936, some 46 burials were exposed. A considerable remaining portion of this village area is still preserved, insulated underneath the heavy railway fill. Definite signs of Indian house floors were uncovered in Unit #3, and elsewhere the occupation is found to have a depth of from 12 inches to nearly 18 inches in thickness.
The narrow, fingering strip of land, precariously wedged between Bull Creek confluence with the Chattahoochee river, has been rapidly cut away by both streams; at least ten feet on the Chattahoochee side, and probably more on Bull Creek side has been eroded during the last twelve years since the site was partially explored in 19361937. Action of the streams is to undercut the banks, and subsequently extensive slumping takes place, huge blocks of embankment and underlying village midden sliding down slope. Unless some salvage procedure is undertaken, in a few years the whole point of land between the two streams will have been cut down to form a rough eroded mass with all original cultural context lost.
The immediate, specific recommendation therefore is that the portion of Bull Creek site indicated, between the creek and the river on city property, be excavated as soon as possible to recover archaeological materials and information that will be irretrievably destroyed by natural forces or by industrial development of the site. This section alone would probably occupy a trained archaeological supervisor and 5 or 6 workmen a whole summer to complete. This, even though not all of the strip indicated might not be completely excavated.
I also recommend that the collections and notes be preserved at the archaeological laboratory of the University of Georgia, to be studied, with the preparation of a report on the archaeological findings. This project of study and publication can be considered apart from the immediate recommendation for the salvage operations to recover the material that will be lost. The material and notes will, of course, be the property of the sponsor, the University offering facilities for storage, preservation, and laboratory analysis.
No treatment of the site with a view to preservation of exhibits in place of a museum is recommended. This hardly seems practicable because of the severe erosion of the area by both Bull Creek and the river. There would be not only the expense of protective covers, museum installations, but a heavy expenditure to erect pilings and insulation from flood damage. The area
88

is too constricted and fingering for effective maneuvering in any permanent planning. The only conclusion possible is to plan on exploration, as complete as possible with entire removal of all specimens to a place of safety, later to be exhibited in a museum building elsewhere.
A specific recommendation is made that the exploration be undertaken for the purpose not only to salvage the relics and data, but to preserve these with a view to their ultimate utilization in appropriate exhibits exemplifying the early aboriginal history of the Columbus area. In short, that the project be undertaken both to save the archaeological values, and in planned contemplation of a museum project that would develop. It is my sincere judgement that the publicity attendant upon the city's undertaking this work would help very materially in interesting your citizens in a museum as a development. There are many other sites, and many other existing collections, valuable materials that loyal people will doubtless make available to you as soon as a definite plan for a museum is matured.
That portion of the Bull Creek site between Bull Creek and Victory Drive, and Weracoba Creek, is also village site; how rich and complicated by house sites and archaeological features remains to be determined. This part of the area is not in as immediate danger from total destruction and might be held in abeyance, pending the results of exploration on the critical area herein described. Possibly, the archaeologist employed in exploring the critical area, could find time during his tenure to make adequate test pits in the adjoining portions. I recommend that this be done if time and nature of the summer's work will permit.
I am appending a suggested budget that would cover 10 weeks of exploration on the Bull Creek Site. This includes provision for an archaeological supervisor, the employment of 5 workers, supplies, photography, engineering recordation. The cost of preparing the materials, studying them, getting out a suitable publication, can only be estimated when the exploration is completed. For that reason I recommend that consideration of this
item be postponed (Kelly 1950:3-7).
Kelly's budget provided for 2000 man hours over a 10 week period. Pay for a supervising archaeologist was to be $75 per week and sixty-five cents per hour for five archaeological laborers. With supplies, the entire project would have cost only $2400. The project was not implemented.

Obviously, the primary purpose of Kelly's 1950 address to the Chamber of Commerce and the subsequent brief report was to convince the city of the need for additional archeology before the site washed into the river or was leveled for construction. The project produced a substantial number of artifacts. The artifacts, presently curated by the University of Georgia, in Athens, have been examined by several researchers. Artifacts have been accessioned by excavation unit but aside from a few handwritten field notes on small fragments of the paper bags, no record of the excavations exists today. It is fortunate that a copy of Kelly's report to the Chamber of Commerce and Patterson's scrapbook have been preserved at the Columbus Museum.
This record of Kelly's brief investigations in 1950 provides valuable information concerning the layout of the village area but is most critically important to interpretation of the "Bull Creek Cemetery." Kelly was able to confidently demonstrate that the cemetery area contained ample evidence of structures. Kelly proposed that the burials excavated by Lester lay beneath the floors and surrounding the walls of a complex of house structures (Kelly 1950:5-6). The possible presence of structures in the cemetery had been considered but dismissed by Lester (Lester 1938:70).
Kelly implies, with specific reference to associated "village middens," that these cemetery area structures were domestic houses. Typically Lamar houses contain burials beneath the floor or just outside the structure walls. Kelly does not address the possibility that some form of ceremonial structure existed within this relatively small portion of the site which appears to contain the vast majority, if not all, of the interred burials.
The supposition that a mound might have once existed above the cemetery appears to have been discarded without further comment as the concept of habitation structures was embraced. Apparently, Kelly found no evidence of basket-loading but conceivably, the

Kelly's 1950 Investigations
scale of Kelly's excavations was not adequate to sufficiently explore the possibility.
Kelly stated that the village midden extended along Weracoba Creek all the way to Victory drive and at the time potentially contained well-preserved deposits. This area would encompass the western half of the Weracoba Creek site which had been judged distinct by the Fenengas during their 1944 survey. Kelly's observation provides important evidence for reconstructing the extent of the Bull Creek village. By incorporating the Weracoba Creek local (presently Georgia State Site Files Number 9Me379) into the Bull Creek site, the village area would extend from the river for a straight-line distance of approximately 500 m along that tributary. While impressive, this boundary estimate does not reach the extent required for a village area covering threequarters of a square mile proposed by both Lester and Patterson.
Documentation accounting for the lack of implementation of Kelly's proposal for additional work at Bull Creek has not been found. The most reasonable likelihood was simply that the cost was too great. However, Mrs. Patterson still continued to press Kelly to prepare a report of Lester's excavations. Perhaps Kelly's final endeavor to undertake the manuscript is referenced in a March 7, 1951 letter to Mrs. Patterson. In that letter Kelly discusses the realization that Bull Creek with its Fort Walton influence actually differs substantially from basic Lamar.
I know you have always been concerned to get Bull Creek into print. I have talked this over with Bill Sears. We are prepared to write up Bull Creek, or to start on it at least, this Spring; working with the materials I brought back from last year's test excavations, and then going into the old notes and materials if we can get access to the stuff at Ocmulgee. I think we can produce a valid report, and study of the Bull Creek stuff I brought back in the laboratory now is indicating a much wider spread of archaeological materials, with regard to source, than we had expected.

89

The Bull Creek Site

For one thing, while the rim treatment, paste, and general morphology of the bulk of the stuff at Bull Creek is definitely in the time level and "feel" of Lamar, the specific designs have a Savannah stamped tradition; also there is a lot of pottery that must refer to influences coming up the Chattahoochee from Florida, Fort Walton influences. This means a very interesting variant in Bull Creek of the widespread Lamar manifestation.
I probably won't have the money to publish the report right away but at least we intend to get the manuscript ready for publication if at all possible in the near future. It has occurred to me that the Columbus people, if they knew an important manuscript had been prepared, with beautiful illustrations and drawings, ready in the press for publication, that some help financially might be forthcoming. I think as much as $500 would insure action by the Press. The expense of preparing the publication can be borne by our Department here, utilizing existing personnel and laboratory facilities. At any rate, I knew you would be interested, and I wanted to keep you informed
(Ledbetter 1995b:49).

Donalsonville, from a soldier at Fort Benning, who is a trained archaeologist, former Carnegie Museum man, and who apparently is permanently stationed at Benning. I have written him to contact you, as I know you would want to meet him, and he might be the boy to help you with any outstanding problems in connection with your museum project. I know you are probably exasperated with me for my tardiness, but I would beg your kindness in remembrance of my good wishes and efforts to
help the best I can (Ledbetter 1995b:53).
Isabel Patterson died in 1955 but the new generation of archeologists continued to conduct excavations at the Bull Creek site. Sgt. David W. Chase conducted limited excavations as part of his ongoing investigations of the Columbus/Fort Benning region and University of Georgia student Frank Schnell, Jr., with the assistance of personnel from the Columbus Museum conducted more extensive excavations. Details of this work will be presented in the following chapters.

Kelly failed to write that Bull Creek report the following spring. Kelly's role in the Bull Creek saga waned following the 1950 excavations. Mrs. Patterson continued to work with the Bull Creek material with efforts directed towards exhibits containing her beloved dog pots for the new Columbus Museum as well as the Smithsonian Institution (Figure 78). Kelly continued to correspond with Patterson and in a letter dated August 6, 1953 he offered support by guiding a new generation of archeological attendants.

In the meantime, I am enclosing a letter received during July while I was in

Figure 78. Photograp .dated 1957 shOWing a Bull Creek dog pot on display at the
Smithsonian Institution (Columbus Museum files).

90

Chapter 5 Schnell's 1959 Excavations

Nearly a decade following Kelly's brief investigation at Bull Creek, Frank Schnell, II. organized and successfully completed a systematic excavation project at Bull Creek. A manuscript which details those excavations, The Bull Creek Site in Western Georgia (Schnell 1963), includes a substantial body of background material and ceramic analysis that has been in-

obtained from a vocational rehabilitation program operating in Columbus for juvenile delinquents, under the direction of the Columbus Craft Shop, Inc. A total of two to three days out of each week was spent excavating on the Bull Creek Site . . . . A total of 950 square feet was excavated in two areas, XU-I and XU-II, in stratigraphic sequence, along with profiling several banks and relocating
excavation units (Schnell 1963).

corporated elsewhere in this report. The following paragraphs relate to the implementation and results of the 1959 field investigation. Pertinent information has also been extracted from Schnell's field notes curated at the Columbus Museum.
Schnell excavated intermittently at the Bull Creek Site during the summer of 1959. His goals were to:

Figure 79 shows a 1957 property map that encompasses the Bull Creek site. The city property markers on the plat were used as fixed reference points for Schnell's grid. The locations of the 1959 excavation units is shown in Figure 80. A series of excavation photographs are shown in Figure 81. These photographs relate primarily to the major excavation unit located near Kelly's old excavation unit 4.

. . . obtain a larger stratigraphic sample ..

. determine if any structures remained in the penin-

sular portion of the site, and to determine if the

site warranted further extensive excavation before destruction by in-

_______.

"-__ _ L .(h I~INA~O",C o,"'o! ---l:~L ~

dustrial development. The pro-

'1"cto!Sv

Ttl tT $),....,...

ject was carried

out as a coop-

erative agree-

ment between

the University of

Georgia Labo-

ratory of Archa-

eology, who pro-

vided the supervision and equipment and the Co-

- \ .... _--

.....

'-"'\ ~<>v~

lumbus Museum of Arts and Crafts, which provided labor. This labor was

,
'\ "~
!j

Df'lAWII\lG

e NO

I~~~ A.

p'"'Of"!!..f'l1v ...,...AIIl~ .. ~P.

ClTV 01" COl.U"'!W~ TO .J W ""0001'10" ~ PAl'IT Of \..UolD \..OT~ ~o, ~'. ':C~f.'Tll. ""'E'!I'Ef'I".'

CITv OF C.Ol..UMe:.uS, (;,t.Or'lGI.o..

o"".et 0" crrv ["G'NEt1"l

I'It""~IQH'!>

~c...ve I". zoo

,;:~ ..~:.'":. ;::;:~. ~ O'!.Ct .. M.'" 1':.'~7

Figure 79. Copy of a 1957 plat of the J. W. Woodruff property containing the Bull Creek site (Schnell's main excavation unit is indicated).

91

The Bull Creek Site

APPROXIMATE LOCATION PROFILE2
o ~/
X-I

\ APPROXIMATE LOCATION PROFILE 3

N
!

0

20

40

I

FEET

GRID POINT N500
ESOO
FIRST USED IN X-2
Figure 80. Plan map showings areas of Schnell's 1959 investigations (adapted from field map on file, Columbus Museum).
92

Schnell's 1959 Excavations Figure 81. Series of photographs showing the 1959 excavations in progress. Field director, upper left; remainder show XU 2 (Columbus Museum files).

The Bull Creek Site
Schnell prepared a series of profile drawings, taking advantage of earlier excavation units which had never been filled. Figure 82 shows a profile made in the railway cut just south of Patterson's House Number 1. This is Profile Number 3 shown in Figure 2. The site's occupation surface is shown to be 1.0 to 1.5 feet below the present surface and overlain by alluvial sand. A portion of the profile had been worn away by fishermen crossing over this bank from the railway cut to the river. This path is visible in the 1950 aerial photograph (see Figure 72 in Chapter 4).

A second profile drawing (Figure 83), shows the depth of the railway fill covering the portion of the site referred to as the peninsula, which extends above the confluence of Bull Creek and the Chattahoochee River. This profile drawing shows the prehistoric occupationallayer beneath four feet of railway fill and four other strata. Schnell notes that "closer to the confluence of Bull Creek and the Chattahoochee River, this fill becomes much thicker" (Schnell 1963). Additional profile drawings are provided in following sections describing Schnell's excavations.

Fishermen's Path Pit

1l2r-:,.qf [JI]
~
1;::-::,'1

Humus Yellow - Tan Sand Occupation Layer Red - Tan Clay Yellow - Tan Clay

N
Tap Root
9ME 1 XU-II
Profile South of Patterson's House # 1
Schnell: 1959

Figure 82. Profile 3 drawing of bank cut near Structure 1 originally excavated in 1936 (redrafted from Schnell 1963).
94

Schnell's 1959 Excavations

Cache of Charred CornCobs

Limit of Profiling

V':~':'~I
r=-=1 t=:::"=:.I 1':.';:~?:::1
~~ .4....f.....f..l..
~
E3

Railway Fill (Loose Sand - Clay Mixture) Heavy Black & Red Marblized Clay Yellow Sand Old Humus Tan Sand Occupational Layer Sterile Red Clay

-5 -6 -7 -8
-9

-10

9ME 1 XU-II

Profile made on River Bank at West of XU - IT

0

1m

0

3ft

Facing East

Figure 83. Profile drawing showing the railroad fill at a location nearer the mouth of Bull Creek and adjacent to XU 2 (redrafted from Schnell 1963).

Schnell's field notes indicate that an old excavation cut near Lester's Structure 1 was examined in some detail. This appears to be the area shown in Profile 3 (see Figure 82) which corresponds to Kelly's Unit 5 excavated

in 1950. A comparison of Schnell's field note entry of June 29, 1959 and Kelly's description of Unit 5 shows the area to have remained essentially unchanged during that decade.

95

The Bull Creek Site
I tested in a hole shown to me by D.W. Chase on Sunday. There are very good indications of a hard packed sun baked floor. This is an old excavation on the opposite side of the railway embankment from XU-2 and slightly north of the excavation.
An area about 1 5 feet long and 2-3 feet deep was cleared. A floor appears to run the full length with a heavy ash lens intrusive through it (fire basin). There is about one foot of midden above this and at least one foot of midden below it. Above the upper midden is a heavy mantle of tan sand with the humus on top of that. There is about 10-1 5 feet of flat area before the railway fill is reached. The railway fill is also very low here. It seems to be a good possibility for excavation. The black midden is deeper here than any other
place profiled (Schnell field notes 6/29/59).
Coupled with Kelly's observations of 1950, Schnell's description provides evidence for another well-preserved structure that would have been located approximately 20 to 30 m south of Lester's Structure 1. An extremely important observation made by Schnell concerns the presence of foot-thick deposits of midden both above and below the house floor. This suggests relatively great intensity and duration of occupation, meaning that the Bull Creek phase occupation extended over a sufficient length of time to allow rebuilding of structures on top of middens associated with earlier structures.
The text from Schnell's manuscript relating to the excavations follows. Information from the field notes are included.
XU-I consisted of four 5 x 5 foot squares excavated in the vicinity of an old well and between two old excavation units, one made by Kelly (Unit 2) and one made by Chase (Test B). This small excavation was made to determine the merits of excavating on the higher ground siightly away from the river {see Figure 80}. This excavation unit, though small, did show that the upper portion of the peninsular area had been too heavily disturbed both by recent occupation and extensive testing by others, to warrant further excavation.
One feature was located in XU-I. This was a shallow basin shaped pit {Figure 84}. This pit was circular, 4.5 ft in diameter, and 0.2 feet deep. This feature contained only a few Lamar
96

plain sherds. It was located in the southeast corner of Square 1 at a depth of 2.5 ft.
Soil stratigraphy in the square, from top to basic red clay, included 0.6 ft of yellow-brown sand immediately below a thin veneer of humic leaf mold, 0.55 ft of dark brown charcoal-flecked midden, and 0.55 ft of yellow-brown sand, resting directly on the basic clay at a depth of 1.9 ft.
After finding that excavation in the area of XU-I would be unprofitable, operations were moved farther out on the peninsula, to XU-II {see Figure 80}. It was there that major stratigraphic testing was undertaken and where the new crew remained for the rest of the period of excavation. A total of seventeen 5 x 10ft tests were excavated here. This area was the only remaining easily accessible and largely undisturbed area on the site. Although this area had been occupied in recent times, as evidenced by the large amounts of modern material, the talus from the railway cut, built before the recent occupation, acted as a "buffer" and protected the lower, prehistoric occupation zone. {Figure 85} depicts a schematic representation of the area.
A grid system was laid out to cover the XU-II area, using the Huscher variant of the Wheeler system. A datum point (labeled N500 E500) was arbitrarily placed to the west and south of the area to be excavated, making all grid coordinate readings to the east and north of that point. As an example, the first test excavated in XU-II came within the coordinates N490-500, E525-530.
SQUARE I OF XU - 1 GRID
N
A
1ft
:======:'30 em
10p of Pit 2.5 ft Below Surface Baseof Pit 2.7 ft BelowSurface
Lamar Pit
Figure 84. Plan map of pit found in XU-I (redrafted from Schnell 1963) .

Schnell's 1959 Excavations

Creek Bank

River Bank ~ W

E

-1:::',::1
~ ~

Railway Fill Talus& Recent Humus Old Humus Tan Sand OccupationalZone Sterile Clay

9ME 1 XU-II

Schematic Profile Schnell: 1959

o

2m

o

5ft

Facing South

Figure 85. Schematic representation of the soil stratigraphy adjacent to XU-2 (redrafted from Schnell 1963).

Individual points referred to in the excavation notes were also designated aerially by this system. As may be seen by reference to the map {see Figure 80}, this placed the imaginary zero point well to the south and west of the whole site. This system is therefore expandable to cover all future excavations on the site. This grid system was tied into a number of temporary bench marks on the site as well as two permanently placed bench marks of the city of Columbus.
The first feature to be uncovered in XU-II was a pit (Feature 1) located at grid coordinates N494.6, E528.3 in the first 5 x 10 excavated. This pit was defined 2.2 ft below the surface, was 1.5 ft in diameter and 1.2 ft deep. This pit contained only a few Lamar sherds, charcoal flecks, stream pebbles and badly decayed bone

which was unidentifiable other than it was not human. To the north of this feature was another, similar pit, located at grid coordinates N500, E428. This pit (Feature 2) was 1.9 ft below the surface, 2.0 ft in diameter and 0.9 ft deep. The most interesting pit encountered in the XU-II excavation was located at grid coordinates N509.9, E528.3. This pit (Feature 3) was encountered at 2.2 ft beneath the present surface, was 1.6 ft in diameter and 1.1 ft deep.
Feature 3 contained one complete Lamar Complicated Stamped bowl with very slight lip notching. This bowl had apparently cracked in firing and was discarded complete (except for the crack) in a refuse pit. Nesting inside this bowl were several large sherds of Lamar Complicated Stamped and Lamar Plain {Figure 86}.

97

The Bull Creek Site

(Schnell 1963). Figure 87 illustrates the bowl in greater detail.

Schnell's excavations provided the largest, systematically recovered collection of pottery known to exist for the Bull Creek site. The screened collections have formed the basis for defining the composition of the Bull Creek phase ceramic assemblage. Schnell's characterization of the assemblage is included in a later chapter.

Figure 86. Field photograph showIng vessel fragmentS exposed in

The XU-II excava-

excavated Feature 3 (Columbus Museum files).

tions also produced a number

of large sherds and some

The Lamar Complicated Stamped vessel fragments appear to have been a large, straight

partially reconstructible vessel fragments that provide excellent examples of the decorative

and high necked jar, while the Lamar Plain vessel

motifs present at the site. Figures 88 and 89

was a wide, dish-like shallow bowl. The complete Lamar Complicated Stamped bowl proved to be

illustrate three examples.

9.25 inches in diameter and 4.8 inches deep

o3
em
Figure 87. Photograph of Lamar Complicated Stamped Bowl from Feature 3 (Columbus Museum files).
98

