Cultural Resources Data Recovery at Bayou Park (8OK898), Eglin Air Force Base, Okaloosa County, Florida

Share Embed


Descrição do Produto

  FINAL REPORT   

CULTURAL RESOURCES DATA RECOVERY  AT BAYOU PARK (8OK898),  EGLIN AIR FORCE BASE,  OKALOOSA COUNTY, FLORIDA 

  CONTRACT NO. EAFB‐62221‐008  TASK ORDER NO. CR‐11‐0051    PREPARED FOR  MID‐BAY BRIDGE AUTHORITY  AND  96TH CIVIL ENGINEERING GROUP, EGLIN AIR FORCE BASE    PREPARED BY  ROBERT J. AUSTIN, CHRISTOPHER MICKWEE, AND JOSHUA M. TORRES    WITH CONTRIBUTIONS BY  ANNA R. DIXON, IRVY R. QUITMYER, DENNIS K. WARDLAW, BRIAN WORTHINGTON,  AND CHAD YOST   

________________________________  ROBERT J. AUSTIN, PHD, RPA  PRINCIPAL INVESTIGATOR   

SOUTHEASTERN ARCHAEOLOGICAL RESEARCH, INC.  WWW.SEARCHINC.COM 

SEARCH PROJECT NO. 2614‐11010V    AUGUST 2014 

This page intentionally left blank 

TABLE OF CONTENTS      List of Figures ........................................................................................................................... vii  List of Tables .......................................................................................................................... xvii  List of Acronyms ..................................................................................................................... xxi  Acknowledgments ................................................................................................................ xxiii  SEARCH Project Participants ................................................................................................ xxiv  Executive Summary ............................................................................................................... xxv    Chapter 1: Introduction ..................................................................................................... 1    Site Location and Present Condition ................................................................................... 2    Summary of Previous Investigations................................................................................... 2    Summary of Phase III Results ............................................................................................ 24    Conclusion ......................................................................................................................... 30  Chapter 2: Environmental and Archaeological Contexts ....................................... 31    Regional Environmental Setting ....................................................................................... 31    Paleoenvironmental Conditions ....................................................................................... 44    Prehistoric Overview ......................................................................................................... 49    Conclusion ......................................................................................................................... 68  Chapter 3: Research Design and Methods ................................................................. 69        Research Design ................................................................................................................ 69    Field Methods ................................................................................................................... 77    Laboratory Methods ......................................................................................................... 95    NAGPRA‐Related Human Remains ................................................................................... 95    Curation ............................................................................................................................ 96    Appendices ........................................................................................................................ 97  Chapter 4: Stratigraphy and Chronology .................................................................. 109    Description of Strata ....................................................................................................... 109    Site Chronology ............................................................................................................... 123    Temporally Diagnostic Artifacts ...................................................................................... 127    Conclusion ....................................................................................................................... 134    Appendices ...................................................................................................................... 139  Chapter 5: Ground‐Penetrating Radar Survey by Dennis K. Wardlaw ................... 147    Methods .......................................................................................................................... 147    GPR Survey Results ......................................................................................................... 149    Discussion of Results ....................................................................................................... 160    Conclusion ....................................................................................................................... 163  Chapter 6: Midden Deposits and Features .............................................................. 165    Midden Deposits ............................................................................................................. 167    Features .......................................................................................................................... 171    Lipid Residue Analysis of Selected Features ................................................................... 245  iii 

           

Functional Interpretations .............................................................................................. 249  Spatial and Temporal Distributions ................................................................................ 255  Structures ........................................................................................................................ 261  Discussion........................................................................................................................ 262  Conclusion ....................................................................................................................... 274  Appendices ...................................................................................................................... 275  Chapter 7: Baked‐Clay Objects .................................................................................... 291    Background ..................................................................................................................... 291    Bayou Park Baked‐Clay Object Typology ........................................................................ 294    Analysis Methods ............................................................................................................ 298    Assemblage Description .................................................................................................. 300    X‐Ray Fluorescence Spectrometry Analysis .................................................................... 311    Spatial Distribution ......................................................................................................... 314    Functional Interpretations .............................................................................................. 316    Technological Interpretations ......................................................................................... 322    Temporal Comparisons ................................................................................................... 323    Conclusions ..................................................................................................................... 328    Appendix ......................................................................................................................... 331 

Chapter 8: Phytolith and Starch Analysis of a Possible Griddle    Fragment from Bayou Park by Chad Yost............................................................... 353         