Schnell's 1959 Excavations

o

3

~~~-.i!iiiiiiiiii~~i

em

:~.':.;',".' : -, : '. :.:. :.' .... ' .. :', . .

...~,,\ -. <...,.:.r."

...-:l':-.~:. '.'t'!-11"-":":-~'',', ."...'. - : "

~'''1.).:r. \4<'.t.:>':1 ~,~~:'':-''':'I$''\\:J~
.

;

'~\\-:;::::::::::;"'" ~t J~\ :').;'i.:.. i"~....'-.~.;;:,:.~:.~:;;r~.
{~~<~'~2;,:~"

.:..~.' ....

- ---..'..' .. . "

........

... ......

....

....

_---- ... ............. ...

---------- -------

. . ,,'"
.."

Figure 88. llIustrations of two Fort Walton-style vessels from XU-2.

99

The Bull Creek Site

o

3

IPSj5-.j5-;;;;;jjjjjjj~~

em

Figure 89. Example of bold incised vessel fragment from XU-2.

Figure 88 shows two Fort Walton-style vessels. The upper, partially reconstructed vessel is a common bowl form with a typical decorative motif of zone punctations. The punctations are found in alternate bands that encircle the neck of the vessel in an undulating pattern. The second vessel illustrated in Figure 88 is an example of a hand-eye motif. This motif occurs consistently on late Fort Walton sites and has been found on other Bull Creek and Stewart phase sites in the region (Broyles 1962; Hally and Oertel 1977).
The rim sherd illustrated in Figure 89 represents a moderately large constricted orifice vessel (estimated to measure 32 em in diameter). The sherd is decorated by deep and very wide incised lines which average 4 mm in width. The motif is quite similar to that found on an effigy vessel found in a later investigation of Bull Creek conducted in 1981 (see Figure 105 in Chapter 6).

Schnell's field notes indicate that a number of postmolds and possible evidence of a burned house floor were found in XU-IT. The location of the postmolds and the larger pits are shown in a composite plan map as Figure 90. These features indicate proximity to a domestic structure if not a direct presence within the excavation unit. The postmolds may represent screens or supports related in some manner to the pits; elsewhere on the site similar pits were found outside of structures.
According to field note entries dated July 10, 1959, evidence of a possible burned house floor consisted of a hard packed floor in squares N51O-520, E520-525, extending into the northwest corner of the adjacent square to the east. Fragments of daub and charcoal were found above the floor. A possible burned postmold was found in square N51O520, E525-530. An entry dated July 12 notes a possible fire pit in the floor.

100

Schnell's 1959 Excavations

recovered during his excavations provided no evidence to support the presence of both an early Bull Creek phase occupation and a late Bull Creek or Stewart phase occupation. Schnell's conclusions, which will be examined in a later chapter of this report, were that only an early Bull Creek phase occupation was present at the Bull Creek site.

N
~

o

10

20

FEET

Schnell's systematic field techniques provide basic information necessary for interpretation of the Bull Creek site. The fundamental step of establishing a site grid tied to property markers allows the reconstruction of an excavation plan map showing locations of most of the excavations on the site through time. Perhaps the most important result of Schnell's excavation was the recovery of a large pottery sample from a controlled (screened and stratified) context. Schnell's interpretation of those data will form a basis of much of the ceramic examination chapter.

Schnell's inves-

tigations were to be the last

professionally conducted ex-

Figure 90. Plan map showing the location of pits and postmolds in XU-2 cavations in the heart of the

(adapted from field note entry).

Bull Creek site. A quarter of

a century later, work on the

Schnell's findings appear to confirm

periphery of the site would be conducted in

Kelly's 1950s interpretations of a structure

connection with the Columbus Riverwalk

based upon the excavation of nearby Unit 4 in

project. A review of that work will be

1950. The richness of the deposits examined

presented in Chapter 7. However, before the

by Schnell provide ample evidence of extended

recent investigations are detailed, a summary

occupation during the Bull Creek phase. How-

of the minor investigations conducted at the

ever, Schnell's analysis of the ceramics

site will be presented.

101

THIS PAGE LEn BLAHK

Chapter 6 Minor Investigations

While the excavations conducted by Lester in 1936-1937, Kelly in 1950, and Schnell in 1959, provide the bulk of the archaeological data needed to interpret the Bull Creek site, there have been several minor investigations conducted from the mid-1940s into the early 1980s which also provide valuable bits of information. These investigations include additional survey work, testing, excavation of features and site assessments related to proposed development of the property. Some of these studies may be considered peripheral to the excavations conducted in the heart of the Bull Creek site but all provide data necessary to reconstruct the extent of the Bull Creek village. One goal of reviewing this work is to confirm or reject the claim of Lester and Patterson that the Bull Creek village covered an area of three-quarters of a square mile (Lester 1938:27). This chapter will begin with the survey of Frank and Barbara Fenenga conducted in 1945 and continue with the known events relating to the other investigations.
1945 Survey by Frank and Barbara Fenenga
While stationed at Fort Benning, Franklin Fenenga and Barbara Wagner Fenenga conducted a surface survey of the Columbus area. Much of this work focused on sites in the lower Bull Creek drainage. Nine of the 15 sites identified lie near the mouth of Bull Creek. The report titled An Archeological Survey of the Vicinity of Columbus, Georgia (Fenenga and Fenenga 1945) contains line .drawings of numerous artifacts and detailed ceramic descriptions which allows temporal context to be readily determined for all sites. The collections are presently curated at the Columbus Museum. This obscure report provides some of the only survey data for an area which has since been affected by extensive commercial development.

The Fenenga's defined site boundaries by surface scatters of artifacts. Apparently, site boundaries were approximated based upon the extent of surface exposure. In certain instances, Bull Creek for example, this resulted in underestimation of site size and splitting into multiple sites. At the time of the Fenenga survey there was no statewide mechanism for assigning site numbers. The Fenenga's identified their sites by numbers 1 through 15 and named each site. As part of the Smithsonian survey (Huscher 1960), David Chase completed Smithsonian Institution site forms for the Fenenga sites and attempted to reconcile their locations with previously known sites such as Bull Creek. This work was only partially successful. As late as 1994 there were still substantial discrepancies within the State of Georgia Site Files concerning the Fenenga sites (Ledbetter 1994b:26).
The Fenengas identified two site locales within the confmes of the Bull Creek site area (Figure 91). These were identified as the Weracoba Creek site (Site 2) and the Riverside Driving Park site (Site 4). Two additional sites (No.3, the Lumpkin Superhighway site and No.5, the Confluence site) were identified across the channels of Weracoba Creek and Bull Creek respectively. Both of these sites also contain Bull Creek phase occupations (Fenenga and Fenenga 1945: 1631). The Weracoba Creek site represents the most important site locale examined simply because the area was documented by no other investigation. The Fenenga description of the Site 2 area follows:
The site is on the right bank of Werocoba Creek at the point where it is crossed by Lumpkin Superhighway. The site is about 150 yds. in diameter, extending on both sides of the highway, the construction of which has destroyed about 60% of the total area of the site.
103

The Bull Creek Site

I

,I

\

\. \

\
\

\1
\\\:~,\ ;~N~~,/

II ""-...:

,,

\ , ~Q""_/

\
\

\
\

,(.i'"'--'U-----<, ,

,I

/J

\

\ \\

I

'\

VtI '~~<;:?t~

\

\/ 5 ,,\ ... <,

Figure 91. Copy of a sketch map showing the locations

of Fenenga Sites 2-5 (source Fenenga and Fenenga

1945:12, Map 3).

Local informants say that a party of archaeologists from the University of Georgia excavated here during construction work which would indicate that this is the Bull Creek Village excavated by Lester. If this is the case "Bull Creek" is an unfortunate misnomer . . .. A great quantity of pottery sherds are found throughout the two feet of occupation zone but flint chips, stones and fragments of shell are extremely rare. No bones
were observed (Fenenga and Fenenga 1945:16).
Although the area collected as Fenenga Site 2 lay well to the east of the richest part of the Bull Creek site and Lester's excavations, a substantial amount of material was still found. The Fenengas collected more than 400 sherds and one projectile point from the site. Figure 92 shows examples of Fenenga line drawings of representative sherds from the site. These sherds clearly conform to Fort Walton and Lamar types of the Bull Creek phase.

. ....!!~.....:"':.:.:=...~..:::::.;... J.:
Figure 92. Examples of ceramics from the Werocoba Creek Site (source Fenenga and Fenenga 1945:17)
104

The Weracoba Creek site represents a portion of the Bull Creek village, with evidence of intensive occupation. The locale is enclosed on three sides by a large bend of Weracoba Creek, which would have formed something of a geographical boundary. We know from other investigations that the Bull Creek village extends from the Chattahoochee River to join this locale, but subsequent research has not been conducted to determine if the village extends farther upstream.
The area identified as the Riverside Driving Park site (Number 4) is a portion of the Bull Creek site which surrounds the borrow pit. Artifacts illustrated by the Fenengas are appropriate and most of the information presented conforms with other reports. Their description of Site 4 follows:
The site is at the junction of Weracoba Creek and Bull Creek and is on the right bank of both streams .... The essential frontage of the site is on Bull Creek, being separated from Weracoba Creek by a long sloping bank. The site is about 100 yds. in diameter. Its central portion has been removed by a borrow pit of recent origin which extends into the subsoil (sandy red clay) to a depth of about 4 ft. The soil of the site is dark and it contains a great many potsherds (290 collected) but very few flint chips or other stones. No special features were observed in connection with the site, the maximum depth of deposit of which seems to be less than 24 inches. All of our collections were made from the exposed face of
the borrow pit (Fenenga and Fenenga 1945:26).
The remaining two sites identified near Bull Creek (sites 3 and 5) are both large and contain artifacts of several components. Bull Creek phase sherds are present on both sites but appear less common on Site 5, which lies directly across the mouth of Bull Creek. Both sites were severely disturbed by development in subsequent years. Site 5 was recently reexamined as part of the Riverwalk survey (Ledbetter 1994). The only additional information on Site 3 relates to Chase's attempt to relocate the area in order to prepare a site form. Chase met with minimal success.

Minor Investigations
Much of the site is covered by "boomtown" type establishments-juke joints and the like. Behind this is a rather thick woods where surface finds are not apparent. Did a little testing but with negative results. May have been a bit off the
beam. Will return (Smithsonian Institution site form: 1959).
1950s Investigations of David Chase
Aside from the information contained in an unpublished manuscript titled The Bull Creek Site, A Type Station of the Middle Lamar Culture, which will be cited as (Chase 1957), relatively little is known concerning these excavations. Chase appears to have excavated at Bull Creek with the goal of acquiring a ceramic collection that could be used to refine the chronological sequence for that portion of the Chattahoochee valley. In the mid-1950s, the artifacts from earlier excavations were stored by the National Park Service and the University of Georgia and were not available for study by Chase. Figure 93 shows the site area as it appeared in a 1955 aerial photograph at the time of Chase's work.
A sketch accompanying Chase's manuscript shows the approximate locations of his test pits (Figure 94). Descriptions of profile stratigraphies are presented for two of the excavations (Trench D and Test E) but notes relating to the excavations could not be obtained. The location of the artifacts has not been determined. The artifacts from this site, like others examined by Chase, should have been curated with the Columbus Museum or University of Georgia in Athens. Most of Chase's papers are stored at the Columbus Museum, although some material has also been distributed to the University of Georgia and the Smithsonian Institution. The Smithsonian comes into play because Chase worked closely with Harold Huscher and the River Basin Survey project. In all probability, and with further effort, the artifacts and additional notes from Chase's excavations will be found at some institution. Because most of Chase's manuscript deals with analytical and theoretical concepts unrelated to fieldwork, only a portion is presented in this chapter.
105

The Bull Creek Site
Figure 93. Portion of a 1955aerial photograph of the Bull Creek site area (Columbus Museum
files). 106

....
" ,"
,', ,'.
. .;'
'," ':','
.... :
....

rr ,~. , ,
~ / OldExcavation
'-~j r--l

I

I

I

I

-:

I

1

I\ JI

-:

'.':
,','

Minor Investigations

J
GN

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

Figure 94. Copy of sketch map prepared by Chase showing location of excavation units (the landmarks are not drawn to scale or accurately portrayed).
107

The Bull Creek Site
Based upon Chase's sketch map, he excavated a total of four test pits and one trench. The tests were excavated adjacent to the erosional feature or gully that parallels the railroad embankment. One of the units (Test B) would have been rather close to Kelly's Unit 2 of 1950. The introductory part of Chase's manuscript provides an intriguing glimpse into the historically pervasive documentary vacuum which has always enveloped the Bull Creek site. As noted, at the time of Chase's writings, most Bull Creek documentation was inaccessible. Chase's comments on the earlier investigations apparently were based on the brief published account found in Patterson's 1950 paper in an Early Georgia. Chase had little or no contact with Isabel Patterson while working at Bull Creek (David Chase, personal communication, 1995). Selected paragraphs from Chase's manuscript follow. Sections of the manuscript that discuss ceramic attributes and cultural contexts will be presented in a later chapter.
This site, located between the Victory Highway and the Chattahoochee River some four miles south of the city of Columbus Georgia, was initially explored by Mrs. Isabel Patterson during the early 1930s. This venture was launched by the discovery of an exposed burial on the south bank of the Bull Creek mouth not far from its junction with the Weracoba Creek. During the course of this exploration, Mrs. Patterson and her assistants explored several burials and house sites. In at least two of the former, effigy dog pottery vessels were found. The bulk of the utility pottery fell into a recognizable culture product of the Lamar Culture. Predominant among the sherd types found at the time were Lamar Complicated Stamped, Lamar Plain and Lamar Bold Incised. Details or notes relating to house plans or post patterns are not known to this writer since the notes in question have never been published nor are they readily available at this time.
Sporadic testing was conducted from the autumn of 1955 to the present (1957) on this site not only to confirm the findings of Mrs. Patterson but also to determine stratigraphy, culture change or any factor not disclosed by her during the course of her explorations ....

Pipes are infrequently found, at Bull Creek only three stems and one bowl fragment came to light, hundreds of pottery disks and one of stone appeared. This is not surprising since the disks prevailed from earliest Lamar times into historic period. Objects of shell were absent as were those of bone. Occasional areas yielding refuse animal bones were found. No projectile points were found during our explorations at Bull Creek, but we suspected that these would be of the small isosceles triangle found elsewhere on sites of the same period and culture . . . .
We suspect that Bull Creek represents an early habitation site of the Middle Lamar Period due to the presence of a small amount of typical Early Lamar pottery included in the lower levels of refuse. At the time of the occupation of this site, which was quite a large town, the Lamar culture had achieved domination of the middle Chattahoochee Valley and were thus able to move from the large inland sites of Patterson and Singer to the River way which they finally controlled. No mounds are associated with Bull Creek and it is suspected that although the tradition of mound building had not perished by Bull Creek times, it was definitely on the way out . . .. Whether the dog pots found by Mrs. Patterson represent a relationship to the Cult is a matter of pure speculation. The general opinion is that they were not.
Bull Creek will be destroyed by the end of 1963 by the River Development Program. This is somewhat unfortunate since much can still be learned from this large site about the Lamar people and their climactic society (Chase 1957).
Chase's excavations were placed to the east of the railroad embankment and beyond the most productive part of the site. The stratigraphy described for Trench D consisted of a humus zone overlying a dark sand (refuse zone) which extended to a depth of 16 inches. Below this, a light sand zone containing charcoal flecks but no artifacts extended to 25 inches. Sterile loam clay lay below. The stratigraphy described by Chase is similar to but more shallow than that described a decade earlier by the Fenengas (Fenenga and Fenenga 1945:26).

108

The second profile described by Chase was for Test E, which lay at the eastern edge of the borrow pit. The stratigraphy here consisted of a thin humus zone above a light grey loam refuse zone which extended to 13 inches below surface. This light zone overlay a dark loamy midden that extended to a depth of 18 inches. Sterile brown clay lay below.
Chase's map shows the locations of a number of old excavations. The excavations to the west of the gully can probably be attributed to the WPA and 1950 excavations. However, the excavations to the east do not appear to match any locations of earlier professionally supervised excavations. These old excavations may be evidence of unrecorded archeological or quasi-archeological digging or quite possibly pothunter looting activity.
Chase's investigations at Bull Creek provided him with sufficient information to characterize many aspects relating to the Lamar period at the site and within the region. The nature of these characterizations will be further examined in a final interpretative chapter of this report.
1960 Assessment by Huscher as Related to Construction of the Walter F. George Reservoir
While Harold Huscher did not personally investigate the Bull Creek site, he did include an assessment of the potential impacts of reservoir construction in his 1960 report Appraisal of Archeological Resources of the Walter F. George Reservoir Area, Chattahoochee River, Alabama and Georgia (Huscher 1960): Relying to a great extent upon the work of Chase coupled with the limited amount of available documentation, Huscher briefly reviewed the history of the site and made a series of recommendations concerning the site. Huscher and Chase were also responsible for preparing the first official site form for Bull Creek. Huscher's assessment follows:

Minor Investigations
9Me1, the Bull Creek Site, is an important Lamar site, regarded as the type site for the Columbus (Georgia) area variant of Late Lamar. The site lies on a point of terrace just north of the mouth of Bull Creek, at elevations between 230 and 240 feet, which throws it into the A3 (important but will not be inundated) category; however, since it lies in the most probable area for harbor development, the dredging of harbor facilities and turn-around basin, and the consequent commercial exploitation, it is regarded as of the highest priority. As indicated under "Previous Archeological Work," the site was excavated extensively in 1936 by Mrs. Wayne Patterson and Mr. Frank Lester, in cooperation with the University of Georgia. Additional excavations were carried out in 1950 by A.R. Kelly of the University of Georgia, assisted by Joseph Mahan. In 1959, Frank T. Schnell, Jr., under a cooperative arrangement between the Columbus Museum of Arts and Crafts and the University of Georgia, mapped the site in detail, relocating, where possible, check points of the previous excavations, and opened two additional squares, uncovering part of a house pattern. The site
should have additional work (Huscher 1960:106).
Huscher was one of many archeologists to stress the importance of fullscale investigation of Bull Creek before complete destruction took place.
We also have records for three additional site investigations conducted in the 1960s. One represents something of a student's project, which recorded information from a looter's pothole. The second relates to archeological remains exposed during commercial development of the area north of the previous Bull Creek excavations, an area to be known as the Go-Kart site. The remains found at Go-Kart by personnel of Columbus Museum are associated with a large historic Creek settlement of the early nineteenth century, which partially overlies the late Mississippian Bull Creek village. A few years later, a second large Creek feature was exposed during construction on the north side of Victory Drive. The area containing the second feature became known for a short time as the Dolly Madison locale after the factory which was being constructed at the time.
109

The Bull Creek Site

Following excavation of the second feature at Dolly Madison, Frank Schnell, Jr., prepared a manuscript describing the excavations (Schnell 1970). Schnell combined the two locales within a single site, 9Me50, which he named the Victory Drive site. In that manuscript, Schnell noted the probability that 9Me50 and 9Me1 overlapped (Schnell 1970).
o XU-B
9Me50

Bull Creek Site

\

9Mel

1960 and 1969 Investigations at the Go-Kart and Dolly Madison locales.

Prior to 1960, excavation at the Bull

Creek site had been limited to the area

extending north from the mouth of Bull Creek

to the area of Lester's House Site 2, approx-

imately 300 m. Lester had inferred that the

village area extended over a much larger area,

but the village portion of the site had been

generally plowed away

(Lester 1938:27). In

1960, a large pit dating

to the historic Creek

N
~

o

50

m

period and a number of postmolds were uncovered south of Victory Drive but well to the north of Lester's excavations. This area became known as the

Go-Kart site. In 1969,

another pit was found

north of Victory Drive

and was named the

Dolly Madison site.

This area also produced

Creek material.

Schnell noted a pro-

bable overlap between

the two areas and

assigned a single site

number, 9Me50.

Figure 95 shows

Schnell's map of

9Me50 and the

locations of the

excavated features.

Figure 96 illustrates the

manner of the initial

discovery of the Go-

Kart features.

Figure 95. Map of the Victory Drive site showing the Feature locations (adapted from Schnell 1970).
110

Minor Investigations
Site of Indian Village Unearthed By Kart Race Track Developers

Kart racing, a modern

fad of a modern people, is

indirectly responsible for a

new archaeological find here.

On the site of what is

soon to be a kart racing

track, the builder yesterday

literally "stumbled across"

the remains of an Indian

village on Victory Drive.

Joe Mahan, curator of

the Columbus Museum of

Arts and Crafts, said the

village is definitely a part of

the Coweta settlement, the

main part of which has been

uncovered previously in

Russell County.

The village which began

taking shape yesterday is

directly across the river from

the Russell County village.

NealWickham, president

of Go Kart Sales Inc., who is

developing the tract across

from Municipal Airport, was

supervising a work crew

yesterday morning in the

building of the race track.

He said he kicked a clod

of dirt and saw a piece of

pottery.

He scratched

around and found other

pieces of pottery.

Wickham, who is a

member of the Coweta

Memorial Assn., a group

devoted to the study of local

Indian lore, quickly

recognized that here was

something which needed

further investigation. He got

in touch with Mahan and

Sfc. David Chase, curator of

the Fort Benning Infantry

Museum.

They arrived in short

order and began investigation

of the site. By noon, they and their helpers had unearthed several sacks full of bits and pieces of pottery and several animal bones.
Chase said the site where the excavation was being made was probably a refuse pit, where the Indian families of the village dumped broken pots, dead animals and other trash.
Further out in the field being developed by Wickham, a road scraper was slowly bringing to light other information about the village.

Mahan and Chase pointed to round, dark spots in the red clay.
"Those are the spots where the posts of houses once stood," Mahan said.
The most important find as of mid-afternoon yesterday, was the barrel of a musket, which Chase said was unusually short compared to muskets of the same period, which he estimated to be between 1780 and 1800.

SFC DAVID CHASl:, JOE MAHAN, NEAL WICKHAM, VIEW GUN BARREL. One of artifacts dug up Saturday on site of Kart Racing track.

Figure 96. Facsimile of unattributed newspaper article describing discoveries at the kart racing track (Colwnbus Musewn files).

111

The Bull Creek Site
The "helpers" mentioned in the newspaper piece were University of Georgia archeologists Bettye Broyles and Frank Schnell, Jr. Figure 97 shows Broyles in the process of excavating one of the large pits. Two large pits, which contained material dating to the late eighteenth or early nineteenth centuries, were eventually excavated.
Figure 98 illustrates the distribution of postmolds in the nearby area mentioned in the newspaper article. The map was carefully prepared by Broyles using an alidade and plane table. While the map accurately depicts the positions of the postmolds, there are no reference points or landmarks which allow the exact location to be known or even the proximity of the postmolds to the large pits. The location can only be approximated as the area of XUA as shown in Figure 95.

The map of postmolds covers an area more than 25 m across. By connecting the dots on the map, it is possible to define a series of arcs which may represent some evidence of structures. However, there are no concentrated or well-defined postmold patterns comparable to the Bull Creek site structures. Also, there are no large Lamar trash pits in the area. Based upon a small collection of sherds found in the parking lot area at the track and now curated at the Columbus Museum, it appears probable that any structure represented by these posts predates both the Creek and Lamar occupations.

1960 photograph showing Bettye Broyles excavathig a large pit in tIiekart track construction 112

Minor Investigations

.10

.9
8
.11
.12

16 .15
.14
.17 18

.6 .13 .7
8

.41 .43

38

.44

.46 .45
.48

.39 .40

.37
36 35

47

54

.32 31

Datum 6.

.21

.1

.30 .26
.29

N

.24

~

.25

o12 3

Meters

Adapted from B. Broyles, 1959

Figure 98. Plan map of postmold pattern in the Go-Kart locale (adapted from field map prepared by Bettye Broyles, on file at the Columbus Museum).
113

The Bull Creek Site
A single large Creek refuse pit was excavated as XU-B in 1969 at the site of the Dolly Madison plant. Figure 99 shows two views of the excavation in progress. These photographs clearly depict the massive site destruction relating to the developing commercial landscape of the time.
This large Creek pit produced substantial amounts of aboriginal pottery and Euro-American trade material dating to the early part of the nineteenth century. The feature, which had been partially destroyed when discovered, produced nearly 2,400 sherds representing a minimum of 59 vessels. At least some of these vessels appear to imitate European vessel forms. The contents provide valuable comparative information for similar material found at Bull Creek. Figures 100 and 101 illustrate selected artifacts.
Figure 99. Two views of pit excavation at the Dotty MawsonplalJf (1969
photographs from files of the Columbus Museum).
114

Minor Investigations
"~;'~'~'it~ -';'i~ji;" L;'i.~"~<~~";;"i"d'~~~
3 em
em Figure 100. Selected aboriginal artifacts from the Creek pit in the Dolly Madison locale.
115

The Bull Creek Site
Figure 101. Selected Euro-American artifacts from the Creek pit in the Dolly Madison locale. 116

Minor Investigations

Commoilly, when a very large feature such as a large Creek trash pit is excavated through soils containing occupation remains from earlier times, some intrusive material becomes deposited within the feature. This was not the case with the Dolly Madison feature as there was no earlier pottery in the pit. This fact provides some evidence that this area lay beyond the boundaries of the Bull Creek phase Lamar village.
1960 Investigations by Tyler and Jackson

As with any well known site,

pothunting continued throughout the decades

that Bull Creek was being professionally

investigated. In one instance, information was

recorded by examining one of these potholes.

In January 1969, youthful Columbus

archeologists Jack Tyler and Terry Jackson

prepared a brief report on their

examination of a pothole at Bull

Creek entitled Report on

Clearance of Pothole A, Bull

Creek Site (Tyler and Jackson

1969). The pothole had been

located along Weracoba Creek in

an area which had not been

previously examined through

professional excavation. The

handwritten manuscript also

contains several photographs and

maps which provide some useful

information for locating the pit in

relationship to other features. Photographs of value show the general landscape of that portion of the site as it appeared in 1969,

,'~.:. .......................

.:::;;-

which was prior to intensive

landfilling.

more than one hundred sherds were recovered. One chert flake and one piece of fire-cracked rock were also found. The present location of the collection is unknown. The most important information presented in the brief report deals with location and stratigraphy.
The location of Pothole A (site A) is given as a lightly wooded bluff edge overlooking the junction of Weracoba Creek and Bull Creek. The face of the bluff was eroding severely at that time and the pothole was found on a slight projection only 32 inches from the edge of the cliff. The area below the cliff was thickly covered with brush (Figure 102). Much of the pothole had been refilled with backdirt (Tyler and Jackson 1969). According to the plan drawing, the pothole measured more than one meter across.

Golf

Driving

Range

...--,

-~ .:

-,

. /r. ~~ !;4:'f,J,/ ,

"
'''',, ,

/

"

,/

Trailer Sales

"
/

"I

// ~:.,.,.<.. : " .

The pothole report deals primarily with the methodology of examining the pothole. A few typical Lamar sherds are illustrated in drawings and one general layout photograph. There is no tabulation of artifacts although

N
t
Figure 102. Redrafted sketch map prepared by Tyler and Jackson showing landmarks in the vicinity of Pothole A.

117

The Bull Creek Site

Backdirt was removed from the pothole down to a layer of blue clay which was believed to be the bottom. The profile drawing is difficult to interpret, but a cultural zone appears to be present that extends to approximately one foot below surface. This overlies an apparently sterile orange clay, which in tum overlies the blue clay.
The report prepared by Tyler and Jackson provides a few additional bits of information helpful in interpreting the extent of the Bull Creek village. The existence of the pothole indicates that someone considered the area to be relatively rich in artifacts. The recovery of more than 100 sherds would seem to be confirmation of that fact. The report helps chronicle the nearly continuous erosion of the Bull Creek site during the present century. The report also provides some data for a portion of the site which has been lost to erosion and is now covered by a modem landfill.
1970s Developments
In the following years the Bull Creek site continued to be commercially developed, with only remnants preserved near the bluff edge. The bluff edges were preserved primarily because trees were needed to prevent further bank erosion. Like other sites which appeared to be destroyed, archeologists viewed the site more from a historical perspective than as a site worthy of continued research. Only Frank Schnell, Jr., archeologist for the Columbus Museum sustained a measure of public awareness.

In Muscogee County Georgia, the lake is normally bound by the previously existing river banks. The Bull Creek site seems to be totally unaffected by the flooding of the lake. A potential detrimental by-product of the lake development is the possibility of the construction of a marina on the site. At the moment, the site is owned by an individual conscious of the intrinsic and scientific value of archaeological sites (Schnell 1973).
1981 Excavations by Jack Fuller
In the 1980s, most of the Bull Creek site was graded for commercial development. Because the development was privately funded, there was no professional monitoring. One avocational archeologist, Jack Fuller, recovered a small amount of material and reported his findings to Frank Schnell, Jr., at the Columbus Museum. Mr. Fuller was contacted in 1995 and asked to provide further details to the information on file at the museum. A sketch map (Figure 103) and brief report were prepared. A portion of that report follows.
Since childhood I have known of, and had an interest in the Bull Creek site. In 1980, the John Deere Tractor dealership opened on the old Indian site .... In 1981, this area was further disturbed by people test driving tractors for the dealership.
The Bull Creek Site

Schnell produced one report in the 1970s which dealt with the Bull Creek site from a planning perspective. In his 1973 report A Preliminary Assessment of Archaeological Resources Remaining in the Walter F. George Lake Area, Schnell considered the effects of the lake on the Bull Creek site.

N

GEORGIA

i

Figure 103. Sketch map prepared by Jack Fuller showing the Bull Creek site in 1981.

118

Minor Investigations

One hot day in June of 1981, I noticed a box blade cut. In the cut lay the remains of two small clay vessels. One lay upside down, the bottom sheared away by the box blade, leaving only a ring of rim sherds in the soil. The vessel diameter was about 3.5 inches, with a white slip, large grit tempering and decorated with two incised lines, crudely executed, forming a Lamar Bold Incised design. The other was upright, but at an angle which allowed 1/3 of the rim to be sheared off. This vessel was of comparable size and temper. A poorly executed Lamar stamp design had been applied. This second vessel was recovered {Figure 104}.

On subsequent visits, the destruction of the site quickened. Only one other feature of note was exposed. That being a trash pit, which was also exposed by the tractor equipment. It was shallow with a diameter of about 3 ft. It contained charcoal, ash and pottery. Lamar stamped and incised wares were dominant. The remnants of a large incised bowl with an effigy rim adorno was recovered from the pit (Figure 105).

o

3

H

em

Figure 104. Small complicated stamped vessel from the Bull Creek site.

..:"~., .v.' .':," .

o

3

"

Figure 105. Artist restoration of effigy vessel from trash pit excavated in 1981 (photograph from files of Columbus Museum). Figure also shows enlarged views of head (2 views) and tail.

119

The Bull Creek Site
What the effigy represented is subject to interpretation; I suggest a dog. There was not enough present to restore but the diameter was estimated at 14 inches. The vessel had a white grit tempered slip.
If there is anything left of the site, I think it will be found on the finger of land that extends out south along the river, under the mantle of clay that was deposited for the railroad bed that used to extend across Bull Creek (Fuller 1995).

Only part of collection made in 1981 survives. The collection consists primarily of rims and decorated body sherds (Figure 106 and 107). This information has been incorporated into the larger collection of pottery from Bull Creek and will be presented in Chapter 8. The collection also contains a few pieces of chipped stone, four pottery discs and two pipe fragments (Figure 108).

B A

o

3

em

o

c
E
Figure 106. Examples of Bull Creek phase sherds from a pit excavated by Jack Fuller in 1981.
120

Minor Investigations B

o

3

P""-.

;

em

G

H Figure 107. Examples of incised and complicated stamped sherds from trash pit excavated in 1981.
121

The Bull Creek Site
Figure 108. Examples of Bull Creek phase pottery objects and Archaic bifaces found at Bull Creek in 1981. 122

Minor Investigations

The material in the

Table 1. Characteristics of Vessel from Fuller's collection

collection lends itself to examination

with respect to vessel form. Many

of the sherds are large and vessel profiles and motifs are reconstructible. Fuller's collection

Vessel Type

Plain

Stamped Incised/ Punctate

includes a minimum of 28 vessels.

Most were recovered from a single

Simple Bowl

3

2

feature.

Table 1 lists these vessels

Carinated Bowl

2

o

15

with respect to vessel type and

Jar

o

5

o

surface treatment. The results are

intriguing primarily because of the

high proportion of incised/punctate

vessels. The high proportion of incised vessels and the presence of certain incised motifs suggest this pit was later than most of the Bull Creek site occupations previously examined. The motif shown in Figure 105b, for instance, is a common Stewart phase motif.

The leveling, filling, paving and associated commercial development of the 1980s destroyed most of the Bull Creek site. Except for a few spots at the bluff's edge where trees were left to retard erosion, little evidence of the site remained. It was not until the mid-1990s that the site was again

examined. The results of that work follow.

123

TJDS PAGE LEn BLANK

Chapter 7 1993-1994 Investigations

The most recent archeological investigations at Bull Creek and the contiguous GoKart site were conducted by Southeastern Archeological Services, Inc, (SAS) and will be summarized in this chapter. The primary goals of the 1993 survey were site assessment and determination of site boundaries to aid in the design of a riverwalk park along the Chattahoochee River.
The SAS investigations included survey, testing and excavation, with most of the excavation occurring at the Go-Kart site. By accomplishing these tasks, the route of the proposed Columbus Riverwalk could be planned so as to minimize disturbance to the sites.
As a result of the survey, relatively small but highly significant preserved deposits were found to remain on both sites. Within the defined boundaries of the Bull Creek site, a route was selected that would not impact any preserved deposits. An alternative route was not possible for the Go-Kart site, and data recovery excavations were required. Because the results of data recovery at the Go-Kart site are detailed in a separate report (Ledbetter 1995c), only those results which directly contribute to an understanding of the primary Bull Creek site components are included in this chapter.
By 1993 Bull Creek was more of a memory than an archeological site. Virtually everything known about the site was to be found in Patterson's brief report in Early Georgia (Patterson 1950) or secured in the file drawers of the Columbus Museum. The only remaining person knowledgeable of the site was the Museum's archeologist, Frank

Schnell, Jr. Schnell had maintained the records of past excavations and fortunately retained much of the "oral history" of the site.
Figure 109 shows the route of the riverwalk and the boundary line used to separate the Bull Creek site from the contiguous Go-Kart site. The Columbus Riverwalk survey was actually the first intensive archeological survey of the Bull Creek site. There was no accurate information available for determining the former site limits or for determining the boundary between the Bull Creek site, 9Me1, and the adjacent GoKart site, 9Me50. Multiple site forms existed for the two sites, but depictions of site boundaries were contradictory. Unfortunately, by 1993 very little of the once extensive Bull Creek site was left to be surveyed. A series of photographs (Figures 110-112) illustrate the condition of the site area at the time of the survey.
Much of the area between the South Commons (Columbus Fair Grounds) and Bull Creek has been extensively modified by commercial development. Modification of the old land surface has resulted from grading for leveling and elsewhere extensive landfilling of low-lying areas. Figure 110 shows a portion of the Bull Creek site which has been both graded and filled to create a level surface following severe erosion and gullying. Figures 111 and 112 show the area adjacent to Bull Creek where renewed landfilling (to a depth of approximately 50 ft) was continuing as our survey was in progress. According to one local informant, this landfilling has been ongoing since the 1960s with only limited beneficial results.

125

The Bull Creek Site

N
i

0

200 Feet

e

0

66m

Figure 109. Route of riverwalk through the Bull Creek and Go-Kart sites. 126

1993-1994 Investigations
Figure 110. Photograph of the Bull Creek site as it appeared in 1993, view to the east.
Figure 111. Photograph of landfilling at the Bull Creek site in 1993, view to the southeast. 127

The Bull Creek Site

Figure 112. Photograph taken in 1993 of the approximately 50 ft thick landfill at the edge of Bull Creek.

As previously noted, the riverwalk corridor through the Go-Kart site could not be designed to avoid preserved cultural deposits along the upper edge of the wooded bluff line. Shovel testing and test pit excavation in that area produced evidence of intact cultural deposits consisting of both midden and features. The primary occupations indicated by artifacts from these deposits were Archaic and Early Mississippian (Averett phase). Minor aboriginal components included late Mississippian, Bull Creek phase, and historic Creek. There were also artifacts that dated from the midnineteenth through twentieth centuries (Ledbetter 1994b:69). Data recovery was recommended and conducted by SAS in 1994 (Ledbetter 1995c).
The survey of the Bull Creek site was substantially more involved due to the large area to be examined and the extensive filling.

The survey methodology and results will be briefly summarized from the survey report (Ledbetter 1994b). Figure 113 depicts the site at the time of the survey and shows the locations of all collection and testing units.
A visual assessment of the site in 1993 was far from encouraging. The site area showed the scars of years of neglect and abuse. The impact of the 1920s construction of a railroad bridge across the mouth of Bull Creek were clearly seen. Large borrow pits shown in mid-twentieth century photographs and maps were probably excavated to provide fill dirt for construction of the.railroad levee and trestle. In the mid-century the area was used for recreational racing and several associated structures were constructed. Modern commercial development has affected the site through grading and filling (Ledbetter 1994b:70).

128

1993-1994 Investigations

Graded Field

Collection Area A
\

Riverwalk Route

N
A

o

SOm

Positive Shovel Test o Negative Shovel Test

Figure 113. Plan map the Bull Creek site in 1993 showing locations of SAS shovel tests and surface collections.
129

The Bull Creek Site
A primary task of the present survey was to determine just how much remained of the site since the last investigation nearly 35 years ago. The field archeologists were initially guided through the site area by Columbus Museum's archeologist Frank Schnell. Mr. Schnell provided substantial assistance for interpreting the site by pointing out the general locations of old excavations. Most of the site had been graded and filled since the last excavations but the approximate location of Schnell's 1959 block was ascertained. The primary reference point remaining on the site is the railroad grade and a large depression left from the 1936 excavation of the cemetery.
The Bull Creek site was examined through surface inspection and shovel testing (see Figure 113). A few small pockets of exposed ground were found on the site. These exposed areas were collected separately and designated as separate collection units. Shovel tests were excavated as systematically as possible. The goals of shovel testing were two-fold. First, shovel tests were required to determine the depth of cultural deposits, when present. Second, shovel tests were use to assess the severity of grading and modern filling ...

In terms of surface

Figure 114. Photograph of 1993 shovel testing (Shovel Test 4) .

exposure, there was generally no

correlation with subsurface re-

mains. Across most of the site, artifacts found on

The photograph shown in (Figure 114)

the surface represent either displaced material from

was taken at the location of Shovel Test 4. Shovel

land clearing or residual material remaining after

Test 4 represents the northernmost extent of the

sheet erosion has removed all soil. The conclusion

preserved site area. Because the slumping gully

resulting from the excavation of shovel tests is that

walls to the south of Shovel Test 4 produced

only one portion of the site remained preserved.

ample evidence of an intact site, further shovel

That area lay beneath the old railroad fill. Unfortu-

tests were not excavated in that direction. The

nately, that portion of the site was being actively

large slump shown in the (Figure 115) photograph

eroded away at the time of survey.

begins approximately 5 m south of Shovel Test 4.

130

1993-1994 Investigations
Figure 115. 1993 photograph showing the large erosional feature impacting the Bull Creek site (the midden zone is being pointed out by the archeologist).
131

The Bull Creek Site
The field crew did take advantage of the site erosion to further investigate the site. This investigation included cleaning a section of the gully wall to examine site stratigraphy in the remaining (preserved) portion of the site (Figure 116). This was accomplished by utilizing the area of most recent slumping which was nearly vertical and extended across all soil zones associated with site (Figure 117). Because of the freshness of this slump, little cleaning was necessary to prepare the profile for mapping and photography.

The slumping midden soil at the bottom of the gully was also examined. Several large sherds were found protruding from the boulder-size chunks of soil that had fallen into the gully but much of soil and associated artifacts had obviously
been washed away (Ledbetter 1994:79). Table 2 lists artifacts found by the survey crew in the midden slump at the bottom of the gully. The artifacts were recovered by troweling through large boulder-sized chunks of midden.

1993 photograph of gully wall following profile cleaning. 132

1993-1994 Investigations

0-

)f/

lOOcm

200cm

I

I

,'/ Surface

~1/

300cm I

400cm I

Red Brown Mottled Clay

------ ------------- -IOOcm-

~------------------Yellow Brown Sandy Clay

-200cm-

Old Humus Light Brown Sand

Dark Brown Sandy Loam (Midden)

- - - - - - - - - - - - -------

Medium Brown s a n d = = - - - - - - - - - -

Figure 117. Profile drawing of a portion of the gully wall examined by SAS in 1993.

Table 2. Tabulation of artifacts recovered from gully midden slump from 9Mel.

I 1> . Prbveniellce .

.... Description .

1

<Contents

I

Surface Collection Area A

Midden slump in large gully at southwest corner of site. Material
recovered by troweling large blocks of midden at base of gully.

1 chert Archaic stemmed PP/K 1 chert utilized flake
2 quartz tertiary flakes < 2 em
1 fragment of ground hematite 1 fire-cracked rock
78 Lamar complicated-stamped sherds 1 Lamar vertically pinched rim sherd on
comp. stamped body 2 Lamar noded rims 1 Lamar folded pinched rim sherd 2 Lamar notched rims, plain body 1 Lamar notched and incised rim sherd 4 simple rims on plain body sherds 1 simple rim on comp. stamped sherd 21 plain grit-tempered body sherds 13 plain sand-tempered body sherds 5 burnished sand-tempered sherds 2 Mercier Check Stamped sherds 1 Fort Walton Zone Punctate rim sherd 1 engraved rim sherd
1 incised rim lug Artifact Total: 140

133

The Bull Creek Site

The profile recorded from the gully wall is comparable to the stratigraphy previously recorded for this portion of the site (see Figure 83 in Chapter 5). In the section of the SAS profile drawing shown in Figure 117, various layers of railroad fill greater than one meter in depth cover an old humus zone. The humus zone is compressed by the fill but is conspicuous because of its coloration and high loam content. The humus zone covers a light brown sand which in turn covers a darker brown sandy loam midden zone. Cleaning of the midden zone produced a few Bull Creek phase sherds but no lithics. The midden zone contained flecks of charcoal but no Mississippian features. Of interest, the Archaic projectile point was recovered from the railroad fill of the profile (yellowbrown sandy clay). This indicates the projectile point was brought in with the fill, perhaps from another site area such as 9Me50.

The features (pits) shown in

Figure 117 are related to the railroad.

These features represent holes cut

through the old humus zone that were

filled with the reddish brown clay. The feature fill is homogeneous with

the lowest zone of railroad fill.

Figure 118 shows a photograph of the Figure 118. Photograph of section of profile shown in previous

southernmost feature.

profile drawing (note pit extending from uppermost railroad fiII

into midden) .

Because previous excavations

of the Bull Creek site have already produced

features are present at the periphery of the

significant information, the Bull Creek site is

cemetery that would provide information for

recommended eligible to the National Register

better interpreting the cemetery area. This

of Historic Places. Portions of the site remain preserved that may contribute more information toward an understanding of the region's

area of the site should be stabilized and protected.

prehistory. The preserved portion of the site

bounds the cemetery excavations of the WPA

era. We do not know at present if the entire

cemetery was actually excavated or if related

134

The SAS survey determined that the remainder of the site has been destroyed. The SAS archeologists were instructed to find a route for the riverwalk and a location for the bridge across Bull Creek that would provide no impact to the site. This can be accomplished by routing the riverwalk and bridge across the creek near the location presently occupied by a pipeline. This area lies near the mouth of Weracoba Creek. If this is accomplished, further archeological field work will not be required (Ledbetter 1994:70-84).
Site boundaries for Bull Creek were reconstructed following the 1993 survey

1993-1994 Investigations
(Figure 119). These boundaries, which were prepared in large measure to meet the needs of cultural resource management, represented a combination of former estimates of site boundaries as shown on site forms and the results of the 1993 shovel testing. Essentially the boundary between site 9Me1 and 9Me50 was placed at a large erosional gully. However, subsequent excavations conducted for data recovery of the Go-Kart site, 9Me50, produced evidence that this boundary was not real. These excavations produced evidence that the Bull Creek village actually extended a substantial distance onto the Go-Kart site. The results of data recovery will be briefly summarized.

Figure 119. Tentative approximation of the boundaries of 9Mel and 9Me50 based upon survey results (photograph source, 1950 SCS).
135

The Bull Creek Site
Data recovery excavation in the fall of 1994 was restricted to a narrow, 7 m wide, corridor through the Go-Kart site. Along this corridor, the site was examined by shovel

testing, text pit excavation, backhoe trenching, and machine-assisted block excavation. Figure 120 shows the larger excavation units of data recovery.

I

I

I

I

I

480E

S20E

560E

600E

640E

- 540N -SOON ,.,/ -460N

GravelParking Lot ExistingBuilding

o

30

m

Test Pit c::::l Backhoe Trench
o Isolated StandingTree
---- River Walk Center Line ---Site Boundary

- 400N

'". / .~
.

-360N

- 320N

- 280N

Park.man Cul-de-sac

Figure 120. Plan map of data recovery area of the Go-Kart site showing trenches and block excavations 136

Excavation in the northern portion of the Go-Kart site yielded material and features associated primarily with an early Mississippian Averett phase occupation. The discovered features were postmolds and small pits. There was also a thin scatter of Archaic lithics. Perhaps the most significant contribution of those excavations was the procurement of several Averett phase radiocarbon dates (Table 3). These assays confirm previous age estimates (Schnell and Wright 1993; Ledbetter 1995a) and support a range for the Averett phase between the late tenth and late thirteenth centuries.

1993-1994 Investigations
A few Bull Creek phase sherds were found in the northern part of the Go-Kart site but there was not evidence of intensive occupation. Quite possibly, the northern portion of the data recovery area represents the fringe of the Bull Creek village occupation. Whether this assessment is true or not, the available information from data recovery does indicate a scatter of Bull Creek phase pottery extending a minimum of 600 m north from the mouth of Bull Creek. The area beyond that point has been totally destroyed.

Table 3. Averett phase radiocarbon dates from the Go-Kart site.
.

Beta 78127 Beta 78128 Beta 78129

1020 70 B.P. 820 80 B.P.
920 50

A.D. 1015 A.D. 1235 A.D. 1065, 1075, 1155

-- ...
........
,,
.< .....
A.D. 980-1040
A.D. 1170-1280 A.D. 1035-1195

Excavation in the southern part of GoKart produced cultural remains associated primarily with the Bull Creek phase and historic Creek occupations, along with a continuing scatter of Archaic lithics. Data from these excavations indicate that structures associated with the Bull Creek village extended a minimum of one hundred meters beyond Lester's northern most excavations. Large excavations designated Block B and Block D produced important information relating to the Bull Creek site.
Block B (Figure 121) was excavated at the southernmost edge of the Go-Kart site. The block actually intruded upon the edge of a

large gully which had tentatively been used to define the boundaries between the two site areas. Machine stripping indicated that the area had been filled within relatively recent times but beneath that fill the ground surface was badly eroded. The only cultural features to survive this erosion were a large historic Creek pit and an accompanying square postmold. The large pit was 160 em in diameter and 110 ern deep. Figure 122 shows views of the pit during excavation. Contents of the feature included partially reconstructible vessels, various Euro-American manufactured items (Figure 123), and ,a relatively large amount of Bull Creek phase pottery.
137

The Bull Creek Site

I 595E

I 600E

I 605E

0 0

0

a

0 0 ()

0

0

00

00

Shallow

0

Depression

a

0
Modem Fill Shallow Depression
0

9ME50 Block B
N
~
o
e m
[23 Filled Gully
o Tree

Figure 121. Plan map of block B at the Go-Kart site. 138

335N 330N -

1993-1994 Investigations
Figure 122. site.
139

The Bull Creek Site
9
--- 1 . em
Figure 123. Examples of vessels (large Chattahoochee Brushed jars and undecorated water bottle, greatly reduced in size) and glass and metal beads (greatly enlarged) from the Creek pit. 140

The large Creek pit also produced more than two hundred Bull Creek phase sherds in the feature fill. These sherds were interpreted as material displaced from the surrounding midden during feature excavation and subsequent use. The number of sherds found within the Creek feature indicated a dense midden deposit was once present.
Block D, located to the north of Block B, was particularly important with respect to the Bull Creek village. Block D was located at a point where the riverwalk corridor made a short jog toward the river. Excavation of the block exposed a nearly complete post pattern representing a Bull Creek phase structure (Figures 124 and 125). Excavation produced a substantial number of postmolds and a few small pits associated with the structure. The arrangement of the postmolds suggests a subrectangular structure approximately 7 by 7 m. The perimeter postmolds were relatively

1993-1994 Investigations
large, generally ranging from 15 to 25 em in diameter and as much as 25 em deep. Postmolds which form interior partitions were of comparable size. Large interior support posts were as much as 30 em in diameter and 35 em deep. In most respects, the Block D structure appeared similar to Lester's Structure 1.
The SAS investigations confirmed that most of the Bull Creek and Go-Kart sites have been destroyed. A small strip of both sites survives along portions of the wooded river bluff but these areas are subject to riverbank erosion and may eventually perish. Portions of the cemetery area also remain preserved but that area appears to be even more at risk from erosion. With the exception of a few extremely large and deep features such as that found in Block B, there appears to be little likelihood of discovering intact remains in the remainder of the former site area of the Bull Creek site.

Figure 124. Photograph ofBull Creek phase structure in Block D of the Go-Kart site, view to the west.
141

The Bull Creek Site

I
564E

I 566E

I 568E

I 570E

I 572E

I 574E

I 576E

I 578E

9ME50

Block D

368N-

N
!
o

o

0

2

E

0

m

a

o 0

0

o

00

o

o 00

o

0

o0 0

o

o
o

o0

o

o

o 0 o 00

0

o

0

0 o

00

0

366N 364N362N360N -

o0
o
o

o (J
o
o
o

358N356N354N-

352N-

Figure 125. Plan map of Block D at the Go-Kart site.

There is a slightly greater potential for feature preservation in the area of the Go-Kart site. The data recovery investigations suggest that a series of structures associated with the Bull Creek village extended along the river for

a minimum of 450 m from the mouth of Bull Creek. The results of the SAS investigations with respect to the extent and layout of the Bull Creek village will be pursued further in Chapter 9.

142

Chapter 8 Bull Creek Site Material Culture

Perhaps the weakest aspect of most previous investigations at the Bull Creek site has been the paucity of artifact analyses and synthesis of material remains within a cultural framework of the late Mississippian period, and particularly the Bull Creek phase. In this chapter we will present tabulations of artifacts and interpretations found in the manuscripts and notes of previous investigations. Information pertaining to the material culture of the Bull Creek site will accordingly include the unpublished compilations of data from sundry investigations and new forms of analyses utilizing the surviving collections. An examination of the Bull Creek phase ceramic assemblage will be presented first. Following that, the relatively small amount of information available for other categories of Bull Creek phase material remains will be examined. The latter will be brief because lithic materials, in particular, accounted for a minor part of the Bull Creek phase material culture. Unlike earlier people, it appears that chipped stone projectile points were not being manufactured. The material remains from other components will be noted but not detailed. The examination of other components will be especially concise because that material is covered in detail in a separate report dealing with data recovery of contiguous site 9Me50 (Ledbetter 1995c).

complicated stamped pottery. The site also lay at the northern edge of a Gulf Coast tradition characterized by pottery decorated with Fort Walton zone punctate designs. Figure 126 illustrates the geographic location of the Bull Creek site in relation to the areas of these two ceramic traditions. It should be noted that the areas shown are the approximations of two researchers and subject to variable interpretations.
o

Bull Creek Phase Ceramics
During the late Mississippian period, the Bull Creek site was positioned at the boundary of two groups who developed distinct ceramic traditions. The site lay at the western boundary of a region with a ceramic tradition characterized by Lamar

N
A

o

ISO

kin

Figure 126. Location of Bull Creek showing boundaries of Lamar and Fort Walton ceramic traditions (based upon maps by Williams and Shapiro {1990:5} and Scary {1985:200}).

143

The Bull Creek Site

The pottery from the Bull Creek site is typical of late Mississippian decorative styles and vessel forms found throughout the southeast. However, the combination of ceramic characteristics based upon two ceramic traditions raises the probability of greater diversity of the Bull Creek site assemblage compared to sites located within the heart lands of either tradition. Based upon analysis of ceramics from other sites this would appear to be the case, but as Schnell has noted:

The site has over the years become a favorite example of a certain variant of Lamar. Despite this repeated use as an example, nothing has been published on the site in the way of a comprehensive study of the site and its implication
in southeastern archeology (Schnell 1963).

After three decades, the essence of Schnell's commentary remains. While results of analysis of ceramics from sites such as Rood's Landing (Caldwell 1955), 9Cy51 (Broyles 1962), and Park Mound (Hally and Oertel 1977), have been interpreted as typical of the Bull Creek phase, the actual comparability of these collections to the Bull Creek type site has remained unknown. In recent years, some steps have been taken to better define the characteristics of the Bull Creek phase. Particularly important are efforts to quantify ceramic assemblages (Knight 1994; Schnell 1986, 1990; Schnell and Wright 1993; Scarry 1985) and define the geographical distributions of sites relating to this time period (Hally 1994; Schnell and Wright 1993). Hally's (1994) approximation of the boundaries of the phase are shown in Figure 127.
The work of these researchers has refined our understanding of ceramic types and ceramic relationships. Much of the following presentation of ceramic data will simply help to quantify and hopefully better describe specific attributes of the Bull Creek site pottery assemblage. Before examining the site ceramics in detail, additional background information will be presented.

N
A

o

150

km

Figure 127. Geographical extent of the Bull Creek phase (adapted from Hally 1994:151).
Part of this background material includes the interpretations of earlier researchers, thus providing historical context for the data presented in this chapter. The background material is also meant to impart some of the diverging points of view which have existed, and still exist, concerning this phase. This divergence eventually culminated in splitting the traditionally defined Bull Creek focus into early and late phases (Schnell 1986, 1990).
The blending of ceramic traits from two areas created a basic set of attributes which once distinguished the Bull Creek focus, and now defines the Bull Creek phase pottery assemblage. The basic characteristics of this blending have been described by several researchers but perhaps most concisely as follows:

144

Bull Creek Site Material Culture

Bull Creek phase ceramics can be considered a blend of generic Lamar and generic Fort Walton of this period [late Mississippian]. The zone punctated type Fort Walton Incised becomes the principal Gulf Tradition decorated type, while Lamar Complicated Stamped becomes common as the principal "South Appalachian Mississippian"
type added to the local assemblage (Knight and Mistovich 1984:224).
The observations of past researchers provide a historical perspective relating to the recognition of the rather unique characteristics of the Bull Creek phase ceramic assemblage. In 1951 Kelly noted:
. . . while the rim treatment, paste, and general morphology of the bulk of the stuff at Bull Creek is definitely in the time level and "feel" of Lamar, the specific designs have a Savannah stamped tradition; also there is a lot of pottery that must refer to influences coming up the Chattahoochee from Florida, Fort Walton influences. This means a very interesting variant in Bull Creek of the widespread Lamar manifestation
(Ledbetter 1995b:49).

similarities to Ft. Walton of northwest Florida, to the Lamar Culture of Central Georgia, and to a lesser extent Safety Harbor of the Florida east Coast. There are even some specific ceramic similarities to Moundville, Alabama, and the Dallas Culture which succeeded Hiwassee Island in eastern Tennessee. The type of notched rim strip, for instance, is found both at Moundville and in Dallas associated with effigy rim adornos.
The Later Period Culture at Rood's Landing is neither Ft. Walton nor is it Lamar, but represents rather a borderland culture between the two, apparently more closely oriented toward Ft. Walton. The variant of Lamar Complicated Stamped found at Rood's is particularly interesting in appearing to be directly derived from the older Savannah Complicated Stamped type (Figure 128) . Lamar Bold Incised, as the type is known in central Georgia, is practically absent, and its place is taken by Rood's Incised which more closely resembles Floridian variants of a Ft. Walton time level (Point Washington Incised and Pensacola Incised) ...
(Caldwell 1955:45).

A few years later Joseph Caldwell made very similar observations concerning a related ceramic assemblage at Rood's Landing. According to Schnell (1986) the material described by Caldwell as the Later Period Culture should date to the latter portion of the traditionally defined Bull Creek phase. While the Rood's occupation may date a few decades later than the primary occupation at Bull Creek, Caldwell's observations are still important for interpreting the general time period.
The Later Period Culture, which we know in some detail from our work on the summit of Mound A, shows ceramic

.8

C

r

,-c/ G. 7

o

.,

i

i

JOESIG/f 1I10TII'"S Or LA,4UR CON,P,L/C'ATE.o
AT /?oo.oir LA#~/#G'

Figure 128. Lamar Complicated Stamped Design motifs from Rood's Landing (redrawn from Caldwell 1955: Figure 7).

145

The Bull Creek Site
The first presentation of Bull Creek pottery to the archeological community became the task of Isabel Patterson even as the WPA excavations progressed (Patterson 1936). Patterson's paper highlighted the more spectacular burial finds, but results of preliminary analysis of more than two thousand sherds were summarized.
A preliminary analysis of 2,240 sherds from the Bull Creek site show a surprising resemblance with the 5,000 sherds studied and tabulated from the Lamar Village site on the Ocmulgee. There are no major differences in the field of design or stylistic treatment. In the Bull Creek collection there are a few sherds, a fraction of one percent which are probably related to the Lower Mississippi types exhibited by James A. Ford in his work in this area. A few similar incised and punctate types found at Lamar may be ascribed to the same influence. These two sites, so far removed on different river systems, show such a striking similarity at a time [far removed] considered to be either early protohistoric, just antedating de Soto, or late prehistoric. Forty-nine percent of the Bull Creek sherds are of stamped ware, forty percent plain, two percent incised and one percent incised and punctate. All other designs are less than one percent and ninety
percent of the pottery is grit-tempered (Patterson 1936).
The first attempt at a detailed description of Bull Creek pottery appeared in the mid-1940s (Fenenga and Fenenga 1945). The Fenenga survey, which focused on the lower Bull Creek drainage, resulted in relatively large collections of pottery. Because the Fenengas were not familiar with the ceramic typologies of the region, local type names (ware groups) were used and detailed descriptions were presented. The Fenenga types consisted of Weracoba Grit Tempered, Moon Dairy Sand Tempered, Columbus Fine Sand Tempered, and fiber tempered wares (Fenenga and Fenenga 1945:4-8). These wares generally equate to Schnell's (1963) Bull Creek variant of Lamar (Weracoba Grit Tempered), several types of Woodland and early Mississippian pottery (Moon Dairy Sand Tempered), several types of late Mississippian and historic
146

Indian pottery (Columbus Fine Sand Tempered), and Late Archaic (fiber tempered). The Fenenga's description of Weracoba Grit Tempered provides an excellent characterization for many attributes of the dominant category of coarse grit-tempered pottery found at the Bull Creek site.
The pottery designated as Weracoba Grit Tempered in this report is a thick coarse ware varying but distinctive in color. The temper is of quartzite grit particles ranging in size from 1/2 to 2 mm. The paste used is very fine and smooth. Color is either a greyish white, salmon pink or buff with an occasional dark to black sherd. The surface is characteristically smoothed giving the effect of a fine hard surface repeatedly broken by grit particles. Construction was by the annular method, and lines of fracture are frequently between the smoothed-together coils. Sherds average about 9 mm in thickness. The ware is almost invariably decorated and complicated paddle stamping is the most commonly used motif. The stamping is not clear but appears to be a combination of curvilinear and straight lines. The stamp was used over the entire surface of the vessel. Incising or incised lines in combination with punctate patterns are also present. These designs usually are placed close to the rim. They do not appear in conjunction with paddle stamping but are sometimes supplemented with pinched-up knobs. Another form of decoration restricted to the neck or rim area is a pinched-up or appliqued collar which is decorated with nodes, rosettes and punctations. This collar seems to be an ela-
boration growing out of the folded rim (Fenenga and Fenenga 1945:8).
Following extensive test excavation at Bull Creek in the mid-1950s, David Chase produced the following description of the pottery from the Bull Creek site.
The predominant pottery type found during the recent tests is, without doubt, Lamar Complicated Stamped. This is a thick, gray to black ware, heavily grit tempered with particles of the tempering medium extruding to the surface. In almost every rim of this type, the noded collar is present. Nodes are either applique or else pinched. A minority of sherds lack this collar. Stamping is poorly controlled and random overstamping with poorly cut blocks or paddles seems to be the rule.

The vessel is globular and the stamping is applied over all. The rim is usually everted and the lip squared or rounded. No handles appear with this type. Lamar Plain is either a variant of the Lamar Complicated Stamped but without the stamping, or else appears in the form of small bowls. The lips are occasionally notched. The temper is sometimes grit and in a few instances sand is used. The grit tempered pots do not always show the grit on the surface. The noded collar sometimes appears on this ware and when it does the rim is everted. Other rims indicate straight sided or inverted bowl rims. A third type is Lamar Bold Incised. Rather deep incised, usually curvate lines, appear in the region of the rim. Sometimes these consist of only two or three parallel lines circumscribing the vessel. Interlocking loops also appear on some pots. In the cassuela bowl, which is making its first appearance in Middle Lamar times, the incising is applied to the inner ledge lip. Notched rims often appear with the Lamar Bold Incised vessels. Rim adornos, usually lugs, nodes and more rarely effigies, usually of birds occur on this type of ware. Included in the Lamar Bold Incised is a variant of the Fort Walton Punctate. In actuality, most sherds decorated in this manner cannot be distinguished from the classical Fort Walton punctate type. Another Fort Walton type, acquired by trade perhaps, is the Lake Jackson Plain. This occurs as a minority ware at Bull Creek. Shell tempering is also a minority manifestation. These last traits suggest initial contact with the Fort Walton culture to the south in Middle Lamar times, a relationship which appeared to have gathered strength in later times in view of the predominance of Fort Walton ceramic types on such sites at Engineer's Landing, Kendrick's and Bickerstaff, all post-dating Bull Creek. This may well be a clue as to the origin of the Lower Creek and Appalachicola alliance in protohistoric times.
A very few sherds found were of the burnished black slip variety. This is another ware which became common in later times. These were often incised, never stamped and never with handles. Adornos in this ware are often seen usually as animal or bird effigies. An acquaintance with the Dallas Focus is suspected here. Pipes are infrequently found, at Bull Creek only three stems and one bowl fragment came to light, hundreds of pottery disks and one of stone appeared. This is not surprising since the disks prevailed from earliest Lamar times into historic period ....
Bull Creek was termed 'Middle' Lamar for two main reasons. First, it did not produce the massive, almost undecorated thick sand tempered ware which characterized the Lamar site at Pat-

Bull Creek Site Material Culture
terson. Secondly, it lacked the abundance of fine burnished black ware, shell tempered buff ware and red and yellow slip pottery so characteristic of ceramics of the proto and early historic periods.
We suspect that Bull Creek represents an early habitation site of the Middle Lamar Period due to the presence of a small amount of typical Early Lamar pottery included in the lower levels of refuse. At the time of the occupation of this site, which was quite a large town, the Lamar culture had achieved domination of the middle Chattahoochee Valley and were thus able to move from the large inland sites of Patterson and Singer to the River way which they finally controlled. No mounds are associated with Bull Creek and it is suspected that although the tradition of mound building had not perished by Bull Creek times, it was definitely on the way out. The Neisler Mounds on the Flint and the Rood's Landing Mounds represent some of the last of the great temple mounds of the southeast. By Bull Creek times the Southern Cult has probably all but died out. Whether the dog pots found by Mrs. Patterson represent a relationship to the Cult is a matter of pure speculation. The general opinion is that
they were not (Chase 1957).
A detailed analysis of ceramics recovered from the Bull Creek site during the 1950 and 1959 excavations was initiated by Schnell in the early 1960s but only partially reported in his Bull Creek manuscript (Schnell 1963). Nevertheless, the analysis completed by Schnell is significant and constitutes a substantial part of the site's ceramic data base.
In terms of actual published data, the most detailed information relating to Bull Creek phase ceramics comes from sites 9Cy51 (Broyles 1962) and Park Mound, 9Tp41 (Hally . and Oertel 1977). Both site reports contain thoroughly described, tabulated and illustrated accounts of ceramic collections which generally conform to Bull Creek. Observations concerning ceramics at Park Mound are especially important because of temporal implications. Based upon several attributes, rim fold width in particular, Hally concluded that the Park Mound Bull Creek phase ceramics might date as late as the late sixteenth and early seventeenth centuries (Hally and Oertel 1977:51).

147

The Bull Creek Site
Schnell now believes the ceramic assemblages from 9Cy51 and Park Mound date to the latter part of the traditional Bull Creek phase (Schnell 1986, 1990). According to Schnell's view, the Bull Creek site (9Me1) and the contemporary occupation at Kolomoki (9Er1), are the primary examples of the earlier Bull Creek phase, which should date predominantly to the fifteenth century, while 9Cy51 and Park Mound, and the contemporary occupation at Rood's Landing (9Sw1) are representative of a predominantly sixteenth century Stewart phase (Schnell 1990:67; Schnell and Wright 1993:21).

Schnell's two phases are distinguished by differences in relative proportions of complicated stamped compared to plain, incised and punctate pottery. Bull Creek phase assemblages contain complicated stamped pottery in excess of fifty percent and low counts of incised/punctate wares. Stewart phase collections contain more plain wares (>50%), with complicated stamping dropping to about twenty percent and incised/punctate counts increasing to approximately fifteen percent. Minor amounts of check stamping occur with both phases (Schnell 1990:67-68). Figure 129 shows a graph originally prepared by Schnell (1985) which portrays these differences.

VARIATION IN LAMAR POTIERY ASSEMBLAGES IN CHATTAHOOCHEE VALLEY SITES

70% 60% 50% 40% 30% 20% 10% 0% ....1--11-
9MEI
Incised

9ERI

9CY51 SITES

9SWI

~ Complicated Stamped

9TP41
B~ PIam.

Figure 129. Graph showing differences in pottery collections from Bull Creek phase (9Me1, 9Er1) and Stewart phase sites (9Cy51, 9Sw1, 9Tp41), adapted from Schnell (1985).
148

Bull Creek Site Material Culture

Stewart phase collections are further distinguished by the addition of certain incised motifs similar to Pinellas Incised and the incised pottery of the Avery and Atasi phases to the north and northwest (Knight and Mistovich 1984:224). This would include the type Rood's incised (Caldwell 1955:27). Figure 130 illustrates an example from 9Cy51 of perhaps the most common Pinellas-like motif attributed to occupations of this latter phase (Hally 1994:154). Only one vessel with this motif has been identified at the Bull Creek site. That vessel was recovered from a pit excavated in 1981 and appears to be a rare example of a Stewart phase feature located on the Bull Creek site.

Figure 130. Drawing of typical Stewart phase incised motif (redrawn from Broyles 1962: plate 7).
Primary Pottery Attributes

In the mid-1980s, John Scarry devised a classification scheme for Fort Walton ceramics which encompasses the Bull Creek phase. Scarry's (1985) type/variety system builds upon the earlier Fort Walton typology (Willey 1949) and provides an extremely important tool for interpreting differences in ceramic assemblages within a tight temporal and regional framework. Scarry's scheme has been incorporated into our examination of the Bull Creek site pottery.
In our examination of pottery from Bull Creek, a variety of attributes will be examined which hopefully will provide a characterization of the ceramic assemblage. Throughout this report an attempt has been made to present examples of decorative styles and vessel forms found on the site. The following sections will include more quantified data. The sources of these data are tabulations from previous work, primarily housed in the National Park Service files and files of Frank Schnell, Jr., curated at the Columbus Museum, and a reexamination of the surviving collections from the Columbus Museum and University of Georgia.

While there was no formal report detailing the laboratory analysis conducted following the 1936-1937 excavations at Bull Creek, a very substantial body of data was recorded and tabulated and much of that paperwork has survived. At the regional WPA laboratory in Macon, detailed attribute analysis was conducted using the McKern System, a procedure which provided standardized techniques for examining all of the site collections excavated during the different WPA-era projects. A thick stack of paper now curated at the Columbus Museum contains coding and analysis sheets for 2240 sherds along with some summary sheets.
Copies of the coding instruction sheets used for Bull Creek have been transcribed and are included as Appendix B. The summary sheets are included where appropriate in the present chapter. Even with the coding sheets, some of the attributes used in the 1930s are difficult to understand and interpret. Some of the attribute terminology, for instance, is no longer used. Copies of the original summary sheets which tabulate the primary attributes of the site's pottery are shown in Figures 131 through 133. Keys have been added to explain the codes.

149

t;; The Bull Creek Site o

$TANlI'EO I'''S

~N~/Sf[L>

"

7 f . 7 1 /'IItTO'I-O~c.
A/No-}."6$.

,.;J-o

Figure 131. Copy of National Park Service summary sheet of 2240 sherds from the WPA excavations at Bull Creek; various sherd attributes.

Bull Creek Site Material Culture

Temper Size:

a) large b) medium c) small

Cross-sectional Appearance of Paste

2) medium fine 4) medium coarse 6) granular 8) contorted 9) compact

(Hardness - no key needed)

Surface Finish

1) very rough 2) rough 3) outer rim 4) glossy 5) polished 6) vermiculated

Execution

1) crude 2) fair 3) good 4) excellent 5) unobserved

(No key for plain through lip)

Figure 132. Copy of National Park Service summary sheet with counts for temper, paste and texture. 151

The Bull Creek Site

/f-/fi/ --3~-

',.': . ::i

~ . ~ . t:J /J~ / _ f) - ~ ../__ A ()

~-~'f~- -; i - -

i-V; tti! /off-:.II~i/. L!..../-.. 7. ;r~--" ., I ~ :'; r'f'f -.. ..-

,!,.---..

, . .-.-.. --- .... .--...i~'I:... ........ -

. ....

_ -:fJ -If tff' -' ~

1.1'

';I/"/1, '

~

.3 I~~

-

I -.........-'--' . .

.

!/fG}#:---I1J

qtt.-------.----..- -

I~" i,f/1:7'!,tJ....-...

. ...-....-----.--.--

I; -At1 -.-

~

?;1 X "
E::

~ '~~--~---':~iY-3~------;.-~-_-~-;--

53?'

: ... _

_{ .A f ,.
,1B.1M ,

If_ : :

/ ' /7'tJ $"...

h;P ~!f.J J~f

C,&.
~ ,~ ~.,_',_,._._.._._--~._._._-_._.._.
1_17-- ._--_._--

('$- 1f.!, 1/ 1 l 3 ; 1

.1' C! r '8.:: .i.*! l if i :3-... .,----. -.-----------------,

v_-i _ - if) l i E I - 5'7-i?" -.-.-..

:.. / /IA)lDL~.

._3 i .. j .._.)t2 ~.~.------- ~ .......i._:~1.

~1

- ~~._.7--L.._L.1_-H~-....-iji-------~----~-~--..--..j--------------.--I.--1.-~---r.. ---~-------.---+-.i:-t-+ -

-

,
i

"""1 ~-L_-.JLJ1J,:--.-----~:.._f--- '---' - .--------.J--.J..-. '."

.! / .. -A--uu-...Ji __. . _J i __.. ____... _ ... _.

! .1. ....-- " __.~

..1-_- A--~PA. ii'-- .: ... :? (.: ..... . .-.. -. - -.-......._: .. 1-..- .----.----.----~

~-E'%.=l=~~~~~JH -=::=-i=--:~~--~ -==---F,-------~

-- ---~;--'jJ~-~t--r--+. ----.---1 . .-------- .--T--~+-d

,

I

.

.,

.,

Key: Decorations: Part of Pot:

Undeciphered 1 sherd from undetermined portion of pot 2 rimsherd 3 rim and shoulder or neck 4 Rim and handle or lug 5 Rim and Body 10 Handle

Quality of Ware: 1 Excellent 2 Good 3 Inferior

Figure 133. Copy of National Park Service summary sheet of 2240 sherds from the 1936-1937 excavations which quantifies attributes of decoration, quality of ware and vessel portion.

152