Background ..................................................................................................................... 353  Methods .......................................................................................................................... 353  Results ............................................................................................................................. 354  Summary and Conclusions .............................................................................................. 357  Chapter 9: Prehistoric Pottery ..................................................................................... 359    Methods .......................................................................................................................... 359    Results ............................................................................................................................. 360    Conclusion ....................................................................................................................... 367    Appendix ......................................................................................................................... 369  Chapter 10: Lithic Artifacts: Resource Exploitation and Use .............................. 375    Summary of Previous Research ...................................................................................... 375    Analysis Methods ............................................................................................................ 378    Lithic Raw Materials ........................................................................................................ 385    Artifact Descriptions ....................................................................................................... 390    Artifact Function ............................................................................................................. 432    Temporal and Spatial Variation ...................................................................................... 434    Technological Organization ............................................................................................ 444    Assemblage Diversity ...................................................................................................... 448    Discussion........................................................................................................................ 448    Conclusion ....................................................................................................................... 457    Appendices ...................................................................................................................... 459  Chapter 11: Bone, Antler, and Shell Artifacts ......................................................... 493    Analysis Methods ............................................................................................................ 493  iv 

         

Results ............................................................................................................................. 493  Functional Interpretations .............................................................................................. 509  Discussion........................................................................................................................ 510  Conclusions ..................................................................................................................... 511  Appendix ......................................................................................................................... 512  Chapter 12: Faunal Analysis by Brian Worthington ..................................................... 517    Materials and Methods ................................................................................................... 517    Results ............................................................................................................................. 522    Habitats and Exploitation Methods ................................................................................ 552    Seasonality ...................................................................................................................... 561    Comparisons with Other Sites ........................................................................................ 561    Conclusions ..................................................................................................................... 564    Appendices ...................................................................................................................... 566 

Chapter 13: The Sclerochronology of Hard Clams (Mercenaria spp.)    Excavated from Bayou Park by Irvy R. Quitmyer ................................................... 605           

Materials and Methods ................................................................................................... 609  Results ............................................................................................................................. 611  Discussion........................................................................................................................ 614  Conclusions ..................................................................................................................... 615  Appendix ......................................................................................................................... 616 

Chapter 14: Macrobotanical Materials from Features at Bayou Park    by Anna R. Dixon ............................................................................................................. 619       

Research Methods and Objectives ................................................................................. 619  Results ............................................................................................................................. 620  Conclusions ..................................................................................................................... 625 

Chapter 15: Interpretation of Archaeological Data: Bayou Park,    Elliott’s Point, and Beyond ..................................................................................... 629    Bayou Park: A Site History .............................................................................................. 629       

The Bigger Picture: Evaluating Elliott’s Point .................................................................. 647  Future Directions ............................................................................................................ 677  Conclusions ..................................................................................................................... 679  Chapter 16: Conclusion .................................................................................................. 681    References Cited ................................................................................................................... 683  Appendix A:  Memorandum of Agreement  Appendix B:  Updated Florida Master Site File Form  Appendix C:  FDHR Survey Log Sheet     



This page intentionally left blank.

vi 

LIST OF FIGURES    Figure 1.1.  Figure 1.2.  Figure 1.3.  Figure 1.4.  Figure 1.5.  Figure 1.6.  Figure 1.7.  Figure 1.8.  Figure 1.9.    Figure 1.10.    Figure 1.11.  Figure 1.12.  Figure 1.13.    Figure 1.14.    Figure 1.15.  Figure 1.16.  Figure 1.17.  Figure 1.18.    Figure 2.1.    Figure 2.2.  Figure 2.3.  Figure 2.4.  Figure 2.5. 

Location of 8OK898 within Eglin Air Force Base, Okaloosa County, Florida .... 3  Aerial photograph showing 8OK898 site boundaries within the former Eglin  Federal Prison Camp ......................................................................................... 4  Map of 1993 shovel testing at 8OK898 (from Campbell and Meyer 1993)...... 5  Map of 1994 shovel testing at 8OK898 (from Mathews et al. 1994) ............... 7  Map of 1996 shovel testing and test pit excavation at 8OK898 (from  Campbell et al. 1996) ........................................................................................ 7  Map showing the location of the 1997 shovel testing and test pit excavation  at 8OK898 (from Campbell and Mathews 1997) .............................................. 8  Map showing the location of the 1998 shovel testing at 8OK898 (from  Campbell et al. 1998) ........................................................................................ 8  Map showing the location of the 1998 block excavation at 8OK898 (from   Campbell et al. 1998) ........................................................................................ 9  Map showing the location of the 2002 shovel testing and test pit   excavation at 8OK898 (from Meyer et al. 2002) ............................................ 11  Map showing the location of the 2007 shovel testing and test pit .....................  excavation at 8OK898 (from Morehead et al. 2008) ...................................... 13  Map showing the location of the 2008 test pit excavation northeast of  Building 635 at 8OK898 (from Morehead and Aubuchon 2008) .................... 15  Map showing the location of the 2008 test pit excavation southwest of  Building 635 at 8OK898 (from Morehead and Aubuchon 2008) .................... 15  Map showing the location of the 2009 shovel testing and test pit  excavation at 8OK898 (from Campbell, Morehead, et al. 2009) .................... 17  Map showing the location of the 2010 shovel testing and test pit  excavation at 8OK898 (from Morehead et al. 2011) ...................................... 19  Map showing the location of the 2010 monitoring and the area at Building  591 where midden deposits were identified (from Speal and Nolan 2011) .. 20  Map of shovel tests excavated by PTA, 1993‐2010 ........................................ 21  Map of test pits excavated by PTA, 1993‐2010 .............................................. 22  Map showing Operations A, B, and C within site 8OK898 as well as the  locations of Strip Areas and Test Units. .......................................................... 25  Portion of USGS Destin, FL 7.5’ quadrangle map (top) showing the location  of 8OK898 within Eglin AFB and a topographic map (bottom) showing the  elevated hill system on which much of the Bayou Park site rests. ................ 32  Physiographic divisions and subdivisions around 8OK898 and  Choctawhatchee Bay (Brooks 1981) ............................................................... 33  Map of geological formations in the southeastern U.S. that contain  knappable toolstone ....................................................................................... 36  Portion of 1935 USGS Villa Tasso 15’ quadrangle map showing Postl Lake ... 39  USDA aerial photograph showing Postl Lake in 1955 ..................................... 40  vii 

Figure 2.6. 

Map of Eglin AFB showing the distribution of wetlands in relation to Bayou  Park (8OK898) ................................................................................................. 43  Figure 2.7.  Seagrass beds in Choctawhatchee bay ca. 1972 ............................................ 44  Figure 2.8.  Reconstructed Holocene sea level fluctuations for the Eglin Air Force  Base/Choctawhatchee Bay region .................................................................. 48  Figure 2.9.  Location of the Bayou Park site in relation to Poverty Point and other     related sites ..................................................................................................... 52  Figure 2.10.  Distribution of Late Archaic sites in the Choctawhatchee Bay region ........... 57    Figure 3.1.  Model of occupational intensity (after Lightfoot 1984) ................................. 75  Figure 3.2.  GPR grids within Operations A, B, and C ........................................................ 78  Figure 3.3.  GSSI GPR unit used for data collection at 8OK898 ......................................... 81  Figure 3.4.  View to the northeast of Operation A showing locations of the concrete  sewer marker, PTA’s 1998 block excavation, and SEARCH’s grid datum ....... 82  Figure 3.5.  Relative locations of SEARCH and PTA 1000n, 1000E datum points within  Operation A ..................................................................................................... 83  Figure 3.6.  View to the “north” of Strip Area K with hydraulic excavator ....................... 86  Figure 3.7.  Strip areas and test pit locations in Operation A ............................................ 87  Figure 3.8.  Strip areas and test pits locations in Operations A and C .............................. 88  Figure 3.9.  Strip areas and test pit locations in Operation B ............................................ 89  Figure 3.10.  Excavating TPs 59 and 61 inside demolished Building 588, Operation A ....... 93  Figure 3.11.  Block excavation in Strip Area Ja .................................................................... 93  Figure 3.12.  Documenting a shell‐filled pit feature (Feature 36) in Strip Area B ............... 94  Figure 3.13.  Water screening bulk feature sample through 1.6‐cm (1/16‐inch) mesh  screen .............................................................................................................. 96    Figure 4.1.  Generalized north‐south profile through Operation A based on profiles in  Strip Areas E, D, and K ................................................................................... 111  Figure 4.2.  Generalized east‐west profile through Operations A, C, and B based on   profiles in Strip Areas .................................................................................... 112  Figure 4.3.  East wall profile, Test Pit 59, Operation A .................................................... 113  Figure 4.4.  East wall profile, Test Pit 60, Operation A .................................................... 113  Figure 4.5.  West wall profile, Test Pit 62, Operation A .................................................. 114  Figure 4.6.  West wall profile, Test Pit 63, Operation A .................................................. 114  Figure 4.7.  South wall profile, Test Pit 85, Operation A ................................................. 115  Figure 4.8.  South wall profile, Test Pit 88, Operation B .................................................. 116  Figure 4.9.  South wall profile, Test Pit 94, Operation C .................................................. 117  Figure 4.10.  Plot of calibrated radiocarbon ages from Bayou Park .................................. 125  Figure 4.11.  Plot of calibrated calendrical dates from Bayou Park .................................. 126  Figure 4.12.  Examples of ceramic sherds from Bayou Park .............................................. 128  Figure 4.13.  Destin points and “Destin‐like” point from Bayou Park ............................... 131  Figure 4.14.  Examples of projectile points manufactured from broken proximal and     distal fragments ............................................................................................ 132  Figure 4.15.  Examples of diagnostic projectile points from Bayou Park .......................... 133  viii 