According to available records, these sherds were apparently analyzed before July 15, 1936 (Walker 1989:15). This would mean that the sherds were recovered from the village area. As previously mentioned, a lack of a descriptive text accompanying the NPS summary sheets makes interpretation of some of the data rather difficult. Fortunately, a number of these pottery attributes were also tabulated by Schnell (1963). Some categories, including temper size and cross-sectional appearance of paste are not precisely defined and cannot be easily compared to other excavations.

Bull Creek Site Material Culture
A number of detailed drawings which illustrate sherds from the site were also generated as part of that WPA analysis process. Figures 134 and 135 illustrate selected examples of these drawings. The drawings are
num- labeled Bull Creek or abbreviated B.C., indi-
vidually numbered using a Park Service bering system, and initialled by the artist. The drawings that survive do not illustrate the full range of Bull Creek pottery, but appear to concentrate upon incised and punctate examples.

.lJt:lll t~eek
*S.2 9.

Bflfle,.e-e/c
;9f 6o~.

S,C, ~/ .
.509/

Figure 134. WPA drawings of Bull Creek phase plain and complicated stamped sherds. 153

The Bull Creek Site
"Bqll ere ek
#775-.
lJu/1 ("r_
./Yo _ 3 _
Co/_ Ga. .
.13. c _ /?8/.9
Figure 135. WPA drawings of incised and punctate sherds from Bull Creek (lower right is shell tempered). 154

Bull Creek Site Material Culture

Probably the most important collection from Bull Creek, in terms of quantifiable data, resulted from Schnell's excavations of 1959. That produced the first screened collection of material from the site (see Chapter 5). The following text and tables are taken from Schnell's Bull Creek excavation manuscript. Illustrations have been added to further define many of the attributes examined.

As may be seen in [Table 4], the basic analysis of the pottery recovered from the 195gexcavations was done from a purely modal standpoint. The following paragraph correlates these modal categories with ceramic types. All of the pottery presented in this table falls within the Lamar series as partially defined in the Southeastern Archaeological Conference Newsletter (I, 2 :4-6, 9-11 ). Some of the type names used here have not been officially designated, but have found their way into common usage ....

Grit Temper BCVL*

o

3

p-;;-

cm

The plain ware

was divided into three

major categories according

to tempering mode; grit, sand and shell [Figure 136].

Sand Temper

For purposes of present

commentary, it might be

mentioned that the grit tem-

pering category is the same

Shell Temper

as the Bull Creek variant of
Lamar [which designates a Figure 136. Enlarged photographs showing tempering agents found at Bull

paste that contains large

Creek (BCVL represents Bull Creek paste variant of Lamar).

particles of garnet bearing

grit extruding from the exterior and sometimes the interior]. The sand tempered category covers a gradient range from very similar to Bull Creek variant of Lamar to a very fine sand tempering. This very fine sand tempering

historic ware which he designated as Coweta Micaceous. The shell tempered category at the Bull Creek Site is primarily a partially leeched form, indicating some, but not considerable antiquity.

finds its analogy in Hurt's (1975) historic and proto-

155

The Bull Creek Site

The complicated stamped ware may be easily assignable to the type Lamar Complicated Stamped, with the exception that it has the Bull Creek variant of tempering. This exception also applies to check stamped, defined by W.H. Sears (1951 :32) as Mercier Check Stamped. The bold incised category falls in more of a coarse sand tempering than of the tempering characteristic of the Bull Creek variant of Lamar. This type is therefore named as Lamar Bold Incised, with no particular distinction being made. The zone incised category finds its stylistic origins in Fort Walton Zone Punctated, but here again, the tempering is characteristic of the Bull Creek variant of Lamar. The punctated category seems to be most closely related to an undescribed punctated type in the original Lamar series. The engraved category is so sparsely represented here that no attempt will be made to discuss it extensively. It will simply be stated that this one example of engraving was

found on a black burnished sherd and is very similar to earlier Mississippian engraved types except for the fact that it is sand tempered rather than shell tempered. Finally, the red painted category, though statistically insignificant, should be mentioned since it is frequently found as a very minor
type in many Lamar sites ... Figures 137 and 138 illustrate examples of the primary types of surface treatment.
In the original analysis, the material was divided stratigraphically to meet one of the original objectives of the 1959 study. This objective was to determine if any differences could be detected stylistically, modally or typologically. No significant distinctions were detected. The material has therefore, for purposes of presentation in this paper, been lumped together into categories only, with no stratigraphic differentiation made.

Table 4. Ceramic counts from Schnell's 1959 excavations.

Plain (undecorated) Grit tempered Sand tempered smoothed surface burnished surface Shell tempered Total Plain
Complicated Stamped Check Stamped
Total Stamped Incised Punctated
Bold Incised Zoned Punctated Punctated Engraved
Total Incised and Punctated Painted
Red Painted Total Painted
Total Sherds in Lamar Series Collection

2460
474 136
8 3078 4903
94 4997
150 99 4
1 254
7 7 8,336

79.92%
15.40% 4.42% 0.26% 100.0% 98.12% 1.88% 100.0%
59.06% 38.98%
1.57% 0.39% 100.0%
100.0% 100.0%

36.92% 59.95%
3.05% 0.08% 100.0%

156

Bull Creek Site Material Culture
o
ps;-
em
Figure 137. Examples of Lamar Complicated Stamped and Mercier Check Stamped (bottom right) sherds from Bull Creek (various collections). The upper left sherd shows typical overstamping.
157

The Bull Creek Site
em Figure 138. Typical examples of incised and punctated sherds from Bull Creek (various collections).
158

Bull Creek Site Material Culture

Schnell also examined the pottery from the 1950 investigations of Kelly.
The ceramic material recovered from Kelly's 1950 excavation [Table 5] was analyzed in order to compare this assemblage with other analyses made of Bull Creek Focus material ... , A total of 3225 sherds were analyzed from Kelly's excavation. This analysis summarization is a combination of all of the units excavated by Kelly. It might be mentioned here that a detailed analysis of the material recovered by Patterson and Lester
in 1936 has yet to be made (Schnell 1963).

Schnell's analysis of Kelly's material shows lower percentages of complicated stamped pottery compared to his 1959 work. Unquestionably, there is variation in percentages of wares from different collections. In part, this represents differing recovery techniques. However, a comparison is instructive. Figure 139 compares the frequencies of these wares from four investigations at Bull Creek (the 1993 SAS collections are included as the fourth collection).

Table 5. Analysis of sherds from Kelly's 1950 excavation (adapted from Schnell 1963).

Bull Creek Variant of Lamar* Plain

1362

42.2%

Bull Creek Variant of Lamar Curvilinear Complicated Stamped

1394

43.2%

Bull Creek Variant of Lamar Check Stamped (Mercier)

41

1.3%

Plain (moderate tempering)

235

7.3%

Incised

67

2.1%

Zone Punctate

53

1.6%

Plain Shell Tempered

13

0.4%

Plain (fine sand tempered

2

0.1%

Burnished Black Incised

4

0.1%

Bull Creek Variant of Lamar Plain Lug

5

0.2%

Bull Creek Variant of Lamar Plain Disc

4

0.1%

Bull Creek Variant of Lamar Complicated Stamped Disc

1

<0.1%

Red Filmed

3

0.1%

Burnished Black Plain

24

0.7%

Brushed (probable Chattahoochee Brushed)

13

0.4%

Negative Painted

1

<0.1%

Burnished Black with Rim Node

2

0.1 %

Simple Punctated

1

<0.1%

Total

3225

100%

*Bull Creek Variant of Lamar designated a type ofpaste which contains large particles of garnet bearing grit extruding from

the exterior (sometimes the interior).



159

The Bull Creek Site
Schnell's tabulations from his 1959 excavations have been used to define the ceramic composition of the Bull Creek phase, at least with respect to surface treatment. Schnell's excavations produced a predominance of complicated stamped wares
(> 50 percent) followed by plain,
incised/punctate and check stamped (Schnell 1990:67).

By accepting Schnell's (1963) assessment that no stratigraphic differences existed for the pottery recovered in the portion of the site investigated in 1959, we may accept Schnell's figures as the definitive characterization of Bull Creek phase pottery with respect to surface treatment. The small SAS collections show strong parallels in percentages to the 1959 excavations.

Comparison of Bull Creek Pottery From Four Excavations

60
50
40
30
20
10
0--'---
Complicated
Stamped

Plain

~ WPA (2090 Sherds)
D Kelly 1950 (3225Sherds)
Schnell 1959 (8336Sherds)
D SAS 1993 (322 Sherds)

Incised/
Punctate

Check Stamped

Figure 139. Comparison of pottery counts from four archeological investigations of the Bull Creek site (the WPA sherd counts exclude the modeled category).
160

Bull Creek Rim Attributes
Bull Creek vessels display a variety of rim forms and rim modifications typical of both Lamar and Fort Walton. The primary rim attributes of Bull Creek pottery have been

Bull Creek Site Material Culture
examined and quantified for the WPA-era excavations and Kelly's 1950 excavations. Figures 140 and 141 show examples of the more common rim forms from Bull Creek. Figure 142 is a copy of the NPS summary sheet of attributes for 280 rims.

B

c

o

3

&.-

i

em

A, folded, pinched; B, folded, cane punctate; C, Applied stip, pinched;

D, applied strip, notched; E-F, applied strip, noded

Figure 140. Examples of appliqued rims from the Bull Creek site.

161

The Bull Creek Site B E

o

3

&.-

em

A, folded; B, thickened, horizontally expanded and notched; C, noded; D-E, vertical lugs; F-G, notched; H, plain, tapered; I, horizontal lug, notched

Figure 141. Examples of additional rim forms from the Bull Creek site.

162

Bull Creek Site Material Culture

- _._~-'- ----_:..-.-----

! _".".. ~ ". ,.i1

"
i

Figure 142. Copy of National Park Service swnmary sheet of Bull Creek rim attributes.

Key:
Lip Shape Code Not Deciphered

The Bull Creek Site

The data presented in the National

cent), and check stamped (0.7 percent). This

Park Service's summary sheet is valuable for

will be examined again in this chapter. These

examining both rim form and vessel de-

figures suggest that the high percentage of

coration. The summary sheet provides counts

complicated stamped represented by individual

of shape and decoration for rims and lips.

sherd counts (approximately 60 percent in

Because the coding instructions (see Appendix

Schnell's 1959 excavations) may result from

B) are not compatible with the category of lip

less than thirty percent of the vessels produced

shape shown on the summary sheet, those

on the site.

tabulations have not been interpreted.

A detailed examination of Bull Creek

According to the summary sheet

rims was also conducted by Schnell (1963).

figures, the most common rim form curves

The results of Schnell's analysis follow.

outward and is decorated with an appliqued

strip. The rim decoration referred to as appliqued is identified on 37.1 percent of all rims. Appliqued rims are formed by adding a strip of clay at the rim or a slight distance below the rim. Generally, this strip is pinched.

To the present, only general statements have been made about the exact nature of Lamar pottery [at the Bull Creek site]. It was decided, therefore, in order to gain a more exact understanding of the nature of Lamar types, and of the Bull Creek Focus in particular, to make a

When the rim collar is wide and extends down

detailed rim analysis. This was done by dividing

from the rim, it is generally referred to as a folded pinched rim and represents a typical Lamar jar style. The decoration category "gridbarfI as used in the summary sheet appears to be the same as check stamped. The

the rims from the 1950 excavation by Kelly, a total of 278 rim sherds, into a number of categories according to rim form and to designate pottery type according to the body portion below the rim [Figure 18]. The following comparative data resulted [Table 6, Figure 144].

remaining ter-

minology appears to

be similar to that used today.
The category of pot decoration is important as it relates to other vessel attributes, discussed later. The summary sheet data may be used to approximate some vessel count traits. The figures comparing pot .decoration and rim decoration show a predominance of plain vessels (57.9 percent), followed by

FOlded,~~:: "i""'~"":""':'''~''~~~~~~~=O=38 26

Overhanging fold

13

Overhanging fold, thinDed 11"1"'1"1''''1''''';' 91

Outeurved lip

13

Thinned,outeurved lip

ThimIed, straight

Roundededge l~~~~~~ggg~~~ Sttaight, rouDded lip

28

Sttaightlipped

14

KeeIed-back Keeled-back, rolled lip

.. s

Keeled-bacl< DCId>ed Rolled lip

Distribution of Rim

Thicl=ed lip

Thickened 0U1Side lip

.s

PiDcbed ]".E,,,~~

Types for All Sherds Recoveredby Kelly

NOlcbed
Flat lip, notched ]z::::::::::S:~~~::J

Noded Roundedlip, notched LughaDdle J~~gg~~=1=1 ==~22

Luted

J~g~~ Lutednotched

6

Lutedpinched

LutedWldifferentiated

PuDctale

Reed punctale ScaUoped ..j:::'.:

-+-

...l-

--'--_ _- - ;

stamped (28.9 percent), incised or

o

10

20

30

40

Number

punctate (12.5 per- Figure 143. Graph showing rim varieties identified from a collection of 278 sherds

(graph produced from figures in Schnell 1963).

164

Bull Creek Site Material Culture

Table 6. Detailed rim analysis of 278 sherds from Kelly's 1950 excavations at Bull Creek.

yi

i

......i<.......

} .......
.... ......

iit . ttt i ......... ........ } ..

'<



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

..:':':

......


i ..
< .:.

.o.: :..,



<

c

.............

....... . .....

1 Folded

25

2 Folded, thinned

29

9

3 Overhanging fold

10

3

4 Overhanging fold, thinned

9

5 Outcurved lip

9

1

6 Thinned, outcurved lip

2

1

7 Thinned, straight

1

8 Rounded edge

23

2

9 Straight, rounded lip

16

1

10 Straight lipped

4

2

11 Keeled-back

1

12 Keeled-back, rolled lip

5

13 Keeled-back notched

4

14 Rolled lip

3

1

15 Thickened lip

1

16 Thickened outside lip

5

17 Pinched

2

3

18 Notched

2

19 Flat lip, notched

5

4

20 Rounded lip, notched

21 Noded

20

2

22 Lug handle

10

23 Luted

2

24 Luted notched

5

25 Luted pinched

8

26 Luted undifferentiated

1

27 Punctate

1

28 Reed punctate

1

29 Scalloped

1

Grand Total

204

30

2

1

4

9

2

4

4

1

3

4

3

1 1 1

25

15

1

26

38

13

9

13

3

1

29

18

14

1

5

4

4

1

5

3

9

2

16

3

22

11

3

6

8

1

1

1

1

4

278

165

The Bull Creek Site

30
25
8' 20
N II
6... 15
1l
~10
5

Rim Type Distribution for Plain Ceramics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 192021 2223 242526272829

':us
'l
II
6... 0)
,t:J
~o

Rim Type Distribution for Complicated Stamped Ceramics
1 Irapl;;! jl~ll71l~ llllrzJ lll~ ~1~ 11r:a 1111111rJ

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 2021 22 23 24 25 2627 28 29

Vl
J : 10
~ ,~, ,~, I : I I I ,Ill, ,

Rim Type Distribution for Incised Ceramics

I IIII I I

I I"',","', , , , I

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 2223 2425 2627 28 29

r J Vl

Rim Type Distribution for Zoned Punctate Ceramics

I

~ '~, : , , , ,", , , ,Bl,m, , I , I I ,",

,I , , , ,I ,

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 2425 2627 28 29

r:J

Rim Type Distribution for Check Stamped Ceramics

I

~ Z 0 rJ I I I I i i I I I I 1 I 1 I I 1 Tj--'1--'1--'1--'1--'1"""'1"""'1"""'1"""'1-r-j-r-j-
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 2627 28 29

Figure 144. Series of graphs comparing rim type and vessel treatment for sherds examined by Schnell (see Table 6 for list of rim types corresponding to the numbers on the graphs).

166

Bull Creek Site Material Culture

Schnell examined rim types in great

Oertel 1977:39). This comparison seems

detail, but his manuscript does not contain

inconsistent with Schnell's dating of Bull

definitions for the rim types. While most of

Creek as an earlier phase and may indicate that

the rim types are easily interpreted, other cate-

certain concepts regarding rim fold width

gories cannot be accurately understood without

might stand further scrutiny, especially with

formal or measured descriptions. There is

respect to Fort Walton influences.

some discrepancy with respect to counts for

ceramic types or pot types. Schnell's counts

A second set of measurements was

indicate that 204 of 278 sherds represent plain

taken while examining all available collections

bodied vessels. This percentage of 73.4

for vessel form data. Sherds of sufficient size

percent is substantially higher than the 57.9

to allow measurement of rim diameter, deter-

percent noted on the NPS summary sheets.

mine vessel shape, and determine vessel de-

This may represent differences in the col-

coration were also measured for width of

lections, differences in analysis techniques, or

folded rims when present. Most of the vessel

simply a typographic error in that manuscript.

form data will be presented in a later portion

A quick reexamination of sherds in Kelly's

of this chapter; however, the rim fold width is

collections, now curated at the University of

pertinent at this point. Table 7 compares ves-

Georgia, produced a figure of approximately

sel rim diameter to the width of folded pinched

53 percent plain vessels, which is more in line

rims for fifty sherds. All sherds represent

with the NPS figures.

typical Lamar constricted rim jars. These data

show that while the smaller jars do display re-

The width of rim folds (applied rim

latively smaller rim folds, there is not a direct

strips) is today considered a critical analytical

correlation indicating increasingly larger rims

measurement with respect to Lamar pottery.

with increasingly larger vessel diameters.

The generally accepted proposal is that these

Beginning at approximately 32 em diameter,

rims get wider through time. Hally's exa-

the rim folds tend to cluster around 21 mm in

mination of folded pinched rims from Park's

width.

Mound led him to conclude that the rims

were sufficiently wide to indicate a late date

(late sixteenth or early seventeenth century) Table 7. Comparison of rim diameter and width of

for the occupation at that site (Hally and

folded pinched rim fold (sample size: 50 rim sherds)

Oertel 1977:39). During the course of the

present project, measurements were taken

Size Range Average Size

on folded pinched rims from various col-

Rim

Width of

Width of

lections. A total of 112 folded pinched rims Diameter Count Rim Fold

Rim Fold

from the NPS collection were measured. That collection also included 59 rims with distinctly applied pinched strips which were generally placed about one em below the lip. This is mentioned because there is

14 em 18 em 20 em 22 em 24 em

1

13 mm

1

16 mm

1

16 mm

2

12-16 mm

2

16-18 mm

13 mm 16 mm 16 mm 14 mm 17 mm

some subjectivity involved in distinguishing sherds with smoothing of the upper portion of an applied strip from sherds that are generally described as folded pinched rims. The folded pinched rims in this collection ranged in width from 11 to 28 mm and

26 em 28 em 30 em 32 em 34 em 36 em 38 em

1

17 mm

4

12-21 mm

2

16-18 mm

8

16-27 mm

8

14-24 mm

8

15-26 mm

3

14-26 mm

17 mm 17 mm 17 mm 21 mm 18 mm 21 mm 21 mm

averaged 20.3 mm, wider than the 18 mm average from Park's mound (Hally and

40 em

5

15-25 mm

21 mm

42 em

3

18-25 mm

21 mm

44 em

1

20mm

20mm

I

--------------------------------------------------~

167

The Bull Creek Site

Two additional attributes of rim decoration, handles and adornos, are inadequately tabulated by the researchers of the Bull Creek site. Chase (1957) noted that handles were not found on Lamar Complicated Stamped vessels at Bull Creek and there are no references in his manuscript to handles on other pottery types. Chase does note that rim adornos, usually lugs, nodes and more rarely effigies, usually of birds, occur on Fort Walton-style vessels and a type of burnished black slip ware (Chase 1957). Schnell's summary tables list 11 lugs while the National Park Service summaries reference six handlesllugs. The handles appear to be vertically and horizontally placed lug handles. An examination of available collections produced several examples of lug handles (see Figure 141, second row) and two or three flattened fragments of pottery which appear to be portions of small strap or loop handles. Large Lake Jackson-style loop handles which appear so frequently on Mississippian sites in the region do not appear to occur at the Bull Creek site. At the Bull Creek site, lug handles appeared to be limited to plain and incised/punctate bowls.

Very remarkable rimsherd, in the form of a woman, with her arms raised to the back of her head. Facial characteristics more or less of conventionalized form. Found 60" below the surface, 27" below the occupation level. 24" N.
6" E. Sta. 4-15L13 (Lester 1937:91).
The old National Park Service photograph fails to do justice to the exceptional features lauded by the preceding description. A search of the NPS collections suggests the artifact has been lost so it cannot be rephotographed.
At Bull Creek, effigy rim adornos appear to be associated with Fort Walton-style bowls. Similar animal style adornos are found consistently on Mississippian sites and have been reported from other late Fort Walton and Bull Creek-related sites in the region (Willey 1949:504; Broyles 1971).

The production of ceramic effigy figures is an important aspect of the Bull Creek pottery complex. This is most apparent in the effigy vessels and pipes found with burials. Effigies are also common on vessels found throughout the village area.

Figures 145 and 146 illustrate examples of effigy adornos from the Bull Creek site. Most of the illustrated effigy objects are animal figures of some type. The object illustrated in Figure 145 has been identified as a human representation. The effigy adorno was recovered from the midden deposits in the cemetery during the WPA excavations and numbered as Find 69. NPS notations describe the object as follows.

o

3

e

cm

Figure 145. Photograph of human effigy adorno (copy of photograph from National Park Service files).

168

Bull Creek Site Material Culture

o

3

em

Figure 146. Animal effigy rim adornos from Bull Creek site collections (size of lower right is unknown).

Vessel Form Attributes
During the course of the present study all available collections of pottery from Bull Creek were examined. As one facet of that examination, all large rim sherds were pulled and selected sherds were analyzed in detail to procure vessel form data that would allow further comparison to other Mississippian site analyses (Figure 147). An attempt was made

to mend vessels and count only one rim sherd per individual vessel. Sherds were included in this analysis if they were sufficiently large to determine rim diameter, rim type, body decoration and vessel profile. Generally, a minimum rim size greater than 4 em was required. Vessel form represents one aspect not sufficiently examined by previous researchers at Bull Creek. Our examination produced data for 219 vessels (Table 8).

169

The Bull Creek Site
Fort Walton Vessels (Scarry 1985)

Coastal Lamar (Braley 1990)

0 CJ C)

A

B

C

C/ CJ C7

0

E

F

0 00

G

H

I

000

,J

I(

~

A) simple bowl; B) carinated bowl; C) cazuela bowl; D) simple restricted bowl; E) collared bowl; F) flaring rim bowl;
G) collared jar; B) evened rim jar; I) evened rim conical jar; J) restricted jar; K) beaker; L) bottle

'C:.7 C )

A

B

00

C

0

00



F

u <::r

G

H

A) simple open bowl; B) carinated bowl; C) constricted rim bowl; D) flaring rim jar;
E) carinated jar; F) flaring rim pot; G) flaring rim plate; B) recessed neck jar

North Georgia Barnett Phase Lamar Vessels (Hally 1986)

c=J 0 C7 t=J Q

A

B

C

0

E

0 Q U0

F

G

H

I

A-B) rounded bowl; C) flaring rim bowl; D) "gravy boat" bowl; E) carinated bowl; F) pinched rim jar; G) Mississippian jar; B) carinated jar; I) bottle

Figure 147. Examples of vessel forms identified from various Mississippian localities.

170

Bull Creek Site Material Culture

Table 8. Vessel forms and counts from Bull Creek.

Vessel Form
Everted Rim Jar Carinated Bowl Collared Globular Bowl Simple Constricted Rim Bowl Open Bowl Effigy Bottle (Dog Pot)

Count
83 36
2 80 15 3

moderately everted to conspicuously flaring
rims (Figure 148). Within the collection, rim
modifications include folded pinched rims (N = 50), applied strip below the rim that is noded (N = 10) or pinched (N = 5), indi-
vidual nodes (N = 6) or lugs (N = 1) at or
below the rim. A small number of jars have unmodified rims (N = 11). Body decoration consists of complicated stamped (N = 65),
plain (N = 17), and check stamped (N = 1).

Total

219

Of the 219 vessels, only four (the three dog pots and one small bowl) are burial objects. The remainder were recovered from the general midden or trash pits and are considered domestic or utilitarian vessels. Each vessel form will be described separately. Measured attributes of individual sherds are presented in Appendix C. That data will be summarized in this chapter.

Everted Rim Jars (N = 83) generally correspond to Hally's pinched rim jar (see Figure 147) and represent an extremely common vessel type on Lamar sites. Our category applies to jars with welldefined neck constrictions and

o

6

,......

i

em

Figure 148. Partially reconstructed Lamar Complicated Stamped everted rim vessel with appliqued pinched rim strip (UGA collection, Bull Creek site provenience unknown).

171

The Bull Creek Site
Within this collection jars account for approximately two-thirds of the vessels decorated with complicated stamping from Bull Creek. The remaining vessels with complicated stamping are simple bowls.
Measurement of rim orifice diameter is the primary means of comparing jar size from different sites. Jars generally reach their greatest diameter a few centimeters below the neck constriction. Relatively few of the Bull Creek rims were of sufficient size to determine maximum vessel diameter but measurements of six sherds showed the increase was only 2 to 4 em greater than rim diameter.
Rim diameters for jars range from 14 to 44 em, Across this spread, 36.1 percent are small to medium jars of 30 em or less and the remaining 63.9 percent are medium to large jars in the 32 to 44 em range. A pronounced peak exists between 32 and 36 em accounting

for 41.0 percent of all jars. Jars account for 37.9 percent of the total vessels in this collection. Figure 149 shows the range of rim sizes for the Bull Creek jars.
Large jars are considered storage vessels, with a rim shape that facilitates pouring and attachment of a protective covering (Shapiro 1990:155). Research by Hally (1986:285-286) indicates the largest jars (40 em and larger) were used for long term storage while medium sized jars were used as general purpose cooking vessels. According to the research of Shapiro the proportions of larger storage jars (> 30 em in rim diameter) within the overall assemblage is an important indicator of the intensity of site occupation. The relative counts for jars compared to other vessel types at Bull Creek and the relatively high frequency of larger jars are comparable to permanently occupied late Mississippian sites (Hally 1986; Shapiro 1990).

EVERTED RIM JARS
12
10
8
~ 6
U 4
2
0'----10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 RIM DIAMETER (em)
Figure 149. Graph showing the range of rim sizes for everted rim jars from Bull Creek. 172

Carinated Bowls (N = 36) are defined as constricted rim vessels with distinct shoulder break, which sometimes forms a well defined projecting keel or carination. Similar vessels from contemporary sites are generally flat-bottomed. The typical Lamar cazuela

Bull Creek Site Material Culture
bowl is included in this category. At Bull Creek, carinated bowls are typically decorated with Fort Walton incised and zone punctate patterns (Figure 150). The lower restored bowl has cane punctation at the shoulder that is not typical of the Bull Creek site.

o
s-.

3

em

Figure 150. Examples of Fort Walton carinated bowls from Bull Creek. 173

The Bull Creek Site
All but one of the carinated bowls in this collection are decorated by incising (N = 11) or a combination of incising and zone punctation (N = 24). The style of the decoration is basically Fort Walton. The presence of Lamar bold incising is questionable. Rim shapes vary from rounded to flattened. Modification includes notching (N = 16) and incising (N = 3). Incised rims are from vessels where the incised body decoration extends to the lip of the rim. The rims of seventeen vessels are not modified. At Bull Creek, rim notching occurs on carinated bowls and simple constricted rim bowls but relative proportions are substantially higher in this vessel type.
Figure 151 shows the distribution of rim diameters for carinated bowls from this collection. The vessel occurs in sizes from 14 to 40 em but there are some indications that some sizes were more common. Almost twenty percent of the vessels measure 24 em in rim diameter. A second peak is evident at 30 to 32 em (30.6 percent of all carinated bowls).
The size distributions of carinated bowls at Bull Creek is similar to that shown

for the Late Lamar Barnett Phase (Hally 1986:275). Hally suggests that carinated bowls were used to cook and serve food that was of liquid consistency and would require a minimum of cooking. Small and large sizes of carinated bowls reflected the quantities of food prepared (Hally 1986:288-289). This vessel type was widely distributed in the midden and trash pits at the Bull Creek site. At least some examples of the vessel type were found on the floor of Lester's second structure.
Collared Globular Bowls (N = 2) are represented by two examples from Bull Creek. The most complete example was found in a trash pit near Lester's Structure 1 (see Figure 34). The vessel form has been described as a deep globular bowl with incurving walls which constrict at the base of a short neck or collar (Willey 1949:498). The previously illustrated vessel is plain with a rim diameter of 26 em. The rim is unmodified. The second vessel is check stamped with a slightly rolled rim and a rim diameter of 36 em. These bowls may have been used as cooking or serving vessels.

CARINATED BOWLS

6
s~ 4
U2
o
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 RIM DIAMETER (em)
Figure 151. Graph showing the range of rim sizes for carinated bowls from Bull Creek. 174

Simple Rounded Bowls (N = 80) occur with essentially the same frequency as jars. The category includes any bowl with a constricted orifice. In some cases the rim is only slightly incurving. This category is

Bull Creek Site Material Culture
distinguished from other simple bowls that slope outward at the orifice. The category combines vessel types similar to Scarry's simple restricted bowl and simple bowl (see Figure 147). Figure 152 shows examples of the range of profile variability.

o

3

I"l~-~--il-_~~

em

,.;:-~.::::.==~::~:.~:~~:::f:~;;:, ...__..,....__...~...,...._.....,,.,.'".:':-::... ;7..

Figure 152. Examples of simple rounded bowls with slight to moderate rim constrictions. 175

The Bull Creek Site
Simple rounded bowls were decorated
with complicated stamping (N = 34), incising
(N = 11), a combination of incising and zone punctation (N = 2) and check stamping (N = 1). The remaining thirty-two bowls were undecorated. Within this collection sixty percent of all vessels are decorated. The percentage of Lamar complicated stamping (42.5 percent) is second only to jars at Bull Creek.
Rim types are highly variable in this vessel category. Fourteen different rim types
were identified, including unmodified (N =
33), notched (N = 22), pinched (N = 2), noded (N = 2), rolled (N = 3), incised (N = 1), applied notched strip (N = 2), applied noded strip (N = 1) folded (N = 3), folded
and pinched (N = 3), folded and notched (N
= 3), and thickened (N = 2). The collection
also contains one T-rim and two rims with effigy adornos.
Rim diameters range from 10 to 42 em with two-thirds being greater than 26 em

(Figure 153). Within this collection, sixty percent of the vessels fall within the range of 28 to 36 em.
Hally's work with Barnett phase vessels suggests that large rounded bowls, which equate in size to the majority of Bull Creek vessels, were used primarily for preparing and serving large quantities of viscous and solid foods but only minimally for heating these foods. Smaller bowls were used for similar functions but in limited quantities (1986:289).
Simple Open Bowls (N = 15) refers to vessels for which the widest part of the bowl occurs at the rim orifice. These bowls differ from the previous category primarily in this one attribute. Figure 154 illustrates a typical open bowl. A restored example from the WPA excavations was shown previously in Figure 152).

SIMPLE ROUNDED BOWLS

10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 RIM DIAMETER (em)
Figure 153. Graph illustrating range of rim diameters for simple rounded bowls from Bull Creek. 176

Bull Creek Site Material Culture

"

-. ~

_- " , ......

...... ~ "

'~- ......... ----------- ~~-

o,........

3
;

em

Figure 154. Example of a simple open bowl from Bull Creek.

Body decoration for open bowls includes complicated stamping (N = 5),
incising (N = 1), and combined incising and zone punctation (N = 1). The remaining eight
bowls are plain. Rims are generally unmodified (N = 6), or thickened (N = 3). Variations include thickened and L-shaped (N = 2), and lug handles (N = 1). One vessel contains incising extending to the lip and one vessel is notched. One vessel contains a simple incised design on the interior lip.
Rim diameters are shown in Figure 150. The range of 10 to 42 cm is similar to other vessel types from the site but unlike the other types there are indications of size groupings. The smaller sized vessels, which range in size from 10 to 24 em include one small vessel from Burial 18, which appears to be a specially produced mortuary vessel. This small bowl, only 10 cm in diameter, is crudely

decorated with a Fort Walton zone punctate design (see Figure 59). That bowl's combination of diminutive size, sloppy decoration and even vessel shape is unique in the site collection. A second small bowl was also recovered from the cemetery, but that vessel has been lost and was not adequately described in the field notes. The small bowls include three plain, two incised, and three complicated stamped vessels. Rims are variable but primarily unmodified (N = 4). The larger bowls are either plain (N = 4) or complicated
stamped (N = 2) with thickened or L-shaped
(N = 3), plain (N = 2), or notched (N = 1)
rims. The one isolated vessel at 28 em is a slightly flaring rim bowl with interior lip incising. Aside from mortuary use for the one example, the differences in sizes may reflect uses similar to large and small varieties of simple rounded bowls.

177

The Bull Creek Site

S~LE OPEN BOVVLS

4
~ 2
o

10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 RIM DIAMETER (em)

Figure 155. Graph illustrating range of rim diameters for simple open bowls from Bull Creek.

Bottles from the Bull Creek site consist entirely of mortuary ves-
sels. Water bottles similar to the category shown by Scarry (see Figure 147) should be expected at the site. Perhaps this omission is the result of sampling error or misidentification of vessel types.

Table 9. Measurements of effigy bottles from Bull Creek.

Vessel Burial Height Length Width

Orifice

Number No. (em) (em) (em) Diameter (em)

V.1I7 3

26.5 30.0 19.0

9.5

V.1I8 7

27.0 27.5 19.5

9.5

V.1I9 16

32.0 28.0 18.0

8.0

This category includes three bottles commonly known as the Bull Creek cemetery dog pots. Figure 156 shows the three vessels and three examples of similar vessels from other sites in the region. The three vessels from Bull Creek and the one vessel from Neisler Mound are considered the only known examples of this one type of negative painted dog effigy vessel (Schnell 1990:67). Selected measurements for the three vessels are shown in Table 9. The vessels will be examined further in the following section which deals with examination of Bull Creek pottery types.

178

BULL CREEK BURIAL 3

BULL CREEK BURIAL 7

Bull Creek Site Material Culture
BULL CREEK BURIAL 16

NEISLER

CEMOCHECHOBEE BELL FIELD

o

3

em

Figure 156. Illustrations of effigy bottles from the Bull Creek site with comparative examples from other western Georgia:" sites.

The Bull Creek Site
Summary of Pottery Types
The pottery from the Bull Creek site might be viewed as something of a hybrid of ceramic styles with some similarities to Lamar and some similarities to Fort Walton. The pottery does not always fit snugly into established type descriptions. Schnell's term Bull Creek variant may be as appropriate as any means for a designation of the pottery series. Basically, this means that vessel types and decorative styles typical of either Lamar or Fort Walton occur on the site, but these vessels are tempered with a distinctive, gametrich grit. The size of the temper varies from 0.7 to 3.0 mm (average 1.4 mm). One problem with this is that we do not know how localized this use of tempering agent is and whether it reflects cultural distinctions. While heavy grit tempering is generally characteristic of late Lamar in western Georgia, the distinctive nature of the site's temper is quite probably a localized phenomenon reflecting the local clay and sand resources.
The recognition of a distinctive Bull Creek variant of temper does not mean that every vessel on the site was tempered the same. At the Bull Creek site, some vessels of all pottery types are tempered with finer grit or sand. This applies most strongly to the incised wares. Still, there needs to be some means to improve upon the use of purely descriptive terminology for analyzing the pottery from the site. As expected, our examination of vessel form clearly indicated that certain surface treatments (decorations) were more common on certain vessel types. This is consistent with other research where the combined attributes of decoration and vessel form have been used successfully in the examination of pottery types (Shnell et. al 1981; Scarry 1985).
Schnell's 1963 manuscript references an appendix that contains formal descriptions of the pottery types found at Bull Creek. Unfortunately, this appendix could not be found
180

and we are limited to the more general characterizations found within the manuscripts of Schnell and Chase. In the following summary, descriptive categories used for Bull Creek pottery will be examined more closely within the context of formally defined pottery types. For the most part, the previously cited descriptions by Chase and Schnell have adequately defined these pottery types, particularly when referenced to the original type descriptions. Lacking in the descriptions of Chase and Schnell are adequate comparisons to Fort Walton ceramics. For this reason, the work of Scarry (1985) will be incorporated into these descriptions.
Plain wares from the Bull Creek site may be considered variants of Lamar Plain or Lake Jackson Plain (Willey 1949:458). As previously noted, Schnell (1963) divided plain wares from Bull Creek into three categories according to tempering. These categories consisted of grit (79.9 percent), sand (19.8 percent) and shell (0.2 percent). Schnell implied that the heavy grit tempered category should be referred to as Bull Creek variant of Lamar. Schnell also noted that the finest sand tempered was similar to historic and protohistoric ware designated as Coweta Micaceous (Hurt 1975). Since we now know that the Bull Creek site does contain an historic Creek component, the association of this pottery type with the Bull Creek phase might be questioned. Schnell noted that the few shell tempered sherds were partially leached, indicating some, but not considerable antiquity (Schnell 1963). There is the possibility that the shell tempered sherds postdate the Bull Creek phase (Abercrombie phase).
There are several characteristics associated with rim modifications at Bull Creek which are common on Lake Jackson Plain vessels but uncommon on Lamar vessels. Rim notching and applied rim strips that are either pinched or noded are particularly common on Lake Jackson Plain vessels (Willey 1949:459).

Bull Creek Site Material Culture

Scarry applies the type Lake Jackson Plain varieties Ingram and Tallahassee to the Bull Creek phase (Scarry 1985:221). Variety Tallahassee contains small amounts of grog temper but has not been recognized in any examination of pottery from the Bull Creek site. Lake Jackson Plain variety Ingram, which does apply, is described as a variety with coarse grit and smoothed surfaces, common to the Rood and Bull Creek phases and in the Appalachicola Valley, the Yon phase (Scarry 1985:221).

Complicated Stamped Wares found on Bull Creek phase sites have been universally described as Lamar Complicated Stamped. This absence of complexity results from acceptance of the proposition that, by definition, any poorly executed complicated stamping should be called Lamar. Schnell's (1963) observation was simply that the complicated stamped ware may be easily assignable to the type Lamar Complicated Stamped, with the exception that it has the Bull Creek variant of tempering (Schnell 1963).

Scarry's variety Ingram is based in part upon the type Ingram Plain which was defined at the Cemochechobee site in Clay County (Schnell et al. 1981:185). The definition for Ingram Plain suggested synonymy with the types Lake Jackson Plain, Pinellas Plain, Coweta Micaceous Utility Ware and a relationship to Lamar Plain. Ingram Plain vessels were primarily open bowls often exhibiting notched lips and sometimes exhibiting applied notched strips. The type was considered "the apparent predecessor of Lamar Plain as it is represented in the Lower Chattahoochee Valley" and further stated that with the addition of pinched folds and fillets, the type probably developed into Lamar Plain of the succeeding Bull Creek phase, but distinguishing the two types is difficult (Schnell et al. 1981:185-188).