Figure 4.16.  Stratigraphic relationships between temporally diagnostic artifacts and  radiocarbon dates ......................................................................................... 138    Figure 5.1.  GPR grids within Operations A, B, and C ...................................................... 148  Figure 5.2.  Top) GPR survey Grid 1 with rectified image showing 1998 PTA block  excavation and mechanical stripping.  Bottom) The reflection profile    along Transect 201 ........................................................................................ 151  Figure 5.3.  Top) Strip Area Ja showing the location of Features 69/104 and 120.    Bottom) Amplitude maps from the western portion of Grid 1 .................... 152  Figure 5.4.  Reflection profiles from GPR Grid 1 .............................................................. 153  Figure 5.5.  Top) Feature 60 located next to a sewer pipe and trench in Strip Area Jb.    Bottom) Amplitude map of GPR Grid 1 ........................................................ 154  Figure 5.6.  Top) Strip Area E showing the top of the midden.  Bottom) Amplitude    map of the northern portion of GPR Grid 2 .................................................. 155  Figure 5.7.  Reflection profile from GPR Grid 6 ............................................................... 156  Figure 5.8.  Top) Strip Area K during mechanical excavation.  Bottom) Amplitude plan   view of the southern portion of GPR Grid 6 ................................................. 157  Figure 5.9.  Top) Profile of Feature 73 in Strip Area K.  Bottom) Reflection profile    from GPR Grid 6 ............................................................................................ 158  Figure 5.10.  Top) Amplitude map from GPR Grid 42.  Bottom) Similar disturbances are   seen in the profile of Test Pit 86, located just to the southeast of GPR    Grid 42 ........................................................................................................... 159  Figure 5.11.  Top)  A reflection profile from GPR Grid 14 with an area of disturbance    just below the ground surface highlighted.  Bottom)  The disturbance     was confirmed through mechanical excavation of Strip Area X ................... 161  Figure 5.12.  Side‐by‐side comparisons of plan views and profiles of Features 119    and 120 ......................................................................................................... 162    Figure 6.1.  View to the northeast of Operation A at Bayou Park showing Strip Areas    K and L (upper left), Ja (lower left), and E (lower right) in various stages of  work.  Strip Areas F and D (center right) have been backfilled and     graded ........................................................................................................... 166  Figure 6.2.  View to the east of Strip Area Ja showing large pit features exposed     during stripping ............................................................................................. 166  Figure 6.3.  View to the east of Strip Area Ja showing large pit features (left) and the   outlines of PTA’s 1998 block excavation (right) ........................................... 167  Figure 6.4.  Estimated extent of midden based on PTA’s 1998 data recovery project    (from Campbell et al. 1998) .......................................................................... 168  Figure 6.5.  East wall profile of TP60 showing discontiguous nature of shell in    “Stratum” IVb ................................................................................................ 169  Figure 6.6.  East wall profile of TP58 showing Stratum IVc ............................................. 169  Figure 6.7.  View to the southeast of Strip Area B showing dark sediments associated  with the top of Stratum IVa.  Arrows point to isolated shell deposits    (Stratum IVb) ................................................................................................. 170  ix 

Figure 6.8.    Figure 6.9.  Figure 6.10.    Figure 6.11.  Figure 6.12.  Figure 6.13.  Figure 6.14.  Figure 6.15.  Figure 6.16.  Figure 6.17.  Figure 6.18.  Figure 6.19.    Figure 6.20.  Figure 6.21.  Figure 6.22.    Figure 6.23.  Figure 6.24.    Figure 6.25.    Figure 6.26.    Figure 6.27.    Figure 6.28.  Figure 6.29.  Figure 6.30.  Figure 6.31.  Figure 6.32.      Figure 6.33.    Figure 6.34.     