Scarry examined complicated stamping more critically, with the caveat that his Lamar Complicated Stamped pottery includes all coarsely tempered complicated stamped vessels found in the Fort Walton area (Figure 157). Scarry notes that Lamar Complicated Stamped pottery probably developed outside the Fort Walton area in the South Appalachian Mississippian style area and was introduced after the Rood phase (Scarry 1985:221-222).

Based upon the results of previous research we would appear to have a choice of what to call the plain ware from Bull Creek. If the site is viewed from a Fort Walton perspective, the type Lake Jackson Plain variety Ingram is appropriate. If the site is viewed from a Lamar perspective, the plain ware may be called Lamar Plain. With respect to vessel form, the Bull Creek site plain wares are associated primarily with simple bowls, accounting for 41.1 percent of those types, and to a lesser extent with jars (20.5 percent).

Figure 157. Examples of motifs for Lamar Complicated Stamped var. Bull Creek (top) and var. Early (from Scarry 1985:223).

181

The Bull Creek Site

Scarry defined two varie-

ties of Lamar Complicated

Stamped that should apply to the

Bull Creek site (see Figure 157).

Lamar Complicated Stamped vari-

ety Bull Creek is a coarse sand-

tempered ware with figure-eight

designs. The type site was not

Bull Creek but site 9Ca51 (Broyles

1962), which Schnell now con-

siders to date to the subsequent

Stewart phase (Schnell and Wright

1993:21). Scarry also defined

Variety Early for pottery con-

taining motifs consisting of sets of

concentric circles, the centers of

which may contain raised dots or

crosses. Variety Early was con-

sidered the dominant type of the

Yon phase in the Apalachicola

Valley and present in the Bull

Creek phase.
At the Bull Creek site, Lamar Complicated Stamping was

o

3

e-- -

em

found on 78.3 percent of the jars Figure 158. Example of coarse grit tempered, bold check stamped

and 41.1 percent of the simple

pottery from the Bull Creek site.

bowls present in the vessel form

analysis collection of 219 rims. Complicated

At Bull Creek, check stamped designs

stamping was not associated with carinated

were identified on three vessel types, jars, col-

bowls, collared bowls or bottles.

lared bowls, and simple bowls. One example

of each was found in a collection of 219 rims

Check Stamping at Bull Creek is

examined for vessel form data.

generally referred to as Mercier Check

Stamped based upon similarity to material

Incised and Zone Punctate Wares

from Kolomoki (Sears 1951). Schnell noted

represent perhaps the most perplexing pottery

that Mercier Check Stamped at 9Me 1 should

styles found at Bull Creek. Various

be viewed as a Bull Creek variant because of

researchers have referred to the pottery as

its distinctive grit tempering (Figure 158).

either Lamar Bold Incised or Fort Walton

Scarry views the pottery as a variant of Leon

Zone Punctate. Schnell (1963) suggested that

Check Stamped, which includes all coarsely

because bold incising at the site occurred more

tempered and boldly check stamped ceramics

with a coarse sand tempering than with the

in the Fort Walton area (Scarry 1985:225).

temper characteristic of the Bull Creek variant

Scarry prefers Leon Check Stamped variety

of Lamar, the name Lamar Bold Incised

Mercier as a name for the bold check stamped

should be used.

pottery (often with checks up to 1 em) of the

Bull Creek and Yon phases.

182

Schnell felt the zone incised and punctate sherds were variants of Fort Walton Zone Punctated, with tempering characteristic of the Bull Creek variant of Lamar. The punctated wares were viewed as more closely related to an undescribed punctated type in the original Lamar series (Schnell 1963). During our examination of available collections we found these distinctions difficult to replicate.
Chase's efforts to distinguish the two styles were rather vague. Chase identified Lamar Bold Incised by deeply incised, usually curvate lines, sometimes consisting of only two or three parallel lines circumscribing the vessel or with interlocking loops. Chase included a variant of the Fort Walton Punctate within Lamar Bold Incised which for most sherds could not be distinguished from the classical Fort Walton punctate type (Chase 1957).
Hally found that a distinguishing factor of Lamar Bold Incising at Park Mound was simply a greater width of Lamar incised lines compared to Fort Walton incising (Hally and Oertel 1977:45). Using the criteria of greater line width Hally was able to identify only 14 Lamar Bold Incised sherds from a collection of 148 incised sherds from Park Mound (Hally and Oertel 1977:45).
While the more southerly located Bull Creek site might be expected to produce even fewer typical north Georgia Lamar incised sherds, an effort was made to test Hally's conclusion that Lamar Bold Incised has measurably wider lines than other types of

Bull Creek Site Material Culture
mcismg. Measurements were taken of incising on 61 sherds from the NPS collection. Line widths for 34 incised and punctate sherds which appeared to be Fort Walton styles ranged from 0.7 to 3.0 rnm (average 1.7 mm). . Line widths for 27 incised sherds which lacked punctations ranged from 0.5 to 3.0 mm (average 1.8 rnm). The 27 incised sherds were then sorted, somewhat subjectively, into Lamar-like and Fort Walton-like designs.
Sherds identified as Lamar-like (N = 16)
averaged 2.25rnm wide while the Fort Walton-like sherds averaged only 1.2 rnm wide. This does indicate some difference may exist within the collections, but a substantially larger sample of sherds should be measured before firm conclusions are reached.
The same collection of 61 sherds were examined to address Schnell's assertion that Lamar Bold Incised sherds at the Bull Creek site differed in temper. The sherds were sorted into three temper types. While our sorting criteria probably differ somewhat from Schnell's, in this one small sample, there does appear to be a predominance of coarse sand and fine sand tempering for the incised pottery. However, this difference appears to relate to both zone punctate (probable Fort Walton) and incised (possible Lamar) pottery. A low proportion of Bull Creek variant tempering in the zone punctate pottery would seem at odds with Schnell's implication that most sherds contained coarse temper (see page 156). The possibility exists that apparent tempering differences reflect, to some degree, variation which exists in different collections from various areas of the site.

Table 10. Comparison of temper for zone punctate and incised pottery from Bull Creek
~%'iI

Incised (possible Lamar Incised)

27

Zone Punctate (probable Fort Walton) 34

18.5% 35.3%

44.5% 20.6%

37.0% 44.1 %

100% 100%

183

The Bull Creek Site

The incised motifs associated with the Bull Creek and Stewart phases are clearly different from "typical" Lamar assemblages. These motif differences are obvious when sites from different areas are compared. Figure 159 illustrates the differences in the most common incised motifs from three phases. These illustrations are based upon Hally's research (Hally 1994: 153), but similar work has been conducted by other researchers (Snow 1990; Braley et al. 1986) and found to be an extremely valuable research tool for characterizing local phase assemblages. Importantly, Hally found that within each phase certain motifs occur with greater frequency. This is especially interesting for the Bull Creek phase because Hally notes that one motif (Figure 159, third row, left) occurs on more than fifty percent of the identifiable sherds in his sample (Hally 1994:154). In light of the fact that illustrated published data available to Hally comes primarily from two sites (Park Mound and 9Cy51), which Schnell now believes postdates the occupation of the Bull Creek site, Hally's observations are particularly important for comparing and contrasting incised motifs of the Bull Creek phase and Stewart phase.

@jj)j;
Barnett Phase
~~
(iW(I))

The identification of Lamar and Fort Walton design motifs is only one way of defining types of incised pottery. Other researchers working in the Chattahoochee valley have developed typologies based at least in part on vessel form (Schnell et al. 1981 and Scarry 1985). Basically, these typological schemes provide a means of extending Fort Walton type descriptions to the Bull Creek area allowing greater interpretation of various incised motifs.

Tugalo Phase

Scarry's type-variety approach to the incised wares of the region provides a useful tool for interpreting the variety of designs found at Bull Creek. Table 11 presents Scarry's breakdown of incised wares which could occur at the Bull Creek site. Figure 160 illustrates examples of each.

Bull Creek Phase
Figure 159. Comparison of common Lamar Incised designs from three Lamar phases (redrawn from Hally 1994:153).

184

Bull Creek Site Material Culture

Table 11. Presentation of Scarry's (1985) incised types and varieties of the Bull Creek phase.

IType

.... .'.'.".1. ,,iyarietY tl

Ilf:scriPFi()ri> ' .. -,- -"".",.".

1

Lamar Bold Incised

t
{

>>. .

it ....'

y{

>

Subsumes all vessels tempered with coarse sand or grit that are decorated with a band of broad, deeply incised lines on the upper portion of vessel (carinated bowls and cazuela bowls). Motifs are continuous bands of curvilinear scrolls or lines of pendant loops composed of 5 to 15 lines (Scarry 1985:221).

Fort Walton Incised

ii>
>...........'...'.. ,.,.....

, Subsumes all ceramics in the Fort Walton area with a sandy
y paste with decorations formed by incised lines with
punctation filled zones or occur on punctated backgrounds.

Fort Walton Incised

Fort Walton

Characteristic design is a band of running scrolls (both curvilinear and rectilinear) superimposed on a band of punctations. Vessel forms include cazuela bowls, beakers and bottles. Late but widespread variety (late prehistoric and protohistoric (Scarry 1985:215).

Fort Walton Incised

Cayson

Characteristic design is a band containing simple incised rectilinear step figures that separate alternating plain and punctation filled zones. Vessel form is carinated bowl (Scarry 1985:215).

Fort Walton Incised

Englewood

Characteristic design formed by narrow, punctation-filled bands that form abstract rectilinear figures including complex interlocking maze forms, diamonds, triangles, zigzag bands or connected chevrons. Found in Rood and Lake Jackson phase context but most common in Safety Harbor (Scarry 1985:219).

Fort Walton Incised

Safety Harbor

Characteristic design is curvilinear equivalent of variety Englewood. Vessel forms include beakers, bottles and bowls. Most common in Safety Harbor area (Scarry 1985:219).

Fort Walton Incised

Sneads

Characteristic design is curvilinear equivalent of variety Cayson (Scarry 1985:219).

......',...
Point Washington Incised

..>> . ,.. > ,>

Subsumes all ceramics from Fort Walton area with sandy paste decorated with medium curvilinear incised lines.

Point Washington Incised Point Washington

Characteristic motif consists of a series of horizontal running or interlocking scrolls. Vessel forms include carinated bowls, cazuela bowls, bottles and beakers. Late variety found in Bull Creek and equivalent phases (Scarry 1985:230).

Point Washington Incised Griffith

Characteristic designs are a variety of representational forms, many of which are part of Southern Cult iconography. Vessel forms include carinated and cazuela bowls. Chronologically late, found in Bull Creek and protohistoric sites (Scarry 1985:230).

185

The Bull Creek Site

A

E

..... '. .' .

. ' , ' ,

.. .. ',: :.':. '.' '"
. " , .', .. . . ; . ' " ,

..' :.'

.

.

F
A, Lamar Bold Incised; B, Fort Walton Incised var. Fort Walton; C, Fort Walton Incised var. Cason; D, Fort Walton Incised var. Englewood; E, Fort Walton Incised var. Safety Harbor; F, Fort Walton Incised var. Sneads; G, Point Washington Incised var. Point Washington; H, Point Washington Incised var, Griffith
Figure 160. Examples of Fort Walton incised pottery types and varieties typical of the Bull Creek phase (taken from Scarry 1985).
186

The original definition of Lamar Bold Incised was based primarily on the WPA excavations in the Macon area, and as a result those definitions reflect largely the variety of pottery found in that area (Kelly 1938:47-48; Jennings and Fairbanks 1939). Willey applied those definitions to the Fort Walton area with only slight modifications:
Definition: A grit-tempered ware decorated with broad incised lines, hollow reed punctations, and rarely, dot punctations. Curvilinear designs (scrolls) usually combined with rectilinear elements, particularly horizontal lines between scrolls. Some rectilinear elements stand alone. A row of hollow-reed punctations is usually placed at base of the incised decoration which is a border around upper portion of the vessel. The forms of cazuela bowls, on which the incised designs occur, are often covered on the base with Lamar-type complicated stamping.
Relationship of type: Is related to Fort Walton Incised. May be ancestral, in part, to Ocmulgee Fields Incised and Aucilla Incised (Willey 1949:493).
Willey defined Fort Walton Incised as follows:
Technique: Lines and punctations incised into soft surface of vessel. Lines are deep, wide, and usually rectangular in cross section. Large round dot or square punctations most common. Hollow-reed punctations sometimes used.
Design: Elements are volutes, interlocked scrolls, running scrolls, circles, trifoil figures, crescentic forms, S-shaped and reverse S-shaped figures, rectilinear stepped figures, pendant loops, and triangles. Elements are usually repeated around vessel in a connected design pattern. Dot punctations used as filler for both backgrounds and for design proper. Incised lines sometimes used as
fillers (Willey 1949:460).
Willey also named several other incised pottery types for the Fort Walton area, including Pinellas Incised and Point Washington Incised, which have at times been applied to the Bull Creek phase and have been incorporated into Scarry's scheme.
Distinguishing Lamar incising from Fort Walton incising is to a great extent con-

Bull Creek Site Material Culture
tingent on sherd size. Large sherds are required to distinguish patterns that may be typical of north Georgia Lamar pottery from those of the lower Chattahoochee Fort Walton pottery. Unfortunately there were very few large incised sherds recovered from Bull Creek.
During our examination of the Bull Creek collections, all incised designs were drawn as encountered. Most were too small to provide any meaningful indication of design. Eventually 79 sherds were determined to be sufficiently large to determine all or part of a design. Several of these were vessels reconstructed by the National Park Service. Because of the wide variability present in the collection, it was necessary in many cases to create general design categories. This was especially true with respect to the zone punctate vessels where designs differed on each sherd. General rectilinear and curvilinear design categories were selected when there was substantial variability. Figure 161 illustrates examples of each distinct design pattern found in the Bull Creek collections. Table 12 describes the criteria used to distinguish these motifs.
In examining the incised motifs found at Bull Creek, one point becomes obvious. Most of the designs are common Fort Walton decorative patterns. There is little indication of typical Lamar designs. Design Category H which consists of nested half circles bordered by parallel lines, is most similar to a Lamar Bold Incised motif. The recovered examples appear to contain two to four lines in the design, which would be most typical of an early Lamar incised style.
Recognition of Design Category L is particularly meaningful because it represents the most common form of Pinellas Incised. As previously noted, Schnell now considers this style of incising to be typical of Stewart phase but not Bull Creek phase. At the Bull Creek site, this single sherd was found in a probable Stewart phase feature.
187

The Bull Creek Site
A

'.. .-z;:.....~

"'

~

B

.8 D
c

u,

...-,--,y,-r
e.::. f~c: ~

I'

: C: ..... FC..

" ,.., .

&.,.~Cl:".o.

F

G

.. r.

~

.".

" r:
"e



~

IF-

. ~~~= "" . J

1 & ' / '/~ ~'.~~1)

L

N

0)'\)l;l~\: ('(/

o

.~

. "'...

p

Figure 161. Incised designs found at Bull Creek (letters correspond to Table 12).

188

Bull Creek Site Material Culture

Table 12. Incised design categories identified from the Bull Creek site collections.

1c::~t~g8l"yl>

..... ...... I)~S8ripti9fI>

I >1 Count

A

Band of interlocking scrolls (guilloche) with dot (punctate) filler (Fort Walton Incised). 9*

B

Simple curvilinear stepped band alternating with punctate band.

2

C

Parallel undulating bands with alternating punctate fillers.

1

D

Simple continuous undulating band with punctate fillers.

1

E

Complex designs of punctate filled oval and rectilinear elements alternating with clear 17*

zones or elements.

F

Rectilinear step elements with alternating punctate-filled and clear zones.

24*

G

Concentric half-circles with a punctate background

1

H

Concentric half-circles bordered by parallel lines.

3*

I

Concentric circles

1

J

Band of interlocking scrolls without punctate filler (Point Washington) .

5*

K

Running or interlocking open scroll.

3*

L

Running Scroll (Pinellas Incised)

1*

M

Simple looping pattern.

1*

N

Multiple parallel lines.

3

0

Simple half circle single line loops at rim.

2

P

Simple discontinuous rectilinear and curvilinear incised elements below rim.

4*

Q

Hand-eye motif

1

*One or more examples recovered from the probable Stewart phase feature excavated in 1981.

Negative Painted Wares relate to painted water bottles at Bull Creek. There are a few references to fragments of painted pottery found in the general midden of the cemetery area excavations (Lester 1937:54), but these sherds were not relocated in our examination of the collections. Schnell does identify one sherd from Kelly's collection.

In 1979 the Bull Creek negative painted pottery was given the type name Nashville Negative Painted variety Columbus and was considered a local copy of similar vessels from the northwest (Williams 1979). More recently Scarry gave the Bull Creek pots a new type status as Columbus Negative Painted variety Columbus (Scarry 1985:213).

The dog pots from Bull Creek included two varieties of painting. The two vessels from Burials 3 and 7 exhibited red spiral designs on a buff background. The third exhibited a black pattern on a reddish background. A third vessel is most similar to a dog pot recovered from Neisler Mound (see Figure 156). All three designs on the Bull Creek dog pots are quite similar running scroll patterns superficially comparable to the rarely encountered Pinellas Incised vessel design category L (Figure 161-L).

Scarry defined Columbus Negative Painted to include all sand-tempered vessels with indirect (negative) painting in the Fort Walton area, as a distinct type from the shelltempered Nashville Negative Painted wares. Scarry noted that the ware has thus far been found only in Rood and Bull Creek phase contexts. Variety Clay was recommended for the negative painted vessels from Cemochechobee, which differ in vessel form and painted design.

189

The Bull Creek Site

Schnell has noted on several occasions that only four examples of the Bull Creek varieties of negative painted dog pots are known to exist (Schnell 1990:69). This conclusion has been reaffirmed through communications of both Schnell and the senior author with individuals knowledgeable of the antiquities market in the region. The recovery of three-fourths of the known examples of this vessel form from a single site, Bull Creek, does represent a unique occurrence.

Bull Creek/Go-Kart site locality also produced other Creek pottery types such as Kasita Red Filmed and Ocmulgee Fields Incised. Excavation at the Go-Kart site produced substantial amounts of earlier Averett phase ceramics. These ceramics are discussed in detail in a separate report of the Go-Kart site excavations (Ledbetter 1995c) and will not be repeated here.

Other Pottery Types found in the Bull Creek collections include anomalous forms of decoration and a few sherds from other occupations. Several sherds with cordage impressions are listed on the National Park Service summary sheets and examples were relocated in those collections (Figure 162 upper). These sherds may be from non-Bull Creek phase occupations, but they may also represent impressions which commonly occur on vessel bottoms.

Cord Marked

The collections also include a single fiber tempered sherd from Kelly's collection (Figure 162, center) and several Chattahoochee Brushed sherds (Figure 162, lower). Excavations conducted by SAS of the Bull Creek site and the contiguous GoKart site indicate a widespread occurrence of Archaic and historic Creek material across these site areas. The fiber tempered sherd reflects the Late Archaic occupation and the Chattahoochee Brushed sherds reflect the historic Creek occupation.

Fiber Tempered

o

3

E-5*e

em

It should be noted that most of the Creek occupation occurred beyond the boundaries of most previous excavations, as is reflected in the extremely small amount of ceramics recovered from these collections. SAS excavations conducted in the larger

Chattahoochee Brushed
Figure 162. Examples of other pottery types from Bull Creek (Full Size).

190

Other Ceramic Objects

Bull Creek Site Material Culture

Examples of ceramic pipes and pottery discs have also been recovered from Bull Creek, but were quite rare in the collections. The low counts of pipes appear to be an accurate reflection of the scarcity of the artifact type on the site, but there is some contradiction with respect to pottery discs. Chase noted "pipes are infrequently found at Bull Creek, only three stems and one bowl fragment came to light but hundreds of pottery disks and one of stone appeared. This is not surprising since the disks prevailed from earliest Lamar times into historic period" (Chase 1957).

Our reexamination of the collections produced two complete pipes from the cemetery excavations and four fragments from the village midden. Of this total, two were recovered from the pit excavated by Fuller in 1981. We were able to record only six pottery discs from the collections. Of this total four were found in the Fuller collection. With respect to ceramic discs, we are left with the possibilities of an error in Chase's assessment of numbers or the actuality that discs were commonly found only in the portions of the site examined by Chase and Fuller.

The most complete pipes and pipe fragments found in the surviving collections from Bull Creek are illustrated in Figure 163. The upper two examples are the two effigy forms recovered from the cemetery (Burials 17 and 18). The lower examples are more simply fashioned examples from the midden.

Pottery discs from Bull Creek ranged in diameter from 30 to 45 mm. Surface treatment was divided between plain and complicated stamping (Figure 163)

One clay bead was reported from the cemetery excavations (Find 49). The artifact was not relocated by the author.

Figure 163. Examples of ceramic pipes and pottery discs from Bull Creek.

191

The Bull Creek Site
Shell and Bone Objects

Chase noted that objects of shell were not recovered from the village midden (Chase 1957). However, beads made from marine shell were relatively common in burial context. Shell beads were found with Burials 3, 18, and 20. A conch shell core was also found with Burial 18. The NPS list of special finds includes one bone awl (Find 74) from the cemetery area.
Stone Objects

Numerous objects of chipped and

ground stone occur in various collections from

Bull Creek, but there are no tabulations. The

Celt

majority of these belong to occupations other

than the Bull Creek phase. For this reason

only a general characterization, with a focus

on the Bull Creek phase, will be presented.

Groundstone includes many objects, such as generalized grinding implements which are shared by many cultures. Objects that are clearly associated with the Bull Creek phase are primarily burial objects. Included are greenstone celts and chisels and stone discoidals or chunky stones (Figure 164). Examples from the cemetery include celts and chisels from Burials 5, 6, 9, and a discoidal from Burial 7. Fragmentary celts and ground pebbles of undefined function were also recovered from midden contexts but no accurate counts are available.
Ground pigment was also recovered from several burials (Burials 6, 28, and 29). The descriptions from the WPA notes appear to indicate red ochre or hematite.

Chisel

Ground Pebble

Little is known concerning other types of ground stone associated with the Bull Creek phase simply because few examples survive in the collections. Listed "Finds" consist primarily of crude chunky stones. There is no available data for large grinding implements comparable to a metate or grinding slab.

Discoidal

o

3cm

15!5!==~

Figure 164. Examples of other ground stone objects from Bull Creek.

192

Chipped Stone Artifacts were substantially less common than pottery at Bull Creek, but apparently did occur in significant numbers. This is reflected primarily in the NPS list of finds which records a number of projectile points. As previously noted, these projectile points date primarily between the

Bull Creek Site Material Culture
Paleoindian and Late Archaic periods. Most are made from quartz and Coastal Plain varieties of chert. Figure 165 illustrates selected examples of projectile points from the WPA collections. The collections also include small amounts of lithic debris and expedient tools which probably date to these earlier occupations.

o,.......--

3

em

Figure 165. Examples of projectile points found on the Bull Creek site which pre-date the Bull Creek phase occupation.

193

The Bull Creek Site
The only chipped stone artifact that can definitely be associated with the Bull Creek phase occupation is a large biface recovered from Burial 6. The large blade is made from a dark chert most similar to material found to the north in the Ridge and Valley province. The biface measures 83 mm in length.
Modern Artifacts
A small number of historic artifacts have also been reported from the Bull Creek site. Some objects, such as a metal thimble mentioned in the NPS notes, may date to the historic Creek occupation. Most artifacts recovered from the site are modern and relate to modern land use. These objects will not be discussed.
In summary, the majority of artifacts associated with the Bull Creek phase are ceramic objects. The collections of ceramics are dominated by vessels of medium to large size. There is, however, a full range of vessel sizes and a moderate range of vessel forms. The remaining ceramics consist primarily of pottery discs and smoking pipes. Ground stone objects are represented by a few well-made ground stone artifacts such as celts, chisels and discoidals. Chipped stone production and tool use appears to have been insignificant during the Bull Creek phase. Little is known concerning objects made from organic materials due to poor preservation. There does appear to have been a significant use of marine shell, which survives primarily in burial context.

o
P"5

3cm i

Figure 166. Chert biface associated with the Bull Creek phase occupation.

194

Chapter 9 Summary and Discussion

Our examination of the records and collections from previous investigations of the Bull Creek site disclosed a wealth of information. While records of most investigations were incomplete, surviving notes, maps, collections, and quite often unpublished manuscripts did allow all past excavations to be documented. The curation of notes, photographs, manuscripts and collections relating to the 1930s excavations of Frank Lester and Isabel Patterson was accomplished through the efforts of numerous personnel of the National Park Service. Preservation of material relating to the other excavations has been the responsibility of personnel of the Columbus Museum. Frank Schnell, Jr., in particular, has diligently preserved records of other excavations at Bull Creek which otherwise could not be documented. The Columbus Museum's files of Isabel Patterson, which includes correspondence and manuscripts, proved to be an extremely valuable source of information. The importance of the Columbus Museum as a curation facility cannot be overstated with respect to the completion of this report.
The primary goals of this project were the publication of the findings of previous investigations and a synthesis of characteristics of the material culture that defines the Bull Creek phase. This has been accomplished to varying degrees of success in the preceding chapters. Our success in researching previous investigations often depended upon the fortunate preservation of a few notes, letters, draft manuscripts and newspaper articles. Figure 167 shows the locations of major excavations outlined in the following paragraphs.
The National Park Service files contain substantial amounts of information and thus made a major contribution to this report. Shortcomings of the NPS records include missing field maps, particularly relating to the

cemetery, a lack of an osteological report, and the absence of a synthetic or interpretive report. Lester's field notes appear to be complete, but his original hand-written notes have not been located. We do not know how much information, if any, is missing in the notes transcribed by NPS personnel. With respect to an interpretive report, the lack of such a final report by A.R. Kelly was the primary motivation for the present project.
The excavation project conducted by the University of Georgia in 1950, under the direction of A.R. Kelly, was extremely important with respect to interpretation of the site, and records of the project survive only because of the public's interest in archeology. Aside from the collections, no records of the excavations survive at the University of Georgia. However, photographs and a brief report were produced by the Columbus Chamber of Commerce and these records have been preserved in the files of the Columbus Museum.
Records of more recent research features the work of Frank Schnell, Jr., who excavated on the site in 1959, and as a graduate student at the University of Georgia, undertook the site as a thesis topic. Schnell did not complete a Bull Creek synthesis due to a change in thesis topic; however, he did accumulate a substantial amount of data, particularly with respect to ceramics. Fortunately much of Schnell's data have been preserved and major portions of that research have been reproduced in the ceramic synthesis of this volume.
Records of other investigators, such as Frank Mulvihill, Margaret Ashley and Frank Schnell, Sr., Frank and Barbara Fenenga, David Chase, Jack Tyler and Terry Jackson, and Jackie Fuller, are found primarily in rare publications and unpublished manuscripts preserved in the files of Columbus Museum.
195

The Bull Creek Site GEORGIA

/

\

A ..

r:

ALABAMA

Kelly (1950) Unit4 and Schnell ----+-~.

(1959) Unit 2

/ - - ",-~

WPA Cemetery Block------lr--4WI\\ and Kelly (1950) Unit 3

N

i

o

300

m

Figure 167. Locations of major archeological excavations at Bull Creek. 196

Accounts of work conducted on the contiguous Victory Drive site were found in a manuscript prepared by Schnell (1970). Those excavations relate primarily to an historic Creek occupation. Site boundaries created by Schnell for the Victory Drive site represent the extent of the Creek occupation. As a consequence of defining a distinct Victory Drive site, the modern site boundaries of Bull Creek differ from the boundaries of the old Bull Creek Village described in the writings of Lester (1938) and Patterson (1936).
Isabel Patterson's major work on the Bull Creek site is a 1950 article in Early Georgia, but a substantial body of related writings are housed in the files of the Columbus Museum. Included are letters to A.R. Kelly and others, drafts of several manuscripts, and several newspaper articles authored by Patterson. Additional records are found in various cultural resource management reports, the most current and site-specific of which was completed as a result of the Columbus Riverwalk project (Ledbetter 1994b).
Our examination of the material culture of the Bull Creek site focused on ceramics. Relatively few artifacts of other types have been collected from the site and little research has been conducted upon material such as lithics and subsistence remains. In the case of lithics, we are dealing with a Late Mississippian component at Bull Creek that made little use of chipped stone. Like contemporary Lamar populations in the Piedmont, it is apparent that small, triangular, chipped stone projectile points were not being produced at this time. While small chipped stone points were commonly found in association with the Early Mississippian Averett component on the contiguous Victory Drive site, there are no confirmed records of the recovery of any of these points in the excavated portions of the Bull Creek site. The only confirmed chipped stone artifact recovered was a large chert biface found in Burial 6 (see Figure 166). A few groundstone celts and chisels were also found

Summary and Discussion
in burial context and broken examples have been found throughout the midden deposits. Few other stone artifacts exist in association with the Bull Creek phase occupation. Small amounts of pigment stone have been recovered, again from burials. The site has produced a number of marine shell artifacts, primarily beads. The majority of shell artifacts have been found in burials. Past excavations recovered sparse quantities of subsistence remains. In the older excavation projects, bone and charcoal were not commonly saved. In the most recent excavations conducted by SAS, no large pits with the potential for preserved organic material were located. Today, subsistence studies remain a primary deficiency in all investigations of the Bull Creek site.
The study of Bull Creek ceramics has been a major concern of this report. An effort has been made to integrate the ceramic studies resulting from earlier investigations with our own analysis, which focused upon vessel form. Problems in this undertaking resulted from variability in the quality of the original data from different sources, incompatibility of analysis categories and terminology, and the loss of some collections, which makes replication of the original work impossible. Also, differences in the frequencies of certain pottery traits, particularly surface treatment, between total sherd counts and minimum number of vessel counts made comparisons difficult. Nevertheless, we were successful in characterizing the Bull Creek pottery assemblage using these sources of data. This study was not exhaustive; however, much of the primary data relating to vessel characteristics have been presented as an appendix and hopefully will be examined further by other researchers. A primary result of our pottery study is the conclusion that the composition of the ceramic assemblage found on the Bull Creek site is consistent with interpretations of Schnell and others and supports the need to distinguish an early (Bull Creek) and late (Stewart) phase (Schnell 1990:67; Schnell and Wright 1993:21).

197

The Bull Creek Site
Our examination confirmed Schnell's interpretation that the Bull Creek site pottery was sufficiently different from sites such as 9Sw51 (Broyles 1962) and the Park's Mound site (Hally and Oertel 1977) to be defined as a distinct phase. At the Bull Creek site, the higher frequency of complicated stamping and the near absence of late Lamar incised and Pinellas Incised motifs does appear to indicate an earlier occupation. Unfortunately, we do not have reliable radiocarbon dates to further verify this interpretation.
Other research issues may also be pursued using the accumulated information. Issues relating to the Bull Creek village, the Bull Creek cemetery, as well as the status of the Bull Creek site within the greater Mississippian community can be addressed.
A greater understanding of the extent and architectural layout of the Bull Creek village can be gained by examining the findings of past research conducted at the site. Unfortunately, the loss of much of the site area to development prior to any thorough archeological survey means that major aspects of this issue cannot be resolved. We have good locational information for the excavated structures, and the locations of other possible structures may be inferred by the recovery of house daub and the discovery of domestic features, such as large trash-filled pits.
Several questions can be addressed with further examination of the Bull Creek cemetery by incorporating the findings of the Lester's 1930s and Kelly's 1950 excavations with modem osteological examination of the human remains. A very important question that remains unanswered is whether the burials represent an actual cemetery laid out separately from the village or whether the burials were within the floors of domestic structures.
Another question relating to the burials is the simple issue of sex and age composition. This question is addressed to some degree with
198

new information from an examination of the surviving remains by Chad Braley. Using Mr. Braley's 1994 observations to supplement the field observations of Frank Lester, a greater level of interpretation is possible.
Finally, we can examine the site from a larger regional perspective. This has been accomplished with the assistance of Adam King, who is currently involved in comparable research relating to the Etowah Mound group in northwest Georgia. The paragraphs relating to this topic have been prepared primarily by Mr. King. The basic research question to be addressed concerns the evolution of the Bull Creek site and the Bull Creek phase as the result of the movement of people or the adoption of basic concepts resulting from interaction of local populations with other groups.
The Bull Creek Village
The most recent investigations conducted on the Bull Creek site confirmed that most of the site has been destroyed (Ledbetter 1994b). Today, only a small strip of the site survives along the wooded river bluff edge above the Chattahoochee River. Data recovery investigations conducted on the contiguous Victory Drive site indicates that a series of structures associated with the Bull Creek Village extended along the river for a minimum of 450 m from the mouth of Bull Creek (Ledbetter 1995).
Based upon the combined work of previous investigations and the recent survey (Ledbetter 1994b), it appears the Bull Creek Village extended for a substantial distance along Weracoba Creek. Minimally, the village extended along the creek as far as the eastern boundary of the Fenengas' (1945) Weracoba Creek site (9Me379). This would put the eastern boundary of the site across Victory Drive and a straight-line distance greater than 500 m from the Chattahoochee River (Figure 168).

GEORGIA

Summary and Discussion

/

\ ;--...

~ ...

."

ALABAMA
BULLCREEK CEMETERY

N

i

o

300

e

m

Figure 168. Map showing estimated extent of the Bull Creek Village.

199

The Bull Creek Site
The aerial extent of the Bull Creek Village may be only partially reconstructed using the results of previous archeological investigations. Because most investigations were conducted in the richest and best preserved areas of the site, most of our knowledge of the site comes from the area nearest the Chattahoochee River and Weracoba Creek. There is little information for the interior portion of the site. There is also no means of accurately determining the eastern limits of the site. For that reason, we cannot determine whether the village existed as a planned "town," with structures encircling an open plaza, or whether the site consisted of a series of structures strung along the two waterways.
Past investigations have produced no information on the presence of ceremonial or sacred structures and relatively little information on domestic structures. There are no confirmed accounts of the presence of a mound on the site. Information concerning domestic structures comes primarily from the WPA and SAS investigations.
Lester and Patterson found portions of two domestic structures. The structures were substantial post-construction houses plastered with daub. The structures were basically square but appeared to have had rounded corners. The only mapped structure (Number 1) appears to have been approximately 6 m wide. A large pit, probably excavated to procure clay for house daub, was located adjacent to the structure. A similarly shaped structure was encountered during data recovery of the Victory Drive site (Ledbetter 1995). That structure measured approximately 7 by 7 m.
While specifics of the village layout cannot be determined, there is ample information to conclude that a substantial number of structures were present on the site. A review of all excavations produced consistent evidence of structures in all areas of the site investigated. In past excavations conducted along Weracoba Creek and in the more interior por-
200

tions of the site, house daub and pits were consistently found. Based primarily upon the work of Chase, Schnell, and Fuller, it is apparent that structures were present along Weracoba Creek and extended into the interior of the site for a minimum distance of 150 m from the waterways.
One of many questions that cannot be answered with available information is the length of occupation. Substantial amounts of midden were deposited across a large site area. The beginnings of this accumulation equate to the Bull Creek phase time period. Apparently, the site continued to be occupied, or was reoccupied, during the subsequent Stewart phase. At present, we are unable to judge the intensity or extent of the Stewart phase settlement.
The Bull Creek Cemetery
The Bull Creek Cemetery was excavated in 1936 and 1937 under the direction of Frank Lester (Lester 1938). Lester exposed 46 burial pits containing the remains of at least 50 individuals in a block excavation that covered approximately 400 m2 (Figure 169). While a sizable block was excavated, there is no evidence that the entire cemetery area was exposed. Today, prospects linger that additional burials are still preserved in the unexcavated peninsula of land that lies above the confluence of Bull Creek and the Chattahoochee River.
Much of Lester's preliminary report focused upon the cemetery excavation and description of individual burials (Lester 1938). Lester's account has been presented in Chapter 3 and additional field note information may be found in Appendix A of this report. Table 13 is a summary of Lester's burial data with new observations based upon the osteological reexamination of the remains by Braley. Observations concerning burial orientations and layout of the cemetery are also presented in this chapter.

Summary and Discussion

4-27 4-28

Find 72

4-25

4-26
B18
I( .,
.'0

~ B19 IJ

B44 0'-'?J).B46
[J
B45
B43

IY'\'
B39
D B42
{}

B5

llf.'\ \,

~

0~)/,,>B7

. 4-21 4-22
'V.~ B14

B36 B40

~ ,- \Wl'~_B37 B35

"
-.(9qo;
B28
B16
..,.
~
B26 <Xl

B25
B17
B30o:,_l,'0.<1>'" "~'_ B31

BI3 t.~ B22

B3
~M""'

4-11
Sl
eI B12

4-12
<'~B2

II "I

t;J,

B24

~q B32

., B34

~ B33

0'

4-4

.ot..v...

L33 L31 L29 L27 L25 L23 L21 Ll9 Ll7 Ll5 Ll3 Lli

L9

L7

L5

L3

Ll

4-1

4-2

RI

R3

Figure 169. Plan map of the Bull Creek Cemetery.

The Bull Creek Site

Table 13. Summary of Bull Creek Cemetery

BlinCll.
Depth" ..2 .,

.,;~

1><

>

::-:-:.:':.:-".:..-:-:.

1

unknown

NAINA

skull only

adult male, skull complete

none

2

contracted flexed 67"fNA

skull elongated

adult, sex and age unknown

none

3

contracted flexed 67"fNA

some post-mortem disturbance

an elderly adult male, teeth worn down to pegs in front, secondary dentin exposed

effigy pot, shell beads

4

flexed

44"/20"

female

sex and age unknown

none

poorly preserved, skull gone Fairly robust tibia and femurs, sciatic notch is

5

contracted flexed 72 "/NA except for teeth, no cervical less than 90 degrees, suggesting male. Adult, greenstone celt

vertebrae or scapular

little attrition of teeth - no cranial fragments.

(7")

Left tibia has thickened mid-shaft; possibly a 1 greenstone

healed fracture. Fragments of long bones celt (4") and 2

poorly preserved,

(legs), vertebra, tarsal, maxilla and mandible chisels,

6 flexed, feet folded 78"/42" and severely damaged during

(no cranial frags). Male suggested by

worked flint,

back to pelvis

excavation

ruggedness of long bones. Person had lost first black flint

premolar and both molars (lower right) before knife, red

death. Moderate attrition on teeth.

pigment

skull, right fibula, both fragmented femurs and long bone fragments, effigy pot,

7

Contracted flexed 78"/42"

femora, tibiae

no cranial fragments

discoidal

good shape, possible cranial robust tibia and femurs (male?), no cranial

8

contracted flexed 68"/36"

deformation

fragments

none

9

contracted flexed 66"/32"

poorly preserved

adult, sex unknown, moderate tooth wear chisel 3l/z"

10

skull only

54"fNA

*****

missing

none

Adult male (cranial morphology). Complete

skull is rather small and the mandible and maxilla, fragments of skull,

11 folded feet back to 73"/36"

individual was of

femurs and tibia. Skull shows frontal

none

pelvis flexed

comparatively small stature. deformation. Antemortem loss of lower right

This is a female burial.

first and second molars. 2 rom pit cavity in

lower left third molar.

fragments of skull and long adult male-skull cap has blunt supraorbital

12 contracted flexed 69"/35"

bones

margins (consist of four long bones)

none

skull fragments, teeth,

13

undetermined 40"/26"

cervical vertebrae

missing

none

Skull is slightly misshapen,

terrible infection of the leg - massive

right parietal shows patholo- thickening of the right femur, tibia and fibula.

14

(Old Man)

51 "/34"

gical enlargement -

Fibula is almost unrecognizable as such.

none

contracted flexed

depression present at the Probable adult male based on left femur robus-

lambdoid inter-section-old ticity. No cranial bone, but fragments of

male.

pelvis and long bones are present.

Elderly person, sex unknown, skull-vertebra

skull crushed, short

fragments, mandible and maxilla. Bone in

15 contracted flexed 78"/53" fragments humeri, ulna and

good shape but incomplete. There was

none

radius, pelvis, traces of

secondary dentin exposure on most teeth.

femora and tibiae

Front teeth so worn they are peg-like.

16

double burial NA/36 "

*****

missing

effigy pot

17 contracted flexed NA/NA

*****

Skull cap and three other cranial fragments are total collection. An adult- gracile char-
acteristics of skull suggest female. There is slight cranial modification of occipital.

effigy pipe

consists of a few fragments 18 contracted flexed 61 "/39" of the skull and sections of
femora

missing

bowl, shell beads, core, effigy pipe

202

Summary and Discussion

Table 13. Summary of Bull Creek Cemetery

I taL 1/>

-~

> .: BUriiil . :.>.-.>-.:"."::".-:::::

.......

.: pePtli* :.

i:. ....... <:.

==

female, skull badly crushed, Sciatic notch of > 90_ probable female.

contracted more

no cervical vertebrae, rest of Inventory: fragments of cranial, pelvis, femur,

19 than any found yet 67"/45" the body well preservation tibia, ulna, humerus, phalanges, tarsals,

none

- knees on chest

scapula, vertebra, ribs. No teeth, maxilla or

mandible (cranial modification).

male - skull, cervical

An adult female(?) - broad sciatic notch,

flexed-folded feet

vertebrae, femora, both

gracile skull. Lower right first molar lost string of shell

20

to pelvis

NAINA tibiae, clavicles, scapulae, antemortem, attrition of teeth suggests young seed beads at

pelvis

adult. Some cranial modification.

neck

flexed - feet

skull, tibiae, femora, ulna,

21

folded to pelvis 36"/21 "

and radii

missing

none

skull crushed but excellent adult male in very good condition but there is

22

flexed

54"/40"

preservation

no skull - a very robust person

none

23

undetermined 57"/33"

bone fragments

missing

none

24

contracted flexed 59"/28"

skull, long bones

missing

none

25

folded flexed 52"/21 "

skull and long bones

missing

none

26

extended flexed 79"/41"

skull and long bones

missing

none

27

contracted flexed 75"/41 "

skull and long bones

missing

none

portions of left tibiae and

pigment, bone,

28

flexed

80"/35" femur and the skull with all

adult, but age and sex unknown

stone, shell,

of the facial bones missing

bone awl

29 contracted flexed 88"/47"

long bone fragments

missing

red pigment

flexed - with

fragments of skull, tibiae, adult - Inventory: cranial (moderate tooth

30

knees farthest 49"/11 " femora, small section of the wear), mandible, long bone fragments-contains

none

away from body

pelvis

mandible fragments of two people

triple burial - 2

31 contracted flexed, 61"/23" (see description in Chapter 3)

missing

none

3rd is skull only

portions of skull, tibiae, probable male, based on robusticity, inventory rocks, flint

32

contracted flexed 52"/34" femora, humeri, and ulna consists of cranial, mandible, and long bone chips and

fragments

animal bones

33

contracted flexed 36"17" frag. skull, teeth, long bones

bad shape, probable adult

none

34

contracted flexed 37"/16" fragments of skull and tibia

bad shape, adult?

none

skull badly decayed,

Cranial deformation of occipital. Skeletal

35

contracted flexed 67"/27" fragments humerus, ulna,

remains of a child - sciatic notch suggests

none

lumbar vertebrae, pelvis, male. Unfused ephipheses of head of femur,

long bones

greater trochanter, and acetabulum.

36

contracted flexed 60"/35"

poorly preserved

missing

none

Excellent shape, possibly an adult male, but

exceptionally well preserved fairly gracile vault. Most of cranium intact, no

37 contracted flexed- 86"/36"

deformation - strong mandible. Lost lower

none

face downward

right first molar, lower left second premolar

and first and second molars before death.

Adult male. Very broad dental arcade. Chin is

38

double extended NA/36 " male and female (guess). unusual, shaped like a "snow plow". Complete

none

burial

See Chapter 3.

mandible and maxilla. This burial also

contains an extra mandible (right fragment).

203

The Bull Creek Site

Table 13. Summary of Bull Creek Cemetery

II ilii'M== ...r"..

P

:

8pP



':"",

skull, vertebrae, humeri, Female with extremely thin skull (teen-aged)
ulna, radii, metacarpal bones. no cranial deformation. Third molars not

39

contracted flexed NA/42 " Portions of pelvis, tibiae, erupted, little tooth wear. Collection includes

none

femora, fibula, metatarsal mandible of a very old person - all teeth of

bones

second person are missing from canine back.

40

skull fragments 61"/42"

skull fragments

an adult? sex unknown.

none

fragments skull, clavicles,

41

contracted flexed NA/39 "

pelvis, long bones

adult? sex undetermined

none

42

contracted flexed NA/26 "

well preserved

A child, lacks third molars, roots of second

molars still unfused, unfused ephipheses of

none

femur and scapula.

NA/50" skull, clavicles, vertebrae, an adult, possibly male based on robusticity -

43

extended flexed

femora, tibiae, pelvis

very fragmented

none

44

contracted flexed NA/26 "

complete

an adult?

none

45

contracted flexed NA/33 "

poorly preserved

an adult male

none

46

contracted flexed NA/28 " teeth, long bones fragments

missing

none

unknown

Box 219. Burial? - Accession number 3923323. Two skull caps, 1 male, 1 female or child, both with fronto-occipital flattening.

unknown

Burial? Accession number 39-23326. An adult male with cranial flattening.

unknown

Burial? Accession number 39-23324. Contains remains of two people, one adult with heavy tooth attrition, one adolescent with
unfused roots on third molars (they had partially erupted, however).

unknown

Burial? Accession number 39-23329. Adult, possibly female - sciatic notch of about 90 0

*Depth: First reading is depth below surface (in inches), second is depth below occupation zone (Lester's measurements).

Our reexamination of the Bull Creek burials provided a few new insights. While there are a few contradictions between field notes and our reexamination concerning the sex and age of a few individuals, the observations are generally consistent. In this respect, a most noticeable finding is the absence of infants and small children. The cemetery appears to be composed only of adolescents and adults. Our reexamination also provided information for cranial deformation which was not included in Lester's notes.
The loss of a few individuals is apparent. This may represent a combination of extremely poor field preservation and cura-
204

tion problems. We do not know if all bone was saved during Lester's excavations. Consistent with NPS laboratory procedures of the times, the Bull Creek skeletons were carefully labeled and chemically treated to insure preservation. However, curation problems may have occurred as the skeletons, or portions of the skeletons, were examined by various individuals resulting in displacement or even loss of some remains. We know, for instance, that the skulls of certain burials were carefully restored and studied following excavation. This is noted in a letter from A.R. Kelly to Isabel Patterson dated June 25, 1937.

In regard to the 4 skulls Dr. Childs has. I saw him after receiving your letter. He will repair them as soon as he gets around to it. Since he is using his own materials, plaster, modeling materials, and his skill, more as a labor of love, I hesitate to press him. Also, you should know that 4 skulls restored is hardly a seriation on which to base any important conclusions regarding racial type, or for use in making comparisons with other Indians. They are valuable as a start toward a skeletal collection. I wouldn't however regard their study as an essential part of the report. Burial data should
suffice here (Ledbetter 1995b:19).
Obviously, Kelly fully intended to have the burials examined in detail. In a letter to Patterson dated July 9, 1937, Kelly stated his plans.
I think now that Marshall Newman (Chicago and Harvard) is our best bet, following suggestion from Setzler. This is the young man, incidentally whom I recommended to do Physical Anthropology on Bull Creek skeletal material (Ledbetter 1995b:21).

Summary and Discussion
Records of Newman's work has not been found, if indeed, any was ever conducted. If records do exist, they would be invaluable in light of the present status of the burial remains.
Questions still remain concerning the initial designation of the Bull Creek Cemetery as a discrete burial grounds separate from the remainder of the village area. Lester (1938) was not able to demonstrate the presence of structures in the area. Lester did note the existence of a thin clay layer, which might be structurally related, but his comments provide minimal information on the feature (see Figure 69). Kelly's excavations near the cemetery produced ample house daub, suggesting the presence of structures in the cemetery area (Kelly 1950). An examination of the orientations of the burials provides one means of assessing the possibility of a planned cemetery (Figures 170 and 171).

/

/

/
,/

4-25 4-26

//
/ /
,/

NE Cluster

-, -,

/

///'<,

-,

/'
/

--- ------ '>" ........ ". .

. ...... ' ...........

.........

........ ....

'-.

<,

\
"""'l- \ .
\ .\ \ \

." '\\.\

4-21 4-22
-,

SWCluster

,\ \

\

\

/'

\)

\' /

GraveLine

GN
A

~m

~

o

5 FeCI

.

I
"..."..

r/"""\". \
, \ .................................

4-11
\
\.

4-12

" f,O~~

y,
/

w.

t

/

"

/

-,,,

/ '" y '"

-,
\

/ '"

. '" \ \/-':

43 4-4

/

/

/

/

L33 L31 L29 L27 L25 L23 L21 LI9 LI7 LIS LI3 Lit

L9 L7

L5 L3

LI

4-1

4-2

RI

R3

Figure 170. Burial orientations in the Bull Creek Cemetery.

205

The Bull Creek Site
Northeast Cluster of Burials (N = 27)
Grid North

Southwest Cluster of Burials (N = 13)
Grid North

'$"I
Z
N=4

EtoW ------+----- WIOE

N=4

N=4

Nw I

N=7

Figure 171. Contrasting burial orientations of two burial clusters in the Bull Creek Cemetery.

Burial orientations have been previously examined by Schnell (1990:67). Schnell noted variation in burial orientations with evidence of clustering along a SSE-NNW axis (24 percent) and NW-SE axis (22 percent). These compass bearings reflect Lester's grid north grid which lies approximately 45 degrees from magnetic north (see Figure 170). This means that many of the burials are oriented roughly perpendicular to the length of the peninsula and the channel of the Chattahoochee River.
The plotted orientations of the Bull Creek burials may be interpreted as evidence of both a planned cemetery and house floor burials. Some groups shown in Figure 169 appear to conform to patterns commonly associated with house floors where they would lie parallel to the walls. However, many of

the graves are aligned along a single axis, which could reflect a cemetery layout.
An interesting result of the plotting of orientations presented in Figure 170 is the apparent presence of two clusters of burials. These have been labeled as the northeast and southwest clusters on the map. Table 14 lists the range of orientations for each group. Another distinction is that burial goods were found only in the northeastern cluster.
One interpretation of this patterning is that the northeastern cluster represents a discrete cemetery group with most of the bodies generally placed parallel, while the southwest cluster conforms more closely to patterning evident with house floors. A second interpretation would be that temporally distinct burial episodes are reflected.

206

Summary and Discussion

Table 14. Contrast of Burial Orientations from two burial clusters from Bull Creek Cemetery

(based on Lester's grid north notations)

I

Bl1tiaLOtielltatioIl> >1

NortheastCluster

I

$ol.ltlj\yesti@1llster .

I

North to South South to North East to West West to East Northwest to Southeast NNW to SSE Southwest to Northeast SSW to NNE Southeast to Northwest SSE to NNW ESE to WNW Total Count in Sample

Count = 0 Count = 0 Count = 1 (3.7% cluster) Count = 1 (3.7% cluster) Count = 7 (25.9% of cluster) Count = 1 (3.7% of cluster) Count = 2 (7.4% of cluster) Count = 3 (11.1 % of cluster) Count = 6 (22.2% of cluster) Count = 4 (14.8% of cluster) Count = 2 (7.4% of cluster)
27

Count = 1 (7.7% of cluster) Count = 4 (30.8% of cluster) Count = 4 (30.8% of cluster) Count = 4 (30.8% of cluster)
Count = 0 Count = 0 Count = 0 Count = 0 Count = 0 Count = 0
Count + 0
13

The existence of a discrete cemetery area at Bull Creek remains a possibility. Excavation in domestic structures elsewhere on the site have failed to produce burials, suggesting that graves may not have been commonly placed in house floors. Further insight into this intriguing issue would be gained through excavation and comparison of other Bull Creek and Stewart phase sites.
The Bull Creek Phase from a Regional Perspective
Since the first work was done at the site, researchers involved with Bull Creek have noted the unique character the ceramic assemblage found there. While not always explicitly stated, most would certaiIlly agree that this blend of ceramic decorative characteristics resulted as people, pots, and ideas from different geographic areas mixed. In their discussion of the Bull Creek phase in the Oliver Basin, McMichael and Kellar (1960a) interpreted this mixing in the following manner:

At about 1350-1400, several forces begin to impinge upon Rood's focus, which eventually alters it into a Lamaroid expression. Coming from the south, moving up the Chattahoochee River, Fort Walton begins to appear in many sites, especially incised types and Lake Jackson Plain. This force is so strong that a movement of peoples is probably indicated. Secondly, coming from the east and north is the resurgent stamping tradition, complicated and check-stamped, from type Lamar, Savannah, and Wilbanks sources; again a movement of people is indicated and the Rood's Focus is acculturated into the Lamar tradition, but with the addition of Fort Walton-like traits. The culmination of this is probably to be seen at Bull Creek, on the south side of the city of Columbus (McMichael and Kellar 1960:215).
With the benefits of 35 years of archaeological research, it is now possible to see this mixing of ceramic decorative styles in a somewhat different light. McMichael and Kellar originally saw these changes taking place between AD 1350 and 1400. More recently, Schnell's (1986) definition of the Singer phase has pushed the beginning of these changes back, by approximately one century, into the Middle Mississippian period.

207

The Bull Creek Site
According to Schnell, forms and surface treatments commonly found in Savannah and Lamar period assemblages appear in the Lower Chattahoochee as early as AD 1300. During the Middle Mississippian Period (ca AD 1250-1400) a great variety of raw materials and finished goods were exchanged widely across the Southeast. In fact, during this time the entire Southeast may have been more connected, in terms of movement of physical goods, than at any other time in prehistory. One of the most important raw materials exchanged at this time period, as well as many other periods in prehistory, was marine shell. Chemical sourcing of shell found in archaeological contexts has repeatedly shown that the Gulf Coast was one of the major sources of the widely distributed shell.
Through sourcing a variety of nonlocal materials found at the Spiro site, Brown (1983) has been able to identify one of the Middle Mississippian corridors of exchange through which this shell and other materials and goods moved. This corridor began at the Gulf Coast of Florida, extended northward into the Tennessee Valley, and continued toward the Mississippi River and points westward. The Chattahoochee River was undoubtedly one of the major arteries in this corridor, because it offers a water route that connects the Gulf Coast to a point in the upper Piedmont a short overland distance to the major riverways in the Ridge and Valley. It is likely no accident that two of the major Middle Mississippian centers in the Southeast, Etowah and Lake Jackson, are effectively situated at either end of this segment of the corridor.
Along with the shell and other goods, people and ideas also must have traveled up and down the Chattahoochee, and during the Middle Mississippian period the traffic on this route may have been at its heaviest. In fact the connectedness of the two areas is confirmed by the appearance, in Middle Missis-
208

sippian contexts, of Lake Jackson pottery types at Etowah (King 1991) and Wilbanks-like Complicated Stamped sherds in the Lake Jackson area (Scarry 1985). Also, almost identical embossed copper plates were found in Middle Mississippian burials at both Lake Jackson and Etowah (Jones 1982, 1994). In this context, it should not be surprising that sites in the Lower Chattahoochee exhibit a mixture of ceramic traditions derived from the two areas that the river effectively connected.
The mechanisms that brought about this blending of ceramic decorative characteristics need not have involved the movement of people or even pots. A good deal of archaeological and ethnographic research has been done on the use of decorative style as a means of symbolic communication. Essentially, people do not decorate themselves or their possessions without a purpose, and one purpose decoration can serve is as a means of communicating certain messages (Braun and Plog 1982; Plog 1990; Weissner 1983, 1984; Wobst 1977). Depending on the items decorated and the social circumstances, communication might be aimed at the interpersonal, local, or regional level. Overall the messages communicated are relatively simple-In some sense defining an "us" and a "them". In relatively small-scale, kin-based societies like those of the prehistoric Southeast, those who were not your kin or recognized somehow as part of the social group were feared and mistrusted (Hudson 1976). In the case of the Lower Chattahoochee, the adoption of ceramic decorative characteristics may have been an actively pursued strategy whereby the inhabitants of the area acknowledged and displayed their ties to other people, especially from the north, as a way of making exchanges flow more smoothly. Undoubtedly, this was one of a set of strategies that included intermarriage, gift exchanges, etc., aimed at the same goal.

After about AD 1400, Mississippian ceramic assemblages in the Lower Chattahoochee again look somewhat different from preceding phases. While the Singer phase assemblages consist of Rood phase pottery with some Savannah/Lamar types and modes, the Bull Creek phase assemblage is essentially Lamar with some Fort Walton types and modes. In a sense, the ceramics are changed from a southern decorative tradition with some northern influences to a northern decorative tradition with some southern influences. As before, these changes are probably best understood from a regional perspective.
After AD 1400, the Southeast as a region begins to change. The extensive distribution of goods and the importance of elaborate costumes and displays so prevalent in the Middle Mississippian appear to fade. As Anderson (1990, 1994) has suggested, appeals to ideology characteristic of Middle Mississippian chiefly control may have given way to more secular control in the Late Mississippian period. Whatever the structural change, the connectedness of the Middle Mississippian seems to break down or at least regionalize at this time. Reflecting these changes, after AD 1400 the political geography of the interior Southeast seems to shift. During the Middle Mississippian the landscape was dominated by a few large, widely-spaced political centers such as Lake Jackson, Etowah, Moundville, Spiro, Winterville, and Lake George. In contrast, during the Late Mississippian there appear to have been a greater number of more closely-spaced political centers on the landscape, each apparently controlling a smaller area while intensely competing with neighboring polities for regional control (King and Freer 1995).

Summary and Discussion
During this time, shell remained an important raw material used by inhabitants of chiefdoms in the interior Southeast (see Muller 1989; Smith and Smith 1989), so it seems reasonable to assume that the Chattahoochee was still an important avenue for the movement of shell from the Gulf Coast of Florida to points northward. However, unlike during the Middle Mississippian, the political geography of the region was quite different. The two powerful centers at either end of the Chattahoochee that presumably played an important role in the flow of goods through the valley, had lost much of their influence (Hally and Langford 1988; Jones 1994; King 1991). At the same time, those who lived in the Lower Chattahoochee now had immediate neighbors with whom they certainly had contact, probably both hostile and friendly.
After AD 1400, Mississippian chiefdoms of the Lamar ceramic tradition become established in the Lower Tallapoosa, Flint, and Middle Chattahoochee River valleys (see Hally 1993 for a dating of Mississippian mounds in the area). Thus for the first time during the Mississippian period, the inhabitants of the Lower Chattahoochee were surrounded on three sides by people making Lamar ceramic assemblages. Under these circumstances, the stylistic changes that brought about the Bull Creek phase are probably best seen as adjustments made to material culture as the inhabitants of the area, probably out of necessity, became actively involved in more local social and political matters.

209

THIS PAGE LEFT BLANK

REFERENCES CITED
Anderson, D. G. 1990 Stability and Change in Chiefdom-LevelSocieties: An Examination of Mississippian Political Evolution on the South Atlantic Slope. In Lamar Archaeology: Mississippian Chiefdoms in the Deep South, edited by M. Williams and G. Shapiro, pp.187-213.
1994 The Savannah River Chiefdoms: Political Change in the Lat Prehistoric Southeast. University of Alabama Press, Tuscaloosa.
Anderson, D.G., R.J. Ledbetter and L.D. O'Steen 1990 The Paleoindian Occupation of Georgia. Georgia Archaeological Operating Plans 1-6, University of Georgia, Laboratory of Archaeology Series, Athens.
Ashley, M.E. and F. Schnell, Sr. 1928 Mound and Village Sites in Harris, Muscogee and Chattahoochee Counties, Georgia. Manuscript on File, Columbus Museum of Arts and Sciences.
Braley, C.O., L.D. O'Steen, and I.R. Quitmyer 1986 Archaeological Investigations at 9MC41, Harris Neck National Wildlife Refuge, McIntosh County, Georgia. Report submitted to the U.S. Fish and Wildlife Service, Atlanta.
Brannon, P.A. 1930 Coweta and the Lower Creeks. Arrow Points. Vol. 17 (4) pp. 37-47, Alabama Anthropological Society, Montgomery .
Braun, D. and S. Plog 1982 Evolution of "Tribal" Social Networks: Theory and Prehistoric North American Evidence. American Antiquity 47:504-524
Brown, J.A. 1983 Spiro Exchange Connections Revealed by Sources of Imported Raw Materials. In Southeastern Natives and Their Pasts: A Collection of Papers Honoring Dr. Robert E. Bell, edited by D.G. Wyckoff and J.L. Holman, pp. 129-162. Oklahoma Archaeological Survey, Studies in Oklahoma's Past, No. 11.
Broyles, B.J. 1962 A Lamar Period Site in Southwest Georgia, 9CLA51. In: Survey ofArchaeological Sites in Clay and Quitman Counties, Georgia: 9CLA2, 9CLA7, 9CLA28, 9CLA38, 9CLA51, 9QU25. University of Georgia Laboratory of Archaeology Series Report No.5. University of Georgia, Athens.
Caldwell, J.R. 1955 Investigations at Rood's Landing, Stewart County, Georgia. Early Georgia 2(1):22-49.
Chase, D.W. 1957 The Bull Creek Site, A Type Site Station ofthe Middle Lamar Culture. Unpublished manuscript on file, Columbus Museum, Columbus.
1959a The Averett Culture. Coweta Memorial Association Papers.
1959b The Baird Site. Coweta Memorial Association Papers No.3.
1963 A Reappraisal of the Averett Culture. Journal of Alabama Archaeology 9:49-61.
211

DeJarnette, D.L. 1975 Archaeological Salvage in the Walter F. George Basin of the Chattahoochee River in Alabama. University of Alabama Press, University of Alabama.
Delcourt, P.A. and H.R. Delcourt 1985 Quaternary Palynology and Vegetational History of the Southeastern United States. In Pollen Records of the Late-Quaternary Nortn American Sediments, edited by V.M. Bryant and R.G. Holloway, pp. 1-37. American Association of Stratigraphic Palynologists Foundation.
Elliott, D.T., J.L. Holland, P. Thomaston, and M. Emrick 1995 Historic Preservation Plan for the Cultural Resources on U.S. Army Installations at Fort Benning, Chattahoochee and Muscogee Counties, Georgia, and Russell County, Alabama. Garrow and Associates, Atlanta.
Fairbanks, C.H. 1940 Indoor Archeology, in Prehistoric Culturesofthe Southeast, National Park Service Popular Study Series History No.4. pp 27-32. United States Department of Interior, Government Printing Office, Washington.
Fenenga, F. and B.W. Fenenga 1945 An Archaeological Survey ofthe Vicinity of Columbus, Georgia. Manuscript on file, Columbus Museum of Arts and Sciences, Columbus.
Fuller, J. 1995 Bull Creek Site. Manuscript on file, Columbus Museum.
Hally, D.J. 1986 The Identification of Vessel Function: A Case Study from Northwest Georgia. American Antiquity 51:267-295.
1993 The Territorial Size of Mississippian Chiefdoms. In Archaeology ofEastern Nortn America, Papers in Honor of Stephen Williams, ed. by J. B. Stoltman, pp. 143-168. Archaeological Report No. 25, Mississippi Department of Archives and History, Jackson.
1994 Ocmulgee Archaeology 1936-1986. University of Georgia Press, Athens and London.
Hally, D. J., and J. B. Langford, Jr. 1988 Mississippi Period Archaeology of the Georgia Valley and Ridge Province. University of Georgia, Laboratory of Archaeology Series, Report 25, Athens.
Hally, D.G. and L. Oertel 1977 Archaeological Investigations at the Park Mound (9TP4I) Troup County, Georgia, 1972 Season. Submitted to Heritage Conservation and Recreation Service, National Park Service, Contract No. 14-10-9-900395. University of Georgia, Athens.
Hawkins, B. 1916 Letters of Benjamin Hawkins. Georgia Historical Society Collections, Volume 9, Savannah.
Hodler, T.W. and H.A. Schretter 1986 The Atlas of Georgia. The Institute of Community and Area Development, The University of Georgia, Athens.
Hudson, C. M. 1976 The Southeastern Indians. University of Tennessee Press, Knoxville.
212

Hurt, W.R. 1975 The Preliminary Archaeological Survey of the Chattahoochee Valley Area in Alabama. In Archaeological Salvage inthe Walter F. George Basinofthe Chattahoochee RiverinAlabama, edited by David Dejarnette, Alabama Press, Tuscaloosa.
Huscher, H.A. 1960 Appraisal of Archeological Resources of the Walter F. George Reservoir Area Chattahoochee River, Alabama and Georgia. River Basin Survey Report, Smithsonian Institution.
Jenkins, N.J. 1978 Prehistoric Chronology of the Lower Chattahoochee Valley. Journal ofAlabama Archaeology (24)73-93.
Jennings, J.D. and C.H. Fairbanks 1939 Newsletter of the Southeastern Archaeological Conference, Volume 1 (2). Lexington, Kentucky.
Johnson, J.H. 1983 Soil Survey ofMuscogee County, Georgia. United States Department of Agriculture, Soil Conservation Service.
Jones, c. B.
1982 Southern Cult Manifestations at the Lake Jackson Site, Leon County, Florida: Salvage Excavations of Mound 3. Midcontinental Journal of Archaeology 7:3-44.
1994 The Lake Jackson Mound Complex (8LEl): Stability and Change in Fort Walton Culture. Florida Anthropologist 47(2): 120-146.
Kelly, A.R. 1937 Lamar and Related Site Exploration in Georgia. Paper presented at the second annual meeting of the Society for American Archaeology, New Haven, Connecticut. Manuscript on File (Ace. 125), National Park Service, Tallahassee.
1938 A Preliminary ReportonArcheological Explorations at Macon, Georgia. Smithsonian Institution Bureau of American Ethnology, Bulletin 119, Anthropological Papers, No. 1. Washington.
1950 Report on the Archaeological Possibilities of the Bull Creek Site, Columbus, Georgia, with Recommendations for Development. Unpublished manuscript on file, Columbus Museum.
King, A. 1991 Excavations at MoundB, Etowah: 1954-1958. Master's thesis, Department of Anthropology, University of Georgia, Athens.
King, A., and J. A. Freer 1995 The Mississippian Southeast: A World Systems Perspective. In Native American Interactions: Multiscalar Analysesand Interpretations in theEastern Woodlands, edited by M. Nassaney and K. Sassaman, pp. 338-362. University of Tennessee Press, Knoxville, in press.
Knight, V.J. 1994 Ocmulgee Fields Culture and the Historical Development of Creek Ceramics. In Ocmulgee Archaeology 1936-1986, edited by D.J. Hally, 181-189. University of Georgia Press, Athens.
Knight, V.J. and T.S. Mistovich 1984 Walter F. George Lake: Archaeological Survey of Fee Owned Lands, Alabamaand Georgia. Report of Investigations 42. Office of Archaeological Research, the University of Alabama, Moundville.
213

Ledbetter, R.J. 1994a Archeological Survey of the South Commons. Report prepared for the City of Columbus by Southeastern Archeological Services, Inc., Athens.
1994b Archeological Survey of Riverwalk South Rotary Park to Fort Benning, Georgia. Report prepared for the City of Columbus by Southeastern Archeological Services, Inc., Athens.
1995a A Summary of Radiocarbon Dates for the Averett Phase. The Profile.
1995b Dear Isabel, Archeological Correspondence of A.R. Kelly and Isabel Patterson, 1934-1953. Lamar Institute Publication 33, Lamar Institute.
1995c The Victory Drive Site, Data Recovery of 9Me50, Muscogee County, Georgia. Draft report prepared for the city of Columbus. Southeastern Archeological Services, Inc.
Lester, F. 1936 Letter to A.R. Kelly dated March 15, 1936.
1938 Preliminary Report of Excavation at Bull Creek Village Site, Muscogee County, Georgia. Unpublished manuscript on file Columbus Museum.
Lupold, J.S. and F.T. Schnell, Jr. 1991 Historic Resources at the Falls of the Chattahoochee. Prepared for Uptown Columbus. Columbus.
Martin, J.H. 1874 Columbus, Georgia, from its Selections as a "Trading Town" in 1827 to its Partial Destruction by Wilson's Raid in 1865. John Gilbert, Columbus.
McMichael, E.V. and J.H. Kellar 1960 Archaeological Salvage in the Oliver Basin. University of Georgia Laboratory of Archaeology Series Report No.2, Laboratory of Archaeology, Department of Sociology and Anthropology, University of Georgia, Athens.
Muller, J. D. 1989 The Southern Cult. In The Southeastern Ceremonial Complex: Artifacts and Analysis, edited by P. Galloway, pp. 11-26. University of Nebraska, Lincoln.
Mulvihill, F.J. 1925 Some Indications of Indian Occupancy Along Bull Creek, Near Columbus, Georgia. Arrow Points Volume 10 (3) pp 36-41.
Patterson, I. G. n.d. Notes on the Exploration of the Bull Creek Site, Columbus, Georgia. Unpublished manuscript on file, Columbus Museum of Arts and Sciences, Columbus, Georgia.
1936 Archeological Survey of the Chattahoochee Valley. Paper presented at the 2nd annual meeting of the Society for American Archaeology, Washington, D.C. Manuscript on file, Columbus Museum of Arts and Sciences, Columbus, Georgia.
1937 Bull Creek Village. Unpublished manuscript on file, Columbus Museum of Arts and Sciences, Columbus.
1950 Notes on the Exploration of the Bull Creek Site, Columbus, Georgia. Early Georgia Volume 1 No. 1, pp. 34-40.
214

Plog, S. 1990 Sociopolitical Implications of Stylistic Variation in the American Southwest. In The Uses of Style in Archaeology, edited by M. Conkey and C. Hastorf, pp. 61-72. Cambridge University Press, Cambridge.
Scarry, J.F. 1985 A Proposed Typology for the Mississippian Period Ceramics of the Fort Walton Area: A Type-Variety Approach. Florida Anthropologist 38:199-233.
Schnell, F.T., Jr. 1963 The Bull Creek Site in Western Georgia. Manuscript on file, Columbus Museum.
1970 The Victory Drive Site: An Historic Indian Settlement in the Chattahoochee Valley. Manuscript on file, Columbus Museum of Arts and Science, Columbus.
1973 A Preliminary Assessment of Archeological Resources Remaining in the Walter F. George Lake Area. Manuscript 108 on File, University of Georgia Laboratory of Archaeology.
1979 A Cultural Resources Assessment of a Proposed Park on Bull Creek, Columbus, Georgia. UGA Laboratory of Archaeology Report No. 152.
1985 SGA presentation talk on Bull Creek Phase (May 11, 1985)
1986 Phase Characteristics for the Middle Chattahoochee River. Paper presented at the 1986 Lamar Conference, Macon, Georiga.
1990 Phase Characteristics for the Middle Chattahoochee River, in Lamar Archaeology, Mississippian Chiefdoms in the Deep South, edited by Mark Williams and Gary Shapiro, pp. 67-69, University of Alabama Press, Tuscaloosa and London.
1991 A Cultural Resource Assessment ofthe Proposed Site ofthe River Club, Columbus, Georgia. Manuscript on File, Columbus Museum. Prepared for Jordon, Jones and Goulding, Inc.
Schnell, F.T. Jr., V.J. Knight and G.S. Schnell 1981 Cemochechobee: Archaeology ofa Mississippian Ceremonial Center on the Chattahoochee River. Ripley P. Bullen Monograph in Anthropology and History 3. University of Florida, Gainesville.
Schnell, F.T., Jr., and N.O. Wright, Jr. 1993 Mississippi Period Archaeology of the Georgia Coastal Plain. Georgia Archaeological Research Series Design Paper 3. University of Georgia Laboratory of Archaeology Series Report Number 26, Athens.
Sears, W.H. 1951 Excavations at Kolomoki, Season I. University of Georgia, Series in Anthropology 2. Athens.
1956 Excavations at Kolomoki: Final Report, University of Georgia Series in Anthropology, No.5, Athens.
Shapiro, G. 1990 Bottomlands and Rapids, in Lamar Archaeology, Mississippian Chiefdoms in the Deep South, edited by Mark Williams and Gary Shapiro, pp. 147-162, University of Alabama Press, Tuscaloosa and London.
Smith, M. T., and J. B. Smith 1989 Engraved Shell Masks in North America. Southeastern Archaeology 8(1):9-18.
Snow, F. 1990 Pine Barrens Lamar, in Lamar Archaeology, Mississippian Chiefdoms in the Deep South, edited by Mark Williams and Gary Shapiro, pp. 82-93, University of Alabama Press, Tuscaloosa and London.
215

Sturges, D. 1818 Map of State of Georgia. Published by Eleazer Early. On file, at Surveyor-General's Office, Atlanta, Georgia.
Swanton, J. 1922 Early History of the Creek Indians and Their Neighbors. Bureau of American Ethnology, Bulletin 73, Washington.
Trimble, S. 1974 Man-Induced Soil Erosion on the Southern Piedmont: 1700-1970. Soil Conservation Society of America, Ankey, Iowa.
Tyler, J. and T. Jackson 1969 Report on Clearance ofPothole A, Bull Creek Site. Unpublished manuscript on file, Columbus Museum, Columbus, Georgia.
Walker, J.W. 1989 Ocmulgee Archeology: A Chronology. Southeastern Archeological Center, National Park Service, Tallahassee, Florida.
Weissner, P. 1983 Style and Social Information in Kalahari San Projectile Points. American Antiquity 49 (2):253-276.
1984 Reconsidering the Behavioral Basis for Style: A Case Study among the Kalahari San. Journal of Anthropological Archaeology 3:190-234.
Wharton, C.H. 1978 The Natural Environments of Georgia. Georgia Department of Natural Resources, Atlanta.
Willey, G.R. 1949 Archeology of the Florida Gulf Coast. Smithsonian Miscellaneous Collections 113. Washington.
Williams, S. 1979 Some Negative Painted Pottery in the Southeast. Paper presented at 1979 meeting of the Southeastern Archeological Conference, Atlanta, Georgia.
Wobst, M. 1977 Stylistic Behavior and Information Exchange. In Papersfor the Director: Research Essays in Honor of James B. Griffin, edited by C. E. Cleland, pp. 317-342. Anthropological Papers 67. Museum of Anthropology, University of Michigan, Ann Arbor.
Wood, K.G. 1991 Archeological and Historical Investigations of the Proposed Riverwalk Project, Columbus, Georgia. Prepared for Jordan, Jones and Goulding, Atlanta. Southeastern Archeological Service, Inc., Athens.
1993 An Archeological and Historical Assessment of the Columbus Waterfront from the City Whaif to Oglethorpe Bridge. Prepared for the Columbus Water Works. Southeastern Archeological Services, Inc.
216

APPENDIX A
Transcription of Lester's Field Notes Bull Creek Site
1936

Mar. 12, 1936. Visited and made a sketch map of a Village Site, situated just North of Bull Creek, at its junction with the Chattahoochee River. The site is rather well defined, in that a roadway has been cut through it and the exposed profile shows a very definite midden accumulation, ranging in thickness from about six inches to about sixteen or eighteen inches in the middle, where the exposed bank shows the signs of a house of some type - timbers and quite an accumulation of briquettes. This deposit shows in the bank for a distance of some three to four hundred feet and appears from the exposed surface, to be very similar to the Mossy Oak type site on the Ocmulgee.
The ridge or hummocky places nearby seem to have been caused by a company making a railroad cut and are nothing more than spoil banks.
3-12-36. During the afternoon Mrs. Patterson and I visited a place across the river on the outskirts of Phenoix City, Ala., where a portion of a skeleton had been unearthed - also associated with it were some trinkets, that had the appearance of early made articles. A complete circle to go in the lobe and a little bell-like pendulum below, also some small buckles of pewter.
The hill from which these were taken has been used as a gravel pit and the top has almost entirely been removed to a depth of about 31h feet. The underlying geological formation appears to be a type of red sandy loam, either of the Greenville or Barnesville formation - some indication of mica in it. It is possible that the rise may have a small mound on top but for the most part, at least, it is probable, in fact almost certain, that it is nothing more or less than a natural elevation.
Also visited two sites on the Bussey plantation, in Ala. on the bank of the Chatt. just below the mouth of Bull Creek. Both of these sites had quite a bit of surface pottery and were rather extensive. Supposed to be Coweta towns.
3-13-36. Started excavation on the Bull Creek Site this morning, with a crew of 3 negro boys and myself. As previously mentioned, the site has been cut into by a roadway on the east and the river has made rather deep inroads from the west. The place where work was begun is on this narrow strip of ridge, on a bluff, about 40' above the river.
Control Trench #1 was staked out - trench 5' wide - stks intervalS' - stakes nos. starting at 1-39 and going to 1-52. Worked four sections, the width of the ridge from 1-39 - 1-40 to 1-49 - 1-50. Trench #1 has an orientation of N. 15 deg. W - magnetic.
The top soil is a river silt deposit of yellow sandy loam, containing an unusual amount of mica; it ranges in depth from 2" at sta. 39 to about 10" at sta. 48. No sherds were found in this level. No finds.
Immediately below the yellow top soil is a midden deposit of rather greasy type of "blue clay. " Quite a number of sherds came from this level, however, they were all exceedingly small pieces. The majority of pottery found is plain, hard, grit-tempered ware, rather greyish in color - the pottery also has a rather appreciable amount of mica in it.
Find #1 - a rim-sherd with a raised notched design, about an inch from the top.
Find #2 - a rather peculiarly shaped piece of sandstone, about an inch in diameter, so nearly circular that it might have been used for a chunky stone.
A-I

Immediately below the blue-clay level which was about 2 to 4" thick, is a sort of greyish sandy clay containing quite a no. of briquettes and some sherds. This level appears to be immediately above the baked clay, belonging to the house debris. In both this and preceding level, practically no flints have been found and none with secondary working. Typical profile to date.
3-14-36. Continued excavation on house site, exposing 10' of red sandy clay mixture, presumably roof accumulation, between sta. 1-45; 1-46 and 1-49; 1-50. Area south of site 1-39; 1-40 is of a grayish sand clay-midden deposit. It appears that the scattered edge of the house site is directly under sta. 1-46; in section 1-45; 1-46 to 1-49; 150. A baked clay ridge was exposed, I" deep in red level, this ridge appears to be floor material, and is shaped something on the order of a fire bowl, except that it has a very small degree of curvature. Further troweling should prove more enlightening on this.
The sherds from this and the immediately adjacent level of greyish clay, are very abundant much larger pieces than previously noted.
Immediately adjacent to the bake ridge mentioned above, are other baked areas appearing at irregular intervals and slightly different levels. It is probable that all of these belong to the wall and roof structure. Quite a number of briquettes, showing reed imprint, were also found in this level.
Find #3
3-16-36. Removed yellow silt deposit and portion of black midden deposit from section 1-49; 1-50 to 157; 1-58. Worked black midden to a depth of 1 to 3 in. - sherds plentiful.
Find #5 - a piece of briquette, showing reed imprint, found one inch in reddish, sandy clay of
house debris - 32" S of sta, 1-49 and 21" west.
Find #6 - a rim-sherd, showing a notched design on top of the rim, an incised line just beneath it and a portion of an incised whirl, showing punctate marks in it.
3-17-36. Excavated black, midden deposit from sta. 1-49; 1-50 to sta. 1-57; 1-58, to a depth of 14 to 30 inches. Quite a large number of sherds in this level. Also started excavation of reddish sand and clay (house debris) sta. 1-46; 1-45 to 1-52; 1-51. Briquettes, charcoal and pottery are scattered throuout this level.
Note for 3-17-36. Found a small piece of bone, 3" deep in reddish sand and clay, - extreme dampness made it impossible to take up. Presumable leg bone of deer, one end of which had been charred. Large "pores" in knuckle section. 29" S of sta. 1-50 to 6" W.
3-18-36. Continued troweling house site, uncovering quite a large section of briquette material between sta. 1-49 and 1-50, irregular in shape and thickness. Quite a quantity of charcoal was evident, but no timbers except a short section between sta. 1-49 and 1-51, nine to sixteen inches N 1-49 - 6" E. 11" below surface, in reddish sand and clay. Presumably a section of matting of reed with a round pine pole. Another short section 10" long was found 21" S of sta. 1-51, II" E, II" below surface; this appears to be a short section of a log, about 4" in diameter.
3-19-36. Further excavation of house debris continues to show briquette material in abundance. Started troweling floor section and uncovered a piece of broken pottery, laying flat on the floor, with a few
A-2

fragments of human bones in it. This is located just on the edge of the bank (road side) 19": S - 16" E of sta. 1-50. Immediately adjacent to this (26" S- 12" E sta. 1-50) is a portion of a small timber with a large briquette moulded around it.
3-20-36. Started excavation of a section between 1-39; 1-39L1 and 1-57; 1-57L1, working yellow silt top soil, and a portion of the black midden underneath. Sherds plentiful in the midden level. Also continued troweling on floor, uncovering quantities of briquette material and charcoal. Bad weather prevented much headway.
3-23-36. Excavation of black midden deposit 1-39; 1-39L1 to 1-57; 1-57Ll. Quite a number ofsherds, Find #11, a broken rimsherd with a very small handle found 10" deep in black midden deposit, 15" below surface, 22" S - 30" W sta. 1-49.
Between Sta. 1-45 and 1-47 a distinct line of red is observed presumably the west edge of the house debris.
Continued troweling the floor, struck a small portion of a timber about 4" long, 28" S and 3" W Sta. 1-52, lying flat on the floor, also a moulded briquette, 40" Sand 15" E Sta. 1-52.
3-25-36. Rain! - no work. Made arrangements with postmaster for storage space in new P.O. and possible office space in old post office.
3-26-36 more rain and Friday.
Sat., March 28, 1936. Continued work on Bull Creek #1 house site, work out floor. Also visited and made surface collection from Coweta #3 and site at Abercrombie Mound, located on Mr. Brannen's farm, on the Ala. side of the Chattahoochee, about 3 miles below Bull Creek. Visitors in town, Dr. Kelly, Hanna, Solomon and Tony Waring.
April 6, 1936. Started N.Y.A. crew of light excavating, a 5' test trench, from sra. 1-87; 187Ll to Sta.
1-123; 1-123L1. This section lies just north of the house site already described, and shows a slight ridge running through it. Removed yellow topsoil from the whole section, topsoil ranging from 4 to 20" inches. Very few sherds, one Find (#12), a quartz arrowhead of the modified spinner type, found between Sta.
April 7th, 1936. Rain - no work
April 8th, 1936. Continued work on test trench. Working from yellow topsoil into a very rich black midden deposit. Quantities of sherds were found in this level. 5 finds (#13 to 17) Quite a number of briquettes were found in the level and it is probable that there is a portion of a house between 1-93 and 1-107. It appears that the midden deposits runs out between 1-107 and 1-111, grading into a yellow subsoil.
Midden deposit also contains things of late historical occupancy; mixture probably due to plowing; bricks, glass and other articles of such nature, mixed in with pottery fragments and briquette material.
April 9th, 10th, 11th. Rain!
A-3

April 13. Unable to work in L1 Test Trench, due to wet weather, started excavation of 5' Test Tr. E L3 line stakes from Sta. 1-95L5 to sta. 1-111L5, worked through yellow topsoil and into black midden to a depth of 9" at 1-95L5 and 13" at sta. 1-111L5, yellow topsoil an overage of 4" deep. Sherds plentiful- 3 Finds (# 18,19,20) at a depth of 13" below surface at Sta. 1-99L3, briquettes became fairly numerous between sta. 1-97L3; 1-97L5 and Sta. 1-99L3; 1-99L5. Found the pieces of a small pot in place and it will be possible to reconstruct in original form. Found 29" S - 8" W Sta. 1-99L3, also found a complete rim, pinched or beaded design, 1 ft in diameter, found 13" S - 22" W Sta. 1-99L3. Started removing section between l-L1 and 1-L3 between sta. 1-95 and 1-103, to expose house debris. Sherds plentiful in both levels.
4/14/36. New crew of20 men. Extended Test Trench ofL3, L5 to sta. 1-143L5. Worked to a depth of 13" at 1-111L5 to 16" of 1-143L5. It is noted that the black midden deposit appears to run out between 1-115L5 and 1-119L5. being replaced by yellowish clay loam. Very few sherds found in clay loam area. Also started excavation of trench Lll, L13, starting at 1-109Ll3 worked black humus and midden deposit to a depth of 9", worked to 1-143L13 to a depth of 5" Black midden deposit appears to run out at Sta. 1-127Ll3. Yellowish clay appears to extend beneath it, back to 1-107L13. Found outline of a pit in yellow clay deposit, just below black midden deposit. Find #21 found in pit just above burned area.
April 15-16. Continued excavations of area between sta. 1-91Ll - 1-91L5 to 1-123L1- L5. Briquette material appearing at a depth of IS" below surface or 1-91Ll and extending from 1-91L1 - L5 to 1103L1 - L5. Found a pot, broken, but pieces in place, found 28" from 1-99L3, 59" - sta. 1-99L5 - 44" sta. 1-99L3. Scattered fragments in immediate vicinity, appear to belong to these two pots. Portion of another pot was found 65" SE 1-97L1, 41" NE 1-95Ll. Pot is of plain type ware, resting rim down, surrounded by briquette material. There seems to be an absence of the huge moulded type of briquettes found in the first house.
April 20, 21 -'36. Worked area between sta. 1-91L5 -L9 to 1-103L5 -L9, down to a depth of 13" below surface, going through top soil and into black midden deposit, also worked a portion of house debris striking floor between 1-99L1 and 1-99L5. Briquette material shows from 1-95 to 1-95L5 to 1-103L1; 1-1-99L5. Briquettes seem to be of moulded variety but do not show as clearly as in house #1, because of plowing through them.
Total lack of timbers is also probably due to plowing. Several large pieces of pottery were found just on and above floor level. There is a total absence of trade material but modern brick are found mixed all through black midden deposit. See find #34.
4-22,23,24 -36. Staked off Con. #2, across R.R. cut, from sta. 2-55-56 to 2-91-92. Worked through topsoil and into yellow sandy clay-loam deposit underneath. No heavy accumulation of midden deposit, but sherds rather well scattered throughout plowed soil and down into yellow loam. Found a pit in Con. #2 at Sta. 2-59, showing up just under plowed soil.
Two boxes of pottery have come from this pit (See Pit #2 material), several pieces show a rather unusual design of punctate, incised body, with notched rim. At present the pit has not been completely troweled out.
Also started removing briquette material from floor and have found several large pieces of pottery resting just above floor and presumably belong to some of the pits already described. No timber supports as yet but briquettes show more of the massive size of the first house. Also removed topsoil and a
A-4

portion of black midden deposit from section between 1-103L5 -L9 and 1-115L5 -L9, worked to a depth of 6" below surface. Several boxes of pottery from this section.
The briquettes material shows use of fiber for binding purposes and there seems to be an absence of the matted reed construction -- on the whole it is very similar to the other house except that it has a harder clay floor, either due to packing or baking to a small extent. The briquettes shell off the floor proper very easily and it is thus easy to work out.
April 27 - May 4. Continued to remove briquette material from what appears to be floor of house site #2. Portions of floor appear to be intact and other portions (in can #1 - Ll) show signs of having been plowed into. Small brick fragments and occasional pieces of metal show evidence of having been plowed under. At present no post mould impressions are showing.
Two burials have been located in the side of the bank, about 100 yards from where Bull Creek runs into the Chattahoochee River. One has been washed from place and has fallen into the yellow clay muck down below but the other is still in place; with a portion of the lower leg bones and thigh bones sticking out of the bank it seems that both the feet and the pelvis bone have been washed away. In the first - the skull was removed from position by the person who found it before notifying me.
May 5-6. Continued to work on house site #2. On the whole it seems to have been plowed into rather badly; very little evidences of any floor material except that already described. Working plowed soil off down to a yellowish clay loam, showing signs of briquettes and small pottery fragments. A few burned areas have turned up that may possibly develop into post-holes.
Excavated a 10 X 10 pit over the burial in the bank of Bull Creek and worked down to it to a depth of below the surface. On the whole the body is in good state of preservation, but the skull is broken in on the left side -- pelvis bone and feet are gone. Troweling out has developed that it is a flexed burial -- male (?) skeleton -The body is flexed, knees drawn up almost under the chin, and arms folded, with the left hand between the legs. All of the upper vertebrae are gone but some of the ribs remain in place.
May 7-18. Continued work on House # 2, with very small group of workers. Removed pottery from floor (?) level and continued on down. Ground so hard that necessary to keep wet all the time. Floor appears to be very uneven, sloping toward the southwest. Found baked area in S.W. corner at least 3" below level of that in N.E. corner. Still finding some pottery in floor level and traces of briquette material. Very few post moulds have turned up at the present time. Floor area is very uneven, showing patches of baked material in spots. All indications point to use of hard packed clay floor, with patched places baked in place.
May 18-22. More work on first house site, uncovering floor level between 46-45 and 52-51. Removing briquette material from hard-packed blue clay. Made a plaster mould of a large molded briquette. Started working midden deposit beyond immediate confines of house site, just below level of floor. The underlying clay seems to slope down just beyond house site. Found a large refuse pit, containing quite a bit of pottery in flooring location.
Also found some small calcined bone on floor of house site between Sta. 1-47 - 1-48 to 1-49- 50. (Above drawing) Bones are very tiny and have been burned completely through. More post moulds are turning up in this house and should present a very nice architectural floor plan.
A-5

May 25-26. Continued work on #1 house site; more evidence of post mould impressions. Removed moulded briquette from floor of house. Find #? Very little pottery showing up on house floor. Floor stilI shows hard-packed clay variety, with layer of charcoal just over floor. What post remain in holes are partially charred, but what fiber remains uncharred is not totally rotted. It appears that the posts charred, all around the outside and the portion stilI remaining in natural state is the center of the charred portions. See drawing of house for more accurate location of individual post-moulds.
May 27-June 2. Lowered 15 X 15 pit at burial on down to level of burial #2 and found evidence of another burial pit in north wall. Then extended pit 10' to the north and worked on down, finding pottery in the waterlaid sand. At a depth of approximately 5', an effigy pot was found made in the shape of a dog - painted ware. The pot is unbroken and is an exceptional piece of ware. The "dog" is complete, with eyes, nose, ears, four feet and short stubby tail. The pot is part of the furniture with a burial, found just below the pot. The body is face downward, and have only found one arm.
June 2-8. Continued excavation on burial # 3, working down to remainder of skeleton. Sherds occur in all of fill element down to burial and some small pieces right in place with skeleton. Body is laying on the left side, head pointing South - contracted type of flexed burial - knees drawn up almost under the chin, with the pot resting between knees and chin. The hands are between the lower leg bones, right hand about 3th " below the knee-cap and the left hand about 10" above the head. Location Burial #3 between Sta. 4-14 and 4-13.
6-8 - 6-9. Started 10 X 10 pit, 10' N of burials #3 and 4, worked down to a depth of 7"; no sherds found in level. Found a skull of a dog 7" below surface, 26" N Sta. "A". Started form for removal of burial # 3 as a complete unit.
6-10 - 6-12. Lowered 10 X 10 pit to a depth of 31". Top level w-l, sand 3" to 11" deep. Old sod (historic) lies 6" deep - accumulated sand - 14" deep. Bone sherds in Band C levels. Also removed Burial #3 intact, by pouring concrete block around burial and forcing 3/8" steel plate under concrete. The body was placed on exhibit at M & M Bank and has caused quite a bit of favorable comment. Location of Burial # 3 -- 4-
6-15 to 6-22. In extending pit Northward, to remove burial #3, also ran into another burial, which was rather badly cut by the person making the discovery; the burial was found at a depth of 44" below surface and only 20" beneath bottom of occupation level.
The body is laying on the left side, head facing N.W., body oriented N.E. and S.W. Such bones as remain are in very good state of preservation, with exception of fingers, toes and ribs. The skull, pelvis and sections of the vertebrae were cut into with the shovel but the rest of the body is intact.

I-$~ r-""r.nr---i1-541
Found 29" below surface at 1-58

e / /_.r.~~~-----J/-.$&
\

A-6

1--'17

~-/~ . . . . .~
.:.~ .I
I
I i
;

1-/3

i-- 4-/5LI
I
I
I I
!

. _ - - - - - - - - - - - - -4-,/.

Field Drawing of Burials 3 and 4. A-7

;
I
. _....__._. Ji
'/

The body is on the left side, arms straight down toward the thighs, with legs folded back under the body - knees in front with heels just below the pelvis bone - slightly different from the other type of flexed burial. As the hand bones are gone, it is impossible to tell whether or not they were between the legs but it seems highly improbable, as the distance is too great from elbow to the thigh bones. The bones are comparatively small and it appears to be the skeleton of a woman. Location Burial # 4 (between Sta. 4-15Ll and 4-15).
Lowered 10 X 10 pit, sta. 4-20 to 4-24 - 4-19Ll 60 4-23Ll, to a depth of 78" at 4-23Ll, 64" at 4-20. Sherds plentiful to bottom of occupation level. Found a burial just under sta. 4-21Ll, in West profile wall, and extended pit to 4-19L3 to 4-23L3, working down to burial pit. A greenstone celt was found with the burial, resting just back of the head. Skull practically gone and remainder of bones in very poor state of preservation. Details later. Also worked section between 4-26 - 4-20 to 4-15Ll - 419L1 to a depth of 64" at 4-20. Found a burial in the wall on present level - (64") just under Sta. 417Ll. Also found a burial in wall 24" W. Sta. 4-19Ll Details later.
6-23 -6-24-1936. Rain!
6-25. Troweled out burial # 5. Burial is of the contracted, flexed type. Body laying on left side, head pointing SSE. Bones in a very poor state of preservation. Burial furniture consists of a greenstone celt, approx. 7" long and 3" wide at cutting end - 2" wide at handle end. All of the skull is gone, except the upper and lower teeth; no trace of shoulder bones and only the 3 lower joints of vertebrae. Right arm folded across the body, left arm going between the thigh bones, with hand at heels. Burial is in tan sand, 36" below bottom of occupation level, 72" below surface at sta. 4-2IL3. Found quite a few sherds in pit fill. (See box of section.)
Also lowered 5' X 10' pit, 4-15Ll to 4-19LI-L3, to a depth of32"; sherds plentiful in occupation level.
June 29. Lowered section 4-15Ll -L3 to 4-19Ll -L3, to a depth of? below surface; found 4 burials in section; one, #8, directly under 4-19L3; one, #7, on line of sta 4-19L1 -L3; one, #9, between 4-17Ll and 4-15Ll; and one, #6, on line of sta. 4-17Ll -L3. Detailed locations later.
June 30th to July 3rd. Burial # 7 has with it another of the dog effigy painted pots. This pot however, is rather badly broken, the neck being found down on the pot itself, and the head and three of the feet broken off - all parts are in place though, and the vessel can be restored to its original shape. From all appearances the pot is an exact duplicate of the one found with burial #3. There is also a small greenstone discoidal stone as burial furniture; the stone is perfectly smooth and round, about 1%" in diameter and about 5/8" thick, in the middle. Removed the pot after taking photos because of danger of theft. Greenstone celt with Burial #5 having been stolen. Location and description of burial later (after further troweling).
Burial # 7 is of contracted, flexed type bones, in very poor state of preservation; body laying on left side, oriented N-S, with head to south. Skull has been crushed and remaining bones consist of sections of tibiae, femora and upper arm bones. Appears that hands went between the legs as in other burials but all trace of bones gone.
Burial # 8 is of a flexed type, with feet drawn up under the body, knees pointing away from body. Bones in rather poor state of preservation, not completely troweled out present time.
A-8

July 6th to July 8th. Moving surplus dirt back to R.R. fill. Started excavation of section 4-7L3 to 415L3. Lowered to a depth of 35" at 4-15L3 21" at 4-7L3. Sherds plentiful in occupation level. There is what appears to be the outline of a burial pit between 4-15L3 -L1 and 4-11L3 L1, Burial #10 first appeared at 20" below surface at 4-11L3. Several pieces of painted pottery are noted in the sherds coming from this level. Also staked out extensions of Control #4 and ran traverse lines from Con. #1 to Con. #4
July 9. No Crew!
July 10. Continued excavation on section 4-7L3 to 4-15L3. Lowered to a depth of? at 4-7L3; at 415L3 Outline of a burial pit shows between 4-7L3 and 4-11. Found Burial #11 at a depth of? below surface No burial furniture.
July 13-14-15. Lowered section Sta. 4-16 to 4-16R1 - 4-24 - 424Rl to a depth of 24" below surface, also removed large tree between sta. 4-16 and 4-16R1. Found bones of large animal 10" below surface, with a small seed bead resting with bones (Find #45). Lowered section 4-7L3 -L5 to 4-23L3 L5 to a depth of 4" at 4-L5 and 6" at Sta. 4-23L5. Note continued rain has caused considerable damage to burial 4-5-6-7-8-9-10-11, but bones can probably be salvaged.
July 17, 1936. Continued excavation on section 4-7L3 -L5 to 4-23L3 -L5, to 4-23L3-L5, working to a depth of 36" at Sta. 4-23L5, 9" at sta. 4-11L5, working through occupation level from 4-15L5 to 423L5, approx. 4" in tan sand. Considerable pottery from occupation level. Three finds #'s 46,47,48. Only 3 in crew.
July 20, 1936. Continued excvation in section 4-7L3 -L5 to 4-23L3 -L5, working through occupation level and into tan sand, working to a depth of? below surface. Pottery fairly well distributed throughout the level; so far no trace of burial pits as yet. Also continued section 4-16 to 4-24, working tan sand to a depth of? below surface. Find #49, a small clay bead found 5" deep in tan sand, 40" below surface, 11" E 4-18, on line sta. Find #50 a rimsherd of unusual design, found 6" deep in tan sand, 40" below surface, 35" S. Sta. 4-18, also found an outline of a burial pit between sta.
July 21, 22, 23, 1936. Lowered section 4-16 -4-16Rl to 4-24 -4-24Ll to a depth of 4.20 Sherds plentiful in occupational level. Found a burial #13, located at a depth of below the surface, between Stations 4-16 -4-16Ll and 4-18 4-18Ll. Not troweled out at present; skull broken rather badly when found. Also found a burial #14, between stations 4-18 -418Rl and 4-20-4 20Rl; bones in excellent state of preservation. Body lying on left side oriented E & W with skull pointing E. Body is in the contracted, flexed position with knees drawn up almost under the chin, with hands passing between the thighs and lower legs. Skull has peculiar indentation on back of occipital, unnatural deformation - none of the ribs remain, however a few sections of the vertebrae remain in place; pelvis bone in fair state of preservation and a few bones of the feet are still in place, showing arch of foot and upper toes. Skull is also enlarged on right side, being slightly misshapen.
Began troweling out Burial #6. Burial is offlexed type laying on right side -- Body oriented NWSE, head to NW with knees pointed away from body and feet drawn up under the pelvis bone. Arms go straight down at side and fingers nearly reach to femur. Body in rather poor state of preservation, having been considerably damaged after discovery, by heavy rains. Burial furniture consists of a small green celt, about 41h" long, 21h" at cutting end and I" at butt end; two small green stone chisels. A very nice flint knife of black flint, some small pieces of worked flint and some very peculiar reddish powder in globule form, apparently some ferrous oxide, that was possibly used for paint. The flint articles and the greenstone chisels are resting on the "paint" material. 67Ih below sur. 36" in tan sand.
A-9

Burial #7 is in a very poor state of preservation, nothing remaining except small section of fibula of right arm, sections of both femur and fragments of tibia. From appearances it seems that body was laying on left side, oriented SE-NW, head to S.E. and in all probability the hands went between the legs; only small section of? and radius of right arm found. 78" below surface 47" in tan sand
July 27 to 30. July 27 Moved skeleton from M&M Bank. July 28 to 30 Troweled out burials exposed and began cleaning excavations. Found a dog-effigy type 7" S. sta. 4-15L5, 32" E. 70" below surface, 36" deep in tan sand. Pot is slightly different from other two in that neck design is slightly different, neck being higher and smaller. Body of pot painted red with black design. Vessel in perfect state of preservation except that end of dog's tail broken off and one leg loose, but in place. The legs are slightly longer and place at a slightly different angle.
July 31 -- Rain!. August 3,4,5. Complete troweling of burials 8. 14, also started excavation of section 4-24 - 4-23L5 to 4-28 - 4-27L5.
August 10. Continued excavation of section 4-24; 4-23L5 to 4-28; 4-27L5 lowered to a depth of 52" at sta. 4-27L5 and 29" at sta. 4-28. Sherds plentiful through occupation level.
Also troweled out burial #11 between sta. 4-15L3 -L5 and 4-19L3 -L5. Body is of the flexed type, head pointing SSE. Arm straight at sides and going about midway between pelvis and knee. Thighs point straight downward with legs back toward pelvis. Bones in fair state of preservation, a few of the ribs still remaining intact, also some of the bones of the arch of the foot. No burial furniture. The skull is rather small and to all appearances this is the burial of a female.
August 11 to 14. Continued excavation of section 4-24; Rl 4-23L5 to 4-28; 4-27L5, lowered to a depth of 83" below surface at sta. 4-27L5. Located a burial. #18, immediately below sta. 4-24. Burial not completely troweled out, but burial furniture consists of effigy pipe, Find #64 A, core of 63 conch shell and large beads, 64B, presumably of seed. 65, and a small pot 3 3/4 in. dia. incised and punctate. Burial found 69" below surface. Also removed Burials #11. 5, 6, 8, 9.
Burial #8 is of the contracted, flexed type. Body lying on left side, oriented E-W with head to East. Skull in fair state of preservation, vertebrae of neck and portion of shoulder blade and collar bones remaining intact; left arm folded across body, hand at pelvis bone, right arm straight, going between thighs, hand bones under right tibia. Pelvis, tibia, femur, fibula and bones of feet in excellent state of preservation. Three sections of lower vertebrae remaining intact. No burial furniture. Burial located just under sta. 4-19L3. 44" below occu. in tan sand.
Burial #9 is of contracted, flexed type, bones in poor state of preservation. Oriented SE-NW on left side. Nothing remains of upper body except teeth, vertebrae of neck, section of upper arm bone about 4" long. Lower body remains - small section of pelvis bone, tibia, femur, fibula and small section
of hand between femur. Burial furniture consists of small greenstone chisel, approx. 3 lh " long (Find
#56) Burial located 66" below surface 32" in tan sand between sta. 4-15Ll -L3 and 4-17Ll- L3.
Burial #12 consists of small fragments of skull and portion of tibia and femur of both legs. Bones in very poor state of preservation. Burial is of contracted, flexed type, knees in front of body, oriented NW-SE Head to NW. Found between sta. 4-9 - 4-9Ll and 4-11 - 411Ll at a depth of 35" below occupation level. Body laying rt side.
Burial #10 consists of broken fragments of a skull, all other traces of burial gone. Skull crushed beyond repair, burial located between sta. 4-15Ll -L3 4-13Ll -L3. Found 27" below occupation level in tan sand 54" below surface.
A-lO

t- I
EI!j(/Y Pof ~f"

I- '~ 4-/f t:

D/sctJrdal fto)'Je ..9 N'yY J{ 11)/.

JI.-/1 J-/

;;r).L~ .-0--:----.-. ---_.------ _ _~--------J.~-7/ J-/
. - ._. l- -- ..- - - - -
J-ocaJ;~" aLl rJ;~( :# S I

~I 1;6 ~\

.....J

~.lp o"t-:;

::::

Field Drawings of Burials 7 and 8. A-ll

I
I
I
I
1
I
1I 4'
r
_.....
';'-17'- /

. ~_ ...
N

i
1.5' p
~ I 4-- :J.() J
_II'~~~

4--/.5"1- 5 ----I
I
+-/71-~

Field Drawings of Burials 9 and 14.

A-12

Burial #13 consists of the fragments of a skull, found immediately below large tree - roots have destroyed all evidence of other bones; skull crushed. From appearances it appears that body was lying on left side, oriented N-S with head to S. Only a few fragments of the skull - some teeth and a few sections of the cervical vertebrae, small section of collar bone. No trace of burial pit - no burial furniture. Found 26" in tan sand, 40" below surface.

Burial #15 is in a very poor state of preservation, body laying on left side, oriented NNW -SSE. The burial is of the contracted, flexed type, arms appear to go below the legs, however bones in such poor state of preservation it is almost impossible to determine exact relation. Remaining bones consist of skull rather badly crushed frontal bone, decayed entirely, short section of both humerus, trace of ulnor and radious of right arm, pelvis bone almost entirely gone, and only short section of femur and tibia, no trace of fibulas or any of bones of appendages.

Burial #16 consists of a few fragments of tibia and femora, bones are scattered over a large area and this is really the remains of two bodies; general orientation is south southwest and north northeast. Burial furniture consists of the 3rd effigy dog vessel. 36" below occupation level.

Burial #20 is of the flexed type, vertebrae straight, knees at right angles to vertebrae. Bones in fair state of preservation; remaining bones are skull, cervalic vertebrae, both femora, portion of both tibia, small section of pelvis, portions of collar bones and shoulder blades. None of lower vertebrae, ribs, hands or feet. Body laying on left side, oriented ESE and WNW, head to the ESE. No sign of burial pit. Arms folded across body - folded at elbow. Burial furniture consists of beads around the neck.

Burial #18 is located directly under sta. 4-24. Bones in very poor state of preservation, only portions remaining - very small section of skull, a few of the molars and both femurs. Body appears to be laying on left side, oriented SE-NW, head to SE. Knees drawn up toward chin, and appears to be typical Bull Co. type of flexed burial. Burial furniture consists of a small pot, Find #65, approx. 31h dia., incised and punctate design, rather poor workmanship; several large beads, Find 64B, of shell or seed, and unbroken clay pipe, Find 64A, human effigy type, and a small piece of conch shell core Find #63.

Burial #19 is of the contracted flexed type, body oriented SE-NW, head to SE laying on right

side. Bones in fair state of preservation. This burial is contracted more than any previously found - very

good illustration of flexed type, with arms going between the legs. All bones remain, except that skull

has been crushed and is rather badly decayed. None of the cervalic vertebrae remain, but all of the

thoracic together with pelvic and pelvis bones. The arms are straight to elbow and then are flexed to go

between the legs. Female burial. Complete overall length of burial, 31". No burial furniture. Burial

pit first appears at top of yellow clay loam. Burial found 67" below surface, 45" below occupation level.

14" in yellow clay loam.

.

Burial #17 is of the contracted flexed type, oriented NW-SE head to NW. Body laying on right side with the knees at right angles to the vertebrae and both arms going between the legs. Skeletal remains in rather poor state of preservation - one side of skull crushed and all of metacarpal and metatarsal bones gone. Only a few joints of the lumbar vertebrae remain but 5 of the thoric ribs. Portion of skull was all that could be removed.

Sept. 10 and 11, 1936. Moved dump started backfilling hole and excavating 4-7L5 -L7 to 4-27L5 -L7,
a also worked section 4-14 - 4-14Rl to 4-20 -4-20Rl down to clay loam, located burial (#22) between
sta. 4-16; 4-16Rl and 4-18; 4-18Rl almost immediately below the location of burial #13. The body is

A-13

Bvy/~I #/0

/f-/.s-J..3

.IJ./~L./

J ).7
6-~-

__ - - - - .-.---_... ._-~./~ 1-/

,-----/.r L . r

-If"- 3

----------------------"'---.

t

l
Field Drawings of Burials 19, 10, and 16.
A-14

~__.___. ./

about 6" in yellow clay loam and the burial pit first showed about 6" above the loam in tan sand. Located another burial (#24) almost under sta. 4-7L1 in tan sand. Details on both of these burials later.
Sept. 14 and 15. Began excavation on section 4-7L5 -L7 to 4-27L5 -L7. Worked through water-laid sand, yellow clay, historic occupation, tan sand, and into Indian occupation level. Very few sherds in upper levels - rather plentiful in tan sand and occupation level.
Also began troweling out Burial #22. The body is laying on the right side, oriented N.W. and S.E. with head to N.W. The skull has been crushed on the left side but otherwise the body is in an excellent state of preservation. The pelvis is slightly higher than the head and has been slightly displaced by root growth but is in and excellent state of preservation. Remaining bones consist of - skull, in fair state of preservation - clavicles - scapular - both humeri - ulna- radii - a few of metacarpal bones practically all of the vertebrae - tibiae - femora - fibulae - metatarsal bones and phalanges. Found 54" below surface 40" in tan sand.
Sept. 16 to 22. Continued excavation on section between 4-7L5 - L7 and 4-27L5-L7, working down into yellow loam. Located 5 burials in section. Burial #25 between 4-9L5 -L7 and 4-11L5 -L7. Burial #26 between 4-11L5 -L7 and 4-13L5 -L7. Burials #27 and 28 on line sra. between 4-15L5 -L7. Burial #29 between 4-17L5-L7 and 4-19L5 -L7. Also troweled out Burial #17 - see notes under Burial # 17 heading.
Sept. 22 to Oct. 14. This period has been marred by much bad weather, rain has prevented the working out of burials 23 to 29. Excavation was begun on the section L7-L9 and have worked section from 4IlL7 -L9 to 4-27L7 -L9 to a depth of 75" at sta. 4-27L7 -L9. Burials #30 31 and 32 have been located between 4-7L7 -L9 and 4-11L7 -L9. At the present time it appears that portions of the Burials are in the unexcavated section between L9 and L11. Burial #31 is a triple burial, found between sta. 4-9L7 -L9 and 4-11L7 -L9. These remains are in a very poor state of preservation and it is impossible to determine the exact relationship of the individual bones. It appears, however, that at least two of the bodies were flexed in the typical Bull Creek manner and the other burial was of the skull only. All of the bodies were oriented SE and NW, with the heads to the NW and were buried with the face upward, heretofore in the majority of cases the head has been turned to one side or the other. Also the skull (a) (buried without the body), highest up, is gone with the exception of portions of the back occipital and lower jaw. All that remains of (b), the next, is portions of the complete skull - frontal bones and facial bones missing and a short section of the humerus (c) consists of portions of the upper and lower jaw - traces of both humeric and radii and sections of both tibiae, fibulae, and femora. No burial furniture.
The profile at this point shows a band of clay from 2" to 6" thick between sta. 4-17L9 and 423L9, suggesting a house floor. There is also some trace of a sod development below the general occupation level, though it shows no definite form at this particular point.
No date. Burial #24 consists of broken portions of the skull and fragments of the humeri, femora and tibiae. The body left side, oriented NW-SE with the head to the SE. No burial furniture. It appears that this is the contracted type of flexed burial. Found 28" in tan sand - 59" below surface. 4-7L1
Burial #25 is in a very poor state of preservation. Body laying on right side, oriented NW and SE, with head to NW. Remaining bones consist of portions of skull and fragments of femora - tibiae and fibulae.
From position of femora and tibiae it appears that the legs were doubled back under the body with the knees in front. 52" below surface - 21" in tan sand.
A-IS

JfS 1/ l:;)iZ/OTY (...IG .....-r -.- .
I l . / ' l _ ..I J AI""''7'''7''7

r - ---_ __--- 4-/8 L oca,L/on 0/ Ovr/Q/#.22 l+= 7 r

."f-/8R/

~

~

"

~

)/

t

. nn r~\

.

~~

~

i
i
.J
f-/~R(

L ocaf/O/7 Q/' 8wr/QU *30-3/

-

~~9

. > : c=d

!JurI '0/#.30

L/

~c::>

Field Drawings of Burials 22, 30-31. A-16

Burial #26 is an extended flexed burial - laying on the left side, oriented Nand S with the head to the south. Skeletal remains in very poor state of preservation - consists of fragments of the skull and portions of tibiae and femora. Burial is located just under sta. 4-13L7, 79" below surface, 41" below occupation level in tan sand.
Burials #27 and 28
Burial #27 is located between sta. 4-13L5 L7 and 4-17L5 L7, 75" below the surface, 41" below the bottom of the occupation level. The remains are in a very poor state of preservation, but it appears that the burial is the typical type of Bull Cr. flexed burial, with the knees drawn up toward the face. Remaining bones consist of fragments of the skull and of the tibiae and femora. Oriented SSE - NNW with head to SSE. No burial furniture.
Burial #28 is located only a foot away from #27, 80" below surface, 46" below surface - 5" in reddish brown loam. The body is laying on the back, flexed to the left - oriented ESE & WNW with the head to the ESE. Remains in very poor state of preservation - Remaining bones consisting of portion of left tibia and femur and skull with all of facial bones gone. This is a female burial. Burial furniture consists of a small amount of red pigment 12" N - 9" E 4-15L7. Find #65 on drawing - a piece of mussel shell 9": E and 16" N. #61, a piece of worked bone fragment, presumably an awl. # 62, same location as 61 and 63, a group of animal bones on top of 61 and 62; 64-A small rock, apparently the same as the paint pigment.
Burial #29 is located between 4-17L5 L7 and 4-19L5 L7, 68" below surface - 47" below occupation level, 6" in reddish clay loam. Bones in very poor state of preservation - remaining bones consist of sections of tibiae, femora and a section of the right humerus. Body oriented SE and NW, with head to SE. No trace of skull. Burial is a typical flexed type. Burial furniture consists of a ball of red paint pigment.
Burial #32 is the contracted type of flexed burial, laying on the right side, oriented Nand S with the head to the N. The remains are in a rather poor state of preservation. Remaining bones consist of portions of the skull; tibiae, femora - humeri and ulna.
Burial furniture consists of several rocks - flint chips and fragments of animal bone all placed at the neck. Body is 52" below the surface, 34" in tan sand. (Btw. sta, 4-7L9 -L7 and 4-5L9 -L7)
Oct. 14 to Nov. 11. Burial #33 was located in yellow loam between sta. 4-3L3 and 4-3L5, when lowering section for drainage. The burial is in a very poor state of preservation, laying on the right side, oriented N and S, with the head to the south. Remaining bones consist of portions of skull and teeth and the leg bones; from the position of these bones it appears that it was the contracted type of flexed burial. No burial furniture. 36" below surface, 7" in yellow clay loam, 29" below occupation level.
Nov. 11 to Dec. 9. Worked section from sta. 4-lL9 -L11 -L13 to 4-17L9 -L11 -113. Found a burial #34 between sta. 4-Ll1 and 4-L13. Worked Lll -L13 down to bottom of occupation level. Also troweled out Burial #30. Located another burial just at edge of what appears to be a house. Several finds from this section.
Burial #30 is laying on the right side, oriented NW and SE, with head to the NW. Remaining bones consist of a portion of the skull which is in a very poor state of preservation, and portions of both tibiae and femora and small fragments of the pelvis bone. Body found 46" below surface, II" below the occupation level in tan sand. The legs are flexed back under the body with the knees the farthest away from the body of any part of it. Bull Creek type #2. (at sta. 4-11L9)
A-17

4-1$1.5 'f
I
..". "
Field Drawings of Burials 27, 28, and 32. A-I8

+ 4-JL5

Field Drawings of Burials 29 and 33.

A-19

+
-I-5L//

~
1
24 \1
i~

~
J

<, ~

3/ n

i
~+

4--3/.11

+
4-17/..//

Exacf Slqh;..
L0 caf-t;n /JOt-
./" I
-lIVen.
Field Drawings of Burials 34-35 and 36.
A-20

~
~
N.
J +
4-/5/.11

Burial #34 is found between sta. 4-3Lll -L13 and 4-5Lll -L13 at a depth of 37" below the surface - 16" in tan sand.
The body is laying on the left side, oriented E & W with the head to the W. The remains are in a very poor state of preservation - remaining bones consisting - skull, which is almost entirely decayed, nothing left except of portion of the right mandible with two of the jaw teeth and a portion of occipital, sections of both tibiae. Both of which are slightly higher than the head. This burial was evidently of the contracted, flexed type with the knees slightly higher than the skull and the arms going between the legs. No burial furniture.
Burial #35 is located between sra, 4-15Lll -L13 and 4-17L11 -L13. 67" below the surface, 27" below occupation level in tan sand. The body is laying on the left side, oriented E & W, with the head
toward the east.
The burial is of the contracted, flexed type with the knees drawn up toward the chest and the right arm crooked at the elbow-not going between the legs as is the usual case.
The skeletal remains are in a rather poor state of preservation, the skull is very badly decayednothing is left of the upper body except a short section of the humerus and ulnar of the right arm. The lower body consists of a few fragments of the lumbar vertebrae, most all of the pelvis and portions of both tibiae, femora and fibulae. No burial furniture associated with the burial.
Dec. 10, 1936. Troweled out Burial #35 and began work on cut from 4-11L13 -L15 to 4-27L13 -L15. Working down to occupational level to examine what appears to be a house site.
Dec. 28 to 30, 1936. Lowered section between 4-11L13 -L15 and 4-27Ll3 -L15, to the Indian occupation level - cleared whole area 4-11L9 -Ll5 to 4-27L9-Ll5. While the ova contains a very rich midden accumulation and several buried and charred areas and a few miscellaneous post mould impressions, there is no definite sign of a house site. There is quite a profusion of pottery and a few briquettes.
Began lowering section from 4-11Lll -4-15Lll to 4-11LI5-4-15LI5. Found a burial #36 between 4-11L13 -LIS and 4-13Ll3 -Ll5 with part of the body extending into the wall (L17). Began excavation of area 4-11LI5 -Ll7 to 4-13Ll5 -Ll7 and working down to burial.
Also started cleaning profile about 20' W of 4-27L15 on the west side of the R.R. fill. Worked down through yellow clay of R.R. fill and found the same component parts of the profile developments extending through to bank of river.
Jan. 7, 1937. Burial #36 found between sta, 4-11Ll5 -L17 and 4-15L15 -L17. 60" below surface, 35"
below occupation level. The body appears to have been placed on the back with the head turned to the left. The body is oriented E & W, with the head to the E. The knees were drawn straight up and are about 6" higher than the skull. Traces of the right humerus and both tibiae and the skull remain. No traces of any other bones.
Jan. 13, 1937. Burial #37 found by lowering section between 4-11L9 -LIS and Burial located between sta. 4-17L9 -L11 and 4-19L9 -Lll.
Burial is in good state of preservation. This is the contracted, flexed type of burial, oriented E & W, with the head to the W.
A-21

-_.. - 4r-~-
I .'

: I
~._..
-I- '17~ /I

i
I
I
. 4-.
-I-15t-11

-
Field Drawings of Burials 37 and 38. A-22

It is somewhat different from the previous burials, in that the body was buried face downward, the knees were drawn up evidently touching the chin and the left arm was flexed downward with the fingers on top of the right thigh, which means that the arm or hand neither went between the legs. The right arm is flexed and the right hand was at the face. The body is in a rather good state of preservation. Found 86" below the surface, 36" below the occupation level, in tan sand. No burial furniture associated with this burial.
Feb. 12, 1937. Removed overburden (R.R. fill) from remainder of area between 4-11L to 4-27L from L15 to river bank, working down to occupation level.
Located Burial #38 between sta. 4-19L17 to 4-23LI7, and 4-19L13 to 4-23L13 at a depth of below surface, 36" below occupation level. This is a double burial, both bodies were extended, lying on the right side, oriented NW and SE with the heads to the NW.
A. - This body is in a rather poor state of preservation - the skull is crushed and the parts are scattered. (Note: the arms of both bodies were folded across the waist) Remaining bones of A. consist of skull fragments - broken parts of humeri, ulna and radii, portions of both tibia, femora, and fibula.
B. - Is in a much better state of preservation. Remains consisting of skull, one side of which is rather badly decomposed - sections of both ulna -humeri and radii - both collar bones and parts of both shoulder blades - all of the vertebrae with a few of the ribs attached; part of the pelvis, both tibia femora and fibulae and most of the bones of both feet.
From the comparative size of these bodies it appears that A. was female - B. male.
No burial furniture was associated with this interment.
Feb. 15, 1937. Extension of the cut from 4-11L17 -L27 to 4-19L17 -L27, yielded Burial #39.
This burial was located between Sta. 4-21L17 and 4-21L19 at a depth of 42" below the occupation level. The body was lying on the left side, oriented SW & NE, with the head to the SW. This burial was contracted, flexed type with the legs drawn up toward the body and the head was tilted toward the knees. It was interesting to note, however, that instead of the hands going between the legs as is the usual case in this type of Bull Creek burial, the arms were flexed at the elbow and hands were at the neck. The skeletal remains were in an excellent state of preservation. Remaining bones consisted of - skull in good state. All of the cervical vertebrae, thoric vertebrae, with most of the ribs intact. None of the lumbar vertebrae left. Both humeri, ulnae and radii - all of the metacarpal bones - portions of the pelvis. Tibiae, femora, and fibulae - metatarsal bones. No burial furniture associated with the burial.
March 9, 1937. Burial #40 consists ofa few fragments of the skull found between sta. 4-9L15 -L13 and 4-11LI5 -L13 at a depth of 61" below the surface - 42" below occupation level in tan sand.
Burial #41, found between sta. 4-9L17 -L19 and 4-11LI7 -L19 at a depth of 39" below the occupation level. The body is lying on the right side, oriented E & W, with the head to the W. It is of the contracted, flexed type, but it is impossible to tell the position of the arms, as only a small portion of the left humerus remains. Remaining bones consist of parts of pelvis, tibiae, femora, fragments of skull and small parts of both clavicles. There is also a small piece of a joint of an animal bone, close to the upper part of the humerus, and two or three joints of the lumbar vertebrae. There is no burial furniture associated with this burial.
A-23

Burial #42, was found on line of sta. between sta. 4-17L19 and 4-19L19, 26" below occupation level in tan sand below surface.
The burial is of the contracted, flexed type, laying on the left side, oriented N & S., with the head to the South. The feet are gone but they must have been in contact with the Pelvis with the knees drawn up toward the chest. The left hand went between the thighs, about 2" from the pelvis and the right arm crooked downward at the elbow, the ulnar and radius being vertical. The head is turned to the left, thus making it face downward. The skull itself is in a rather poor state of preservation, the top or back of the head being rather badly crushed. On the whole, the body is in a fairly good state of preservation. Remaining bones consist of skull, vertebrae, some ribs, both clavicles, both ulna radii, humeri, both tibiae, femora and fibulae, two or three fingers and practically all of the pelvis.
From the general characteristics of the bones, it would seem that this is an adult female individual. No burial furniture associated with the burial.
Burial #44 is found between 4-19L19 -L21 and 4-21L19 -L21, at a depth of 26" below the occupation level in tan sand. Oriented E & W with head to W.
The body was buried face downward, with the legs flexed to the left. The arms were straight at the sides and the legs are laying on top of the left hand. The skull was rather badly crushed in the back, but otherwise is in an exceptional state of preservation. The remains on the whole are the best and most complete that have come from the Bull Creek Village, every part of the skeleton being perfectly preserved with the possible exception of a few finger joints, toes and one or two ribs. This was the interment of a male adult individual of middle age.
3-9 to 3-23, 1937. Moved surplus dirt on river side of cut - occupation level exposed as shown in preceding diagram (insert)
Worked section between 4-21L25 -L27 to 4-27L25 -L27, to a depth of 52" below occupation level.
Also troweled out Burials #43 -45. Found a small unbroken vessel, Find #72, at the excavation of section 4-21; - 4-23, at a depth of 27" below occupation level; below surface - 33" N. Sta. 4-17L21 on line of stks. There is no evidence of burial pit in this location and no trace of any bones, but this vessel must have been an article of burial furniture associated with a burial.
Burial #43 is of the extended, flexed type, found 50" below occupation level, on line of stats. between 4-11L21 and 4-11L23. The body was laying on the left side, oriented E & W with the head to the E. The feet are flexed back almost touching the pelvis and the knees are down and away from the body. It is impossible to tell the position of the arms, as nothing is left except traces of them. The skull has been rather badly crushed - Remaining bones consisted of skull - crushed - practically all of the vertebrae - both femora - tibiae, parts of the pelvis and sections of the clavicles.
Burial #45, was found between 4-17L21 -L23 and 4-19L2l -L23 at a depth of 33" below the occupation level. The remains were in a very poor state of preservation. All of the body gone except a portion of the skull and short sections of both tibiae and femora. The body was laying on the left side, oriented E & W, with the head to the E and from the position of the remains of the tibiae and femora it appeared that it was of the contracted, flexed type.
A-24

3-23 -3-30, 1937. Worked area from 4-13L27 - 4-21L27 to edge of river bank, 4-13L33 to 4-21L27 to a depth of 67" below occupation level at 4-13L27 and area between 4-9L21 -L25 and 4-27L21 -L25. Found a Burial #46 between sta. 4-17L23 -L25 and 4-29L23 L25.
Found a bone awl, Find #74, 10" deep in occupation level. 10" N -6" E. sta. 4-13L25. 4-1-'37. Completed excavation of area between 4-13L25 -L27 and 4-21L25 -27, with the possible exception that this will have to be lowered.
Burial #46 was found 28" below occupation level, in tan sand. Nothing remained of the body except the crown of a few teeth and short sections of both femora, tibiae. It was impossible to get any accurate picture of the burial, but it appeared that it was ofthe contracted, flexed type, laying on the left side, oriented N & S with the head to the South.
A-25

Bull Creek

.:~##~!~.:_ Qrim~i~~



2

26-27-28

W-E

R

3

33-34

S-N

L

4

SW-NE

L

5

44-45

SSE-NNW

L

6

50

NW-SE

R

7

46-47-48-49

S-N

L

8

55

E-W

L

9

56-57

SE-NW

L

10

11

SSE-NNW

L

12

62

NW-SE

R

13

S-N

L

14

58 to 61

E-W

L

15

SSE-NNW

L

16

SSW-NNE

17

NW-SE

R

18

64

SE-NW

L

19

64-66

SE-NW

R

20

67 to 69

ESE-WNW

L

21

NNW-SSE

R

22

76-77

NW-SE

R

23

24

SE-NW

L

25

NW-SE

R

26

S-N

L

C.F.

4-10; 4-10R1

4-12; 4-12Rl

C.F.

4-13; 4-14

36"

4-15; 4-16

E.F.

4-13; 4-13L1

20"

4-15; 4-15L1

C.F.

4-21L3

36"

E.F

4-17L1

47"

C.F.?

4-19L1-L3

36"

C.F.

4-19L3

44"

C.F.

4-15L1-L3

32"

4-15L1-L3

27"

4-13L1-L3

F

4-17L3-LS

38"

4-19L3-LS

C.F.

4-7;4-7L1

35"

4-11; 4-11L1

4-17; 4-18

26"

4-15; 4-16

C.F.

4-20; 4-20Rl

34"

4-18; 4-18Rl

C.F.

4-17L1

53"

4-17L3

4-13L3-LS

36"

4-15L3-LS

C.F.

4-9L3-LS

36"

4-11L3-L5

C.F.

4-24

39"

C.F.

4-24; 4-24Rl

45"

4-26; 4-26Rl

E.F.

4-13L1; L3

E.F.

4-17LS

21"

E.F.

40"

33"

C.F.?

4-11L9-L11

28"

4-13L9-L11

C.F.

4-11L9-L11

21 "

4-13L9-L11

E.F.

4-13L9

41 "

A-26

-27
28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

SSE-NNW

2[. - C.F.

4-13L5-L7

41 "

40 17L5-L7

ESE-WNW back flexed

C.F.

4-15L5-L7

46"

to L

4-17L5-L7

78-79

SE-NW

L?

C.F.

4-1715-17

47"

4-1915-17

NW-SE

R

E.F.

4-11L9

11"

82-83-84

NW-SE Triple burial

4-9L7-L9

23"

4-11L7-L9

87-88

N-S

R

C.F.

4-5L7-L9

34"

4-7L7-L9

85-86

S-N

R

C.F.

4-3L5

20"

W-E

L

4-3L11-L13

16

4-5L11-L13

89-90

E-W

L

C.F.

4-15L11-L13

27"

4-17L11OL13

91-92

E-W

L

4-11L15-L17

35"

40 15L15-L17

93-94

W-E

Face

C.F.

4-17L9-L11

38"

downward

4-19L9-L11

95-96-97

NW-SE

R

Double 4-19L13-L17

36"

burial

4-23L13-L17

98-99

SW-NE

L

C.F.

4-21L17

42"

4-21L19

100

Skull only 4-9L13-L15

42"

101

W-E

R

C.F.

4-9L17-L19

39"

4-11L17-L19

102-103

S-N

L

C.F.

4-17L19

26"

4-19L19

E-W

L

E.F.

4-11L21

50"

4-11L23

104-105-106

W-E

face down Semi C.F. 4-19L19-L21

26"

flexed to L

4-21L19-L21

E-W

L

C.F.

4-17L21-L23

33"

4-19L21-L23

No Picture

S-N

L?

C.F.?

4-17L23-L25

28"

4-19L230L25

A-27

Bull Creek
Skull Only Contracted flexed Contracted flexed Folded feet back to
pelvis flexed Contracted flexed Folded feet back to
pelvis flexed
Contracted flexed Contracted flexed Contracted flexed
Skull only Folded feet back to pelvis flexed (Female) Contracted flexed Near #3 (Below tree) (Old Man) contracted
flexed Contracted flexed
(West of #6) Double burial Contracted flexed Contracted flexed
Contracted more than any found yet - knees
on chest (Female) Folded feet to pelvis -
flexed Folded feet to pelvis -
flexed Flexed Fragmentary burial Contracted flexed Folded flexed type Extended flexed Contracted flexed

~4ti~~ ;;]~epth~~i6W~tf~~~

1

2

67"

3

67"

4

44"

No furniture No furniture (1) effigy pot and pierced seed beads No furniture

5

72"

Greenstone celt 7" long

6

78"

Greenstone celt 4" long, 2 small greenstone

chisels, several pieces worked flint, black

font knife, red pigment

7

78"

(2) effigy pot, discoidal stone

8

68"

No furniture

9

66"

Greenstone chisel 31h" long

10

54"

No furniture

11

73"

No furniture

12

69"

13

40"

14

51"

No furniture No furniture

15

78"

No furniture

16

(3) effigy pot

17

18

61"

Small bowl 31h" in diameter, several large

beads of shell, piece of conch shell core,

clay pipe human effigy handled type

19

67"

No furniture

20

21

36"

22

54"

23

57"

24

59"

25

52"

26

79"

27

75"

String of seed beads
No furniture
No furniture No furniture No furniture No furniture No furniture No furniture

A-28

Flexed

28

Contracted flexed

29

Flexed - with knees

30

farthest away from

body

Trible burial - first of

31

this sort found - 2

bodies contracted

flexed, other only the

skull

Contracted flexed

32

Contracted flexed

33

80"

Red pigment & small stoned, mussel shell,

evidently bone awl & other pieces of animal

bone

88"

Ball of red pigment

49"

No furniture

52"

Rocks, flint chips, fragments animal bones,

grouped around neck

36"

Rocks, flint chips, fragments animal bones,

grouped around neck

Bull Creek #1
Finds

Find #1 (39-22198)

3-13-36

Rimsherd with raised, notched design, about Ih" below rim. Found 24" N., 23" W. Sta. 1-46;

8" below surface, I" in black midden deposit.

Find #2 (39-22199)

3-14-36

A piece of a broken arrow-point with haft type end. Found 18" N. 8" E. Sta. 1-47. Ph" deep

in greyish sand, midden deposit, 11" below surface.

Find #3 (39-22200)

3-14-36

Small piece of rimsherd, showing raised, notched design, about 1/4" below rim; Found 14" S.,;

4" W. Sta. 1-48; 14" below surface, 5" in greyish sand, midden accumulation.

Find #4 (39-22201)

3-16-36

Peculiar type of rimsherd, showing notched design on top of rim and incised and punctate design

on the body of the pot. Found 22" N -12" W. Sta. 1-52; 6" deep in yellow sandy clay topsoil.

Find #5 (39-22202)

3-17-36

A quartz arrow point - imperfectly made, with portion of the point section broken off. Found

18" S. Sta. 1-56. 23" below surface, 3" deep, in black midden deposit.

Find #6 (39-22203)

3-17-36

Piece of broken flint, showing secondary working, possibly a portion of a scraper or chisel.

Found 6" N.; 16" E. Sta. 1-49; 13" below surface, 3" deep in reddish sand clay.

Find #7 (39-22204)

3-18-36

A small broken piece of stone, elliptical in shape with a hole drilled through it. Found 22" S;

26" E. Sta. 1-49; 16" below surface, 6" deep in house debris.

A-29

Find #8 (39-22205)

3-20-36

Small fragment of rimsherd with small handle attached Found 29" N., 4" E. Sta. 1-47LI; 12"

below surface, 7" deep in black midden deposit.

Find #8A (39-22206)

3-23-36

Small rimsherd with handle, found 15" below surface, 10 in black midden deposit. 22" S; 30"

W Sta. 1-49.

Find #9 (missing at time-no Park Service Number assigned)

3-21-36

A moulded clay briquette, showing timber mould and use of vegetable fiber, as tempering agent,

found resting on floor of house, 41" S. 5" E. Sta. I-52, 26" below surface, 5" deep in house debris.

Find #10 (39-22207)

3-21-36

A piece of broken celt or chisel, found 3" S., 23" W. Sta. 1-53. 18" below surface, 7" in black

midden deposit.

Find #11 (39-22208)

3-26-36

A hafted scraper of quartzite - fair workmanship of material. Found directly under Sta. 1-47L1,

8" below surface, in black midden deposit.

Find #12 (39-22209)

4-6-36

Quartz arrowhead, spinner type, found 2" deep in yellow topsoil, in 5' exploration trench, 36"

E, 18" S. Sta. i-n n.i.

Find #13 (39-22210)

4-8-36

Piece of broken knife or spear-head, found 4" deep in black midden deposit, 12" below surface.

31" e., 17" S. Sta. 1-107Ll.

Find #14 (39-22211)

4-8-36

Quartzite arrowhead, very poor workmanship, found 8" deep in black midden deposit, 13" below

surface. 46" S, 7" E Sta. 1-IIlLI.

Find #15 (39-22212)

4-8-36

Chunky stone of granite type rock, found 6" deep in black midden deposit, 9" below surface.

36" N., 12" E. Sta. 1-91Ll.

Find #16 (missing at time-no Park Service Number assigned)

4-8-36

Pieces of broken pot, found massed together, 13" deep in black midden deposit. 19" below

surface, 35" E., 4" S. Sta. 1-95Ll.

Find #17 (39-22213)

4-8-36

Portion of rimsherd, with very small handle attached; found 9" deep, in black midden deposit.

12" below surface, 18" E. - 18" S. Sta. 1-91Ll.

Find #18 (39-22214)

4-13-36

Small piece of worked flint, possibly used as an engraving tool. Found in 5 ft test trench, 7"

below surface, in black humus, 7" E., 9" N. Sta. 1-103L5.

Find #19 (39-22215)

4-13-36

A-30

Portion of a rimsherd, made in effigy form, of a human face. Found in black humus, 3" below surface. Found 52" N., 31" W. Sta. 1-95L1.

Find #20 (39-22216)

4-13-36

Chunky stone of granite type rock, rather poor workmanship. Found 4" deep in black midden

deposit, 5" below surface. 27" S.; 42:E. Sta. 1-103L5.

Find #21 (missing at time-no Park Service Number assigned)

4-14-36

Large pieces of two broken pots, found in Pit #1, 16" to 18" below surface, in pit fill. 42" S.;

31" E. Sta. 1-119L13.

Find #22 (39-22217)

4-16-36

A piece of worked quartz - knife or side scraper; found 3" deep in yellow topsoil. 39" S. - 52"

W. Sta. 1-99L1.

Find #23 (39-22218)

4-15-36

Broken piece of quartz arrowhead, imperfectly worked. Found 2" below surface, in black

humus. 8" S. - 52" W. Sta. 1-99L1.

Find #24 (39-22219)

4-15-26

A small quartz scraper, found 6" deep in black midden deposit, 9" below surface. 5" N. - 25"

W. Sta. 1-107Ll.

Find #25 (39-22220)

4-15-36

Broken arrowhead or scraper, poor workmanship, found 8" deep in black midden deposit, just

above house debris, 15" below surface, 22" S. - 2" W. Sta. 1-99L1.

Find #26 (39-22221)

4-16-36

Small pottery chunky stone, found 7" below surface, in black humus. 15" S. - 28" W. Sta. 1-

95Ll.

Find #27 (39-22222)

4-16-36

Small spinner type arrowhead, excellent workmanship. Point of bluish flint. Found 9" below

surface, 4" in black midden deposit, on level of house debris. 43" S. - " E. Sta. 1-99L5.

Find #28 (39-22223)

4-20-36

Broken quartz arrowhead or scraper, imperfectly worked, probably a reject. Found 13" below

surface, 7" deep in black midden deposit. 22" N., 12" E. Sta. 1-91.

Find #29 (39-22224)

4-20-36

A portion of what appears to be a bird effigy of some type. Seemingly a portion of the rim of

a pot. Found 16" below surface, 10" in black midden, directly under Sta. 1-94.

Find #30 (39-22225)

4-20-36

A small pottery chunky stone, found 10" below surface. 5" deep in black midden deposit. 11"

E., 17" N. Sta. 1-99L7.

Find #31 (missing at time-no Park Service Number assigned)

4-20-36

Several large pieces of one pot - plain, with beaded rim design, found 13" below surface, 7th "

in black midden deposit, in what appears to be a small pit. 20" W. - 9" N. Sta. 1-93L5.

A-31

Find #32 (39-22226)

4-20-36

Flint scraper, very excellent workmanship, found 13" below surface, 7 1h" in black midden

deposit. 20" W. - 9" N. Sta. 1-93L5, in what appears to be a small pit.

Find #33 (39-22227)

4-14-36

Broken hafted scraper or knife (flint), found 41" N. - 38" E. Sta, 1-119L13, 3" below surface

in black humus.

Find #34 (39-22228)

4-21-36

Small metal sewing thimble, found 12" below surface, 4" deep in black midden deposit, in what

appears to be slump, near edge of bank. 16" W. - 8" S. Sta. 1-100.

Find #35 (39-22229)

4-22-36

Small flint arrowhead, found 8" below surface, 4" in black midden deposit, 26" N. -15" W. Sta.

1-95L7 on west edge of bank.

Find #36 (39-22230)

4-22-36

Quartzite hand-chopper, found 5" S. Sta. 2-70 on line of stas., 4" below surface in topsoil.

Find #37 (39-22231)

5-14-36

Broken pieces of a pot, found on floor level of house site #2. Bowl and rim found 42" E. - 7"

N. Sta. 1-95L1. Bottom found 32" E. - 14" S. 1-95L1. Pot approximately 9" in diameter, plain design,

with fluted rim.

Find #37A (39-22232)

4-29-36

Small arrowhead, found 11" below surface, 6" deep in midden deposit, 32" W - 12" S. Sta. 1-

99L5.

Find #38 (39-22233)

5-14-36

Broken pieces of a large pot, notched rim, incised and punctated design on body of pot. Pieces

found grouped together on floor level of house site #2, between Sta. 1-97Ll -L3 and 1-99L1 -L3.

Find #38A (39-22234)

5-25-36

Flint arrowhead, poor workmanship, found 12" below surface, 8" deep in midden deposit, 18"

S. - 10" E. Sta. 1-41.

Find #38B (39-22235)

5-11-36

Small quartzite arrowhead, found 27" N. - 40" E. Sta. 1-119L9, 10" below surface in black

midden deposit.

Find #39 (missing at time-no Park Service Number assigned)

5-14-36

Broken pieces of a small pot (approx. 9" dia.) resting on floor level of house site #2, 7" W - 31"

S Sta. 1-99L3. Pot has notched rim and plain body. Body notched due to firing.

Find #40 (39-22236)

5-14-36

Several pieces of a large pot found on floor level of house site #2. Found 24" S. - 18" W. Sta.

1-99L3, 13" below surface.

A-32

Find #41 (missing at time-no Park Service Number assigned)

5-14-36

Portion of broken rim of large pot (13" dia.). Found resting on floor level of house site #2, 13"

below surface - 8" W. Sta. 1-99L3. Pot shows painted or beaded rim and circular stamped design.

Find #42 (missing at time-no Park Service Number assigned)

5-14-36

Portion of rim of large pot (I3 "dia.). Found resting on floor level of house site # 2-13" below

surface, 13" S. 18" W. of Sta. 1-99LE.

Find #43 (39-22237)

4-21-36

Hafted type flint scraper, rather poor workmanship. Found 5" below surface, in humus. 451h"

N. Sta. 1-135L9 on line of stakes.

Find #44 (39-22238)

6-17-36

Small pottery chunky stone. Found 6" N. 6" E. Sta. 4-19L3. 25" below surface, 11" in tan

sand.

Find #45 (this is note on sheet in upper left corner) Small bead

Find #46 (39-22239)

7-17-36

Rimsherd, plain design, with small type handle. Found 3" deep in black occupation level, 30"

below surface, 10" N. - 26" E. Sta. 4-19L5.

Find #47 (39-22240)

7-17-36

Small cylinder of baked clay. Found 36" E. Sta. 4-19L5 on line of station, 36" below surface,

9" deep in occupation level.

Find #48 (missing at time-no Park Service Number assigned)

7-17-36

Small briquette showing end of reed imprint, found 53" N. - 9" E. Sta. 4-19L5. 33" below

surface, 7" deep in occupation level.

Find #49 (39-22241)

7-20-36

Clay bead, found 5" deep in tan sand, 40" below surface, 11" E. Sta. 4-18, on line of sta.

Find #50 (39-22242)

7-20-36

Rimsherd with unusual flanged design on neck of pot. Found 6" deep in tan sand, 40" below

surface, 31" S. Sta. 4-18.

Find #51 (missing at time-no Park Service Number assigned)

7-30-36

Small greenstone celt, 4" long, 2Vz" at cutting end, 1" at butt end, found as.part of burial #6.

Found 67Jh" below surface, 35" in tan sand, 7" S. 7" E. Sta. 4-17Ll.

Find #52 (39-22243)

7-30-36

Small greenstone chisel found 671h: below surface, 35" in tan sand, 17" N. 8" E. Sta. 4-17Ll

part of burial furniture with Sta. burial #6.

Find #53 (missing at time-no Park Service Number assigned)

7-30-36

Small greenstone chisel found as part of burial furniture with Burial #6. Found 671h" below

surface, 35" in tan sand, 26" N. 5" E. Sta. 4-17L1.

A-33

APPENDIX B WPA Coding Sheets

PART OF POT REPRESENTED

1. Sherd from Undet. portion of pot 2. Rimsherd 3. Rim and shoulder or neck 4. Rim and handle or lug 5. Rim and body 6. Rim, Body, and Base

7. Body 8. Body and Base 9. Base 10. Handle 11. Whole Pot 12. Undetermined

TEMPER Material

1.

Shell

2.

Sand

3.

Crushed Rock or Grit

4.

Vegetal

5.

Potsherd

6.

Ground Micaceous stone

7.

Hole Temper

SIZE

A.

Large

B.

Medium

C.

Small

CROSS SECTIONAL APPEARANCE OF PASTE STRUCTURE

1.

Fine

2.

Medium fine

3.

Medium

4.

Medium Coarse

5.

Coarse

6.

Granular

7.

Laminated

8.

Contorted

9.

Compact

Criteria (1.
(2.
(3.

Size of Temper Distribution Relatively homogeneity of paste

HARDNESS

22-2.52.533.5-

Gypsum Fingernail Cryolite Calcite Celestite

4-

Flourite

4.5- Chabazite

5-

Apatite

6.- Adularia

SURFACE FINISH - Inner or Outer Surfaces

1. Very rough - (1. From extrusion of temper or lumpish granular exterior knobby)

2. Rough

(2. From pitting, holeing, intaglic effects from absorption or dissolution of temper

or paste)

3. Smooth

(3. Note by extent to which smoothing process is evidenced by mechanical striation

from the pottery smoother)

4. Glossy

5. Polished

6. Vermiculated - Theta, Type sherd EXECUTION of design

7. Glistening -finish- pseudo glaze

1. Crude

2. Fair

3. Good

4. Excellent

5. Unobserved

AREA OF DECORATION
1. Inner Rim 2. Lip 3. Outer Rim 4. Shoulder

5. Body 6. Undetermined 7. Base
x. Denotes design on inside of sherd

DECORATION A. - POSSIBLY related technologically to surface finish.

Cord Marking

1. Irregular Criss-Cross 2. Checker Board 3. Diamond 4. Straight line series or parallel patterns 5. Single strands occur at sparse intervals 6. Undetermined

SMOOTHED-BURNISHED

COMBED, BRUSHING

1. Combing - Definition 2. Brushing - Definition 3 Scratching - Definition

V. Decoration possibly applies with Grass

Phi - Stoneware

Theta - Grass Impressions

Mottling from Firing

X. Denotes design or Execution on inside of sherd.

DECORATION B. - Specialized stamped or paddle marked technique.

Alpha- Large residual class of general stamped or paddle marked sherds in which specific designs are apparent.

1

6

2

7

3

8

4

9

5

10

Sigma - A general class of crudely stamped or impressed sherds characterized by irregular linear or crisscross impressions whose exact technological origin is not known, but which give a taxonomic resemblance when regarded as a whole.
Delta - A fine line cameo effect pattern (Possibly a large and small division).
Pi 1. Checkerboard design squares 2. Rectangular 3. Diamond
a. Large b. Medium c. Small

Decoration C
Technique a. Incising
1. Narrow 2. Medium 3. Wide
b. Trailing 1. Medium 2. Wide 3. Very ?Wide
c. Engraving
d. Punctate a. Small b. Medium c. Large
(Punctate Design) 1. Solid rounded 2. Cylinder 3. Crescent 4. Stippling 5. Triangular 6. Rectangular 7. Bi-sected cones 8. Undetermined
T.N. Thumb nail
R. Roulette

DESIGN
RECTILLINEER
1. Vertical 2. Rectangles 3. Horizontal 4. diagonal or oblique 5. triangle herringbone 6. cross hatching 7. Zigzag 8. Undetermined 9. Triangle
CURVILINEER
1. Concentric Circles 2. Scroll 3. Guilloche 4. Undetermined

Nodes

1. Extruded 2. Applique 3. Moulded

F. Filming

1

2

3

4

1. Painting Red Yellow orange and black

2. Application of slip

a. put on before firing

b. Put on after firing

Clay-wash; not the same as definite slip preparation

Modeling 1. Fish 2. Frog 3. Owl 4. Human 5. Cat 6. Snake 7. Eagle
1. Fabric Impressed

8. Geometric 9. Naturalistic 10. Whole pot 11. Rim 12. Body 13. Bass 14. Lip 15. Bird
Fabric Impressions 1. plain plaiting 2. plain twining - wicker 3. twilled twining 4. zigzag twilled twining 5. Chevron 6. undetermined

DECORATION D - SECONDARY FEATURES

L.H. Loop Handles a. Attached to lip b. Attached below lip c. Project above lip d. Lateral width e. Thickness f. Height g. Handle projection h. Button knob i. Bifurcate teats or knobs at apex (triple knobs at apex) j. Ribbed or single rib handle k. Nodes on middle handles 1. Depression at handle apex m. Rim raised at handle n. Moulded at base o. Riveted at base p. Handle

S.H. Strap Handles a. Attached to lip b. Attached below lip c. Project above lip d. Lateral width e. Thickness f. Height g. Handle projection h. Moulded at base i. Riveted at base

L. G. Lug Handles a. Length b. Height c. Thickness d. Attached below lip e. Attached at lip f. Attached at shoulder g. Rounded h. Pointed or knob? i. Elongate or flange

R.S. Applique Rim Strip (Rim moulding or luted rim strip or band)

1. Moulded to or joined with lip 2. Moulded to or joined below lip c. Teats or nodes d. Beaded e. Pinched? Indented? f. Medium thumbing or swelling g. Cylinder impressions h. Bi-sected impressions i. Rounded band at lip

j. Crescent k. Stamped 1. Scalloped m. Sigma p. Plain

GENERAL VESSEL MORPHOLOGY

Lip 1. Pointed 2. Narrow and Rounded 3. Flattened a. Flat horizontal b. Flat slopes in c. Flat slopes out 4. Everted a. In b. Out c. Both sides 5. Folded 6. Rounded

LIP DECORA nON 1. Scalloped 2. Notched 3. Parallel grooved 4. Incised 5. Punctate 6. Cylinder Impression 7. Undetermined

RIM SHAPE Pot -

A vessel with curvilinear sides and partially constricted orifice. 1. Straight 2. Slightly flaring 3. Flaring 4. Pointed 5 . Incurving 6. Outcurving rim

Beaker Beaker proper Straight side cup

Bottle Globulat body cylinderical neck

Bowl

A. Straight B. Incurving rim C. Cazuela D. Outcurving rim

Plate

Straight rim Outcurving rim

Dish Pans

Base 1. Conoidal 2. Oval 3. Globular 4. Truncate 5. Flattened 6. Annular

7. Outer Base Legs. a. Three legs b. Four legs
8. Inner Legs a. Three legs b. Four legs
9. Footed
Size 1. Lip - thickness 2. Rim - thickness 3. Shoulder - thickness 4. Body - thickness 5. Rim - height 6. Lip diameter 7. Shoulder diameter or greatest width of body 8. Base diameter (if flattened) 9. Height 10. Neck Diameter 11. Neck Height 12. Undetermined 13. Base Thickness

Execution of Design

Excellent

4

Crude

5

Fair

2

Good

1

Unobserved

5

The Quality of the ware

1. Tenacity - Does pottery hold together or is it brittle, flexible, easily broken? 2. Porosity - is it porous; does it show crackling markedly? 3. Firing - Are firing effects evenly distributed or does ware show in cross section areas showing
greater or less effects of heat? 4. Surface finish - Crude, rough finish show poorer technique in pottery manufacturing. 5. Relation of mass to tensile strength
1. Excellent 2. Good or fair 3. Inferior 5. Very crude

APPENDIX C Bull Creek Vessel Form Data

..
..

2

Schnell 1959

23

Fuller 1981

26

WPA

28

WPA

36

WPA

41

WPA

42

WPA

43

WPA

44

WPA

45

WPA

46

WPA

48

WPA

50

Kelly 1950

51

Kelly 1950

53

Kelly 1950

55

Kelly 1950

69

WPA

70

WPA

71

WPA

72

WPA

73

WPA

74

WPA

75

WPA

77

WPA

79

WPA

82

Kelly 1950

85

Kelly 1950

92

Kelly 1950



32 em 18 em 14 em 40 em 32 em 36 em 34 em 42 em 32 em 36 cm 36 em 38 em 40 em 28 cm 44 cm 28 em 14 cm 26 em 36 em 32 em 26 em 34 cm 28 em 22 em 16 em 16 em 22 cm 26 em 22 em

Applied Strip (Pinched) No Modification
Folded Pinched (13 mm) No Modification
Folded Pinched (20 mm) Applied Strip (Pinched)
No Modification Folded Pinched (25 mm) Applied Strip (Pinched) Applied Strip (Pinched) Folded Pinched (24 mm) Folded Pinched (26 mm) Folded Pinched (15 mm) Folded Pinched (21 mm) Folded Pinched (20 mm) Folded Pinched (12 mm)
Applied Strip (Noded) Applied Strip (Noded)
No Modification Applied Strip (Noded) Applied Strip (Noded) Applied Strip (Noded) Folded Pinched (14 mm) Folded Pinched (12 mm)
No Modification Noded
Folded Pinched (16 mm) No Modification No Modification

Complicated Stamped Complicated Stamped Complicated Stamped
Plain Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped
Plain Complicated Stamped
Plain Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped
Plain Plain Plain Plain Complicated Stamped Complicated Stamped

><

00

Li,
.../: .:

"lil'!!11

97

Kelly 1950

98

Kelly 1950

99

Kelly 1950

100

WPA

101

WPA

102

WPA

103

WPA

104

WPA

113

WPA

120

WPA

121

WPA

122

WPA

123

WPA

124

WPA

125

WPA

126

WPA

127

WPA

128

WPA

129

WPA

130

WPA

131

WPA

132

WPA

133

WPA

134

WPA

135

WPA

145

Schnell 1959

147

Schnell 1959

150

Schnell 1959

154

Schnell 1959

~
34 cm 32 em 36 em 42 em 32 em 38 em 32 em 26 em 26 em 38 em 28 em 32 em 20 em 24 em 36 em 34 em 34 em 34 em 36 em 36 em 34 em 36 em 30 em 32 em 30 em 40 em 34 em 36 em 14 em

/:11

) ~

~I-i-- ill ill

~)

...4t1
<

Folded Pinched (19 mm) Complicated Stamped

Folded Pinched (16 mm) Complicated Stamped

Folded Pinched (23 mm) Complicated Stamped

Folded Pinched (20 mm) Complicated Stamped

Folded Pinched (27 mm) Complicated Stamped

Folded Pinched (14 mm) Complicated Stamped

Noded

Plain

Noded

Plain

Noded/Scalloped

Plain

Applied Strip (Noded) Complicated Stamped

Folded Pinched (20 mm) Complicated Stamped

Folded Pinched (18 mm) Complicated Stamped

Folded Pinched (16 mm) Complicated Stamped

Folded Pinched (18 mm) Complicated Stamped

Folded Pinched (26 mm) Complicated Stamped

Folded Pinched (16 mm) Complicated Stamped

Folded Pinched (16 mm) Complicated Stamped

Folded Pinched (14 mm) Complicated Stamped

Folded Pinched (20 mm) Complicated Stamped

Folded Pinched (15 mm) Complicated Stamped

Folded Pinched (22 mm) Complicated Stamped

Folded Pinched (22 mm) Complicated Stamped

Folded Pinched (21 mm)

Plain

Folded Pinched (23 mm)

Plain

Folded Pinched (18 mm)

Plain

Folded Pinched (25 mm) Complicated Stamped

Folded Pinched (19 mm) Complicated Stamped

Folded Pinched (20 mm) Complicated Stamped

No Modification

Plain



< i 0 /..

> .......

.. i


<

155

Schnell 1959

160

Schnell 1959

161

Schnell 1959

166

Schnell 1959

168

Schnell 1959

171

Schnell 1959

172

Schnell 1959

173

Schnell 1959

176

Schnell 1959

177

Schnell 1959

181

Schnell 1959

183

Schnell 1959

184

Schnell 1959

186

Schnell 1959

187

Schnell 1959

191

Schnell 1959

192

Schnell 1959

193

Schnell 1959

194

Schnell 1959

207

Fuller 1981

208

Fuller 1981

209

Fuller 1981

211

SAS 1993

213

SAS 1993

214

SAS 1993







<

B.

}

28 em 26 em 38 em 24 em 26 em 36 em 40 cm 40 em 18 em 40 cm 34 em 40 em 32 cm 40 em 38 em 42 em 34 em 34 cm 34 em 32 em 24 em 42 em 40 em 32 em 30 em

Applied Strip (Noded) Folded Pinched (17 rom) Folded Pinched (22 rom)
Noded No Modification Folded Pinched (15 rom) Folded Pinched (20 rom) Folded Pinched (25 rom) Folded Pinched (16 rom) Applied Strip (Noded) Applied Strip (Pinched)
Lug Handles Folded Pinched (25 rom)
Noded No Modification Applied Strip (Noded) Applied Strip (Noded) Folded Pinched (15 rom) Folded Pinched (24 rom) Folded Pinched (22 rom) Folded Pinched (16 rom) Folded Pinched (18 rom) Folded Pinched (20 rom) Folded Pinched (25 rom) No Modification

Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped
Plain Plain Check Stamped Complicated Stamped Plain Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped

n >/ < ! it~;;11

a',~Ci(Cd u-ddc
I)i~

~()Wl: of _ > <

> /<1



3

Schnell 1959

4

Schnell 1959

10

WPA

11

WPA

35

WPA

49

Kelly 1950

56

Kelly 1950

57

Kelly 1950

83

Schnell 1959

84

WPA

91

WPA

106

WPA

114

WPA

115

WPA

138

WPA

142

WPA

146

Schnell 1959

148

Schnell 1959

149

Schnell 1959

156

Schnell 1959

157

Schnell 1959

163

Schnell 1959

164

Schnell 1959

167

Schnell 1959

169

Schnell 1959

179

Schnell 1959

180

Schnell 1959

196

Fuller 1981

200

Fuller 1981

24 em 24 en 14 em 16 em 24 em 30 em 36 em 28 em 32 em 26 cm 24 em 24 em 26 em 22 em 32 em 18 em 40 em 30 em 30 em 36 em 38 em 32 em 30 em 32 em 40 em 30 em 34 em 18 em 24 em

Notched Notched
Plain Notched Notched Incised
Plain Notched
Plain Notched Notched Notched Notched Incised
Plain Plain Incised Plain Plain Notched Plain Notched Notched Plain Plain Plain Notched Plain Plain

;

%1l

illl
Incised/Punctate

Incised/Punctate

Incised

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised

Incised

Incised

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised/Punctate

Incised

Incised/Punctate

Incised/Punctate

Incised

Incised

Incised/Punctate

Incised

Incised

II. .
.
201 202 203 204 205 206 212

Fuller 1981 Fuller 1981 Fuller 1981 Fuller 1981 Fuller 1981 Fuller 1981 SAS 1993

hil

. .. .

, ... rUU1..

..

32 em 30 em 24 em 28 em 34 em 26 cm 36 em

Notched Plain
Notched Plain
Notched Plain Plain

Plain Incised/Punctate Incised/Punctate Incised/Punctate Incised/Punctate Incised/Punctate
Incised

.....:

.

. .

...

7 20
... .... . .
5 6 12 14 15 17 18 19 24 25 27

WPA
..
Schnell 1959 WPA
Kelly 1950 WPA WPA
Schnell 1959 Fuller 1981
WPA WPA WPA WPA

...

. ...

.:
.

36 em 26 cm

Rolled Plain

.1

32 em 32 em 20 em 40 em 36 em 24 em 36 em 36 em 42 em 36 em 38 em

Applied Strip (Notched) Plain Plain Noded Plain
Notched Effigy Adorno
Notched Notched
Plain Plain

Check Stamped Plain
=
- Plain Incised Plain Plain Incised/Punctate Complicated Stamped
Incised Complicated Stamped Complicated Stamped
Incised/Punctate Complicated Stamped

TIl fJIii i <I

28

WPA

30

WPA

31

WPA

32

WPA

33

WPA

34

WPA

37

WPA

39

WPA

40

WPA

47

WPA

52

Kelly 1950

54

Kelly 1950

58

Kelly 1950

59

WPA

60

WPA

61

WPA

62

WPA

63

WPA

64

WPA

65

WPA

66

WPA

67

WPA

68

WPA

76

WPA

78

WPA

80

WPA

81

WPA

86

WPA

87

WPA

III !II _ I siriM ~lii@iillliiiWj.iitll

l. illf~~t~.

~ti .....

30 em 26 em lOcm 22 em 38 em 40 em 34 em 26 em 30 em 28 em 26 em 24 em 26 em 34 em 40 em 36 em 40 em 32 em 34 cm 32 em 26 em 36 cm 30 em 22 em 34cm 28 em 32 em 30 em 32 em

Notched Notched
Plain Plain Notched Notched Fold Plain Notched Notched Folded Pinched Rolled Plain Plain Notched Plain Plain Applied Strip (Noded) Plain Folded Notched Plain Notched Notched Notched Notched Notched Notched Plain Plain Plain

Complicated Stamped Complicated Stamped
Plain Plain Complicated Stamped Plain Plain Complicated Stamped Complicated Stamped Complicated Stamped Incised Plain Complicated Stamped Complicated Stamped Plain Complicated Stamped Plain Plain Complicated Stamped Plain Complicated Stamped Complicated Stamped Complicated Stamped Complicated Stamped Plain Plain Plain Complicated Stamped Complicated Stamped

.
88 89 90 93 94 95
96
105 107 111 112 116 117 136 137 139 140 141
143 144 151 152 153 158 159 162 165 170 174

.' .. '... .. .' I

WPA WPA WPA WPA WPA WPA WPA Kelly 1950 Kelly 1950 Kelly 1950 WPA Kelly 1950 Kelly 1950 WPA WPA WPA WPA WPA Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959 Schnell 1959

32 em 28 cm 16 em 30 em 36 em 30 em 22 em 24 em 30 cm 42 em 28 cm 36 em 28 em 22 em 26 cm 10 em 34 cm 40 em 36 em 28 cm 28 em 38 cm 36 em 36 em 32 em 18 cm 32 em 24 em 18 cm

. .
.

Notched Notched Rolled
Plain Folded Pinched
Plain Notched
Plain Plain Thickened Effigy Adorno Notched Plain Notched Plain Plain Plain Plain Folded Plain Plain Plain Plain Pinched Thickened Applied Strip (Notched) Notched Plain Rolled

Complicated Stamped Complicated Stamped
Plain Incised Complicated Stamped Plain Complicated Stamped Incised Incised Plain Plain Complicated Stamped Incised Complicated Stamped Plain Plain Complicated Stamped Plain Incised Plain Complicated Stamped Plain Plain Complicated Stamped Plain Plain Complicated Stamped Complicated Stamped Plain

II

J2"
.......
1111

........

....... r



l36



I>



178

Schnell 1959

182

Schnell 1959

185

Schnell 1959

188

Schnell 1959

190

Schnell 1959

195

Fuller 1981

197

Fuller 1981

198

Fuller 1981

199

Fuller 1981

210

WPA

215

SAS 1993

216

WPA

32 em 34 em 28 em 18 em 22 em 14 em 18 em 20 em 36 em 22 em 30 em 28 em

Plain Noded Folded Pinched (16 mm) Pinched Plain Incised Folded Notched Folded T-rim Folded Notched Plain

Complicated Stamped Check Stamped
Complicated Stamped Complicated Stamped
Plain Incised Incised Plain Incised Incised Plain Plain

8

WPA

9

WPA

13

WPA

16

WPA

21

Fuller 1981

22

Fuller 1981

38

WPA

108

Kelly 1950

109

Kelly 1950

110

Kelly 1950

16 em 24 em 20 em 10 em 10 em 14 em 18 em 34 em 34 em 36 em

L-shaped Plain
Lug Handles Incised Plain Plain Plain Notched Plain
Thickened

Incised Complicated Stamped
Plain Incised/Punctate Complicated Stamped
Plain Plain Complicated Stamped Plain Plain

..
.. \/ '/.

/.' i} \ '.....'. ,.....

118

Kelly 1950

119

WPA

175

Schnell 1959

189

Schnell 1959

2.~111
42 em 34 em 38 em 22 em

.. '.

. <: "
i
L-shaped Plain
Thickened Thickened

\
.,......
i<

" i
<<:

Complicated Stamped

Plain

Plain

Complicated Stamped

< '.'.'...,.
:S~tl

,i/
,
/
<


i\\
i

217

WPA Burial 3

218

WPA Burial 7

219

WPA Burial 16

i' lily .......

<: <:
'. i< <.i..>.. '.\

<:

9.5 em 9.5 em 8.0 em

30 em long, 26.5 em high 27.5 em long, 27 em high 28 em long, 32 em high

,

Negative Painted Negative Painted
Red?