Profile of north wall of Strip Area D showing dense midden deposits and  a large pit feature at far right ....................................................................... 170  Mapping of Stratum IVc in Strip Area A.  View is to the northeast .............. 171  Strip areas and midden deposits (identified and interpreted) in  Operation A ................................................................................................... 172  Topographic map of Operation A with midden boundaries ......................... 173  Plan and cross‐section shapes of features.................................................... 176  General distribution of features in Operations A and C ............................... 177  General distribution of features in Operation B ........................................... 178  Features identified in Strip Area A, Operation A .......................................... 179  Features identified in Strip Area B, 3.40‐3.27 m amsl, Operation A ............ 181  Features identified in Strip Area B, 3.09‐2.97 m amsl, Operation A ............ 182  Feature 27 in Strip Area C, Operation A ....................................................... 185  Features identified in Strip Area D and TP 60, 2.95‐2.90 m amsl,  Operation A ................................................................................................... 186  Features identified in Strip Area D, 2.88‐2.80 m amsl, Operation A ............ 187  Features identified in Strip Area D, 2.71‐2.65 m amsl, Operation A ............ 188  Features identified in Strip Area D and TP 59, 2.47‐2.29 m amsl,   Operation A ................................................................................................... 189  Representative plan views of Feature 44 at various elevations ................... 191  Strip Area E at an elevation of 2.84 m showing the location of Feature  46 .................................................................................................................. 195  Strip Area E at an elevation of 2.49 m showing the location of Feature  47 .................................................................................................................. 196  Strip Area G at an elevation of 2.74 m showing the location of Feature  32 .................................................................................................................. 197  Features identified in Strip Area H at 3.34 m (top) and 2.86‐2.74 m  (bottom) amsl, Operation A .......................................................................... 198  Features identified in Strip Area Ja, 3.20‐3.09 m amsl, Operation A ........... 200  Features identified in Strip Area Ja, 2.98‐2.83 m amsl, Operation A ........... 201  Features identified in Strip Area Ja, 2.62‐2.41 m amsl, Operation A ........... 202  Features identified in Strip Area Ja, 2.12‐1.86 m amsl, Operation A ........... 203  Reproduction of PTA’s 1998 feature map showing the locations of  Features 1, 2, 3, and Post Hole 1 (adapted from Campbell et al.  1998:Figure 24) ............................................................................................. 205  Profiles of postmolds/possible postmolds associated with a small  structure in Strip Area Ja, Operation A ......................................................... 207  SEARCH excavation block in Strip Area Ja, Operation A.  Bottom   illustration shows unit numbers with shaded squares indicating  excavated units.  Top illustration shows the relationship of the backhoe  trench and features to the block of units ..................................................... 208  Figure 6.35.  View to the east‐northeast of feature excavations in Strip Area Ja,    Operation A ................................................................................................... 210  Figure 6.36.  “East” profile of Feature 106A ...................................................................... 210  x 

Figure 6.37.  Figure 6.38.  Figure 6.39.    Figure 6.40.  Figure 6.41.  Figure 6.42.  Figure 6.43.    Figure 6.44.  Figure 6.45.    Figure 6.46.  Figure 6.47.  Figure 6.48.  Figure 6.49.    Figure 6.50.    Figure 6.51.    Figure 6.52.    Figure 6.53.  Figure 6.54.    Figure 6.55.    Figure 6.56.  Figure 6.57.  Figure 6.58  Figure 6.59.  Figure 6.60.      Figure 6.61.      Figure 6.62. 

View to the southeast of Features 69/104, 120, and 64 .............................. 211  View to the south‐southwest of Features 64 and 120 ................................. 212  Profiles of Features 69/104 and 120: top) “north” wall of TPs 65, 70, and   71 across the backhoe trench; bottom) “west” wall of TPs 74, 75, and 76 . 212  Profile of Feature 120 showing layer of baked‐clay objects ......................... 214  Features identified in Strip Area Jb North, Operation A ............................... 216  Features identified in Strip Area Jb South, Operation A ............................... 217  View to the northeast of the top of Feature 60 after exposure in Strip  Area Jb North, Operation A ........................................................................... 218  Profile of Feature 60 showing various depositional lenses.  Compare with   drawn profile in Figure 6.41 .......................................................................... 218  “West” profile of sample test in Feature 60.  Arrow points to Lens 5  pocket that contained an incised sherd ........................................................ 219  Features identified in Strip Area K, 3.23‐2.94 m, Operation A ..................... 222  Features identified in Strip Area K, 2.25‐2.19 m, Operation A ..................... 223  North, west, and south wall profiles of PTA’s ST162 showing Feature 5.   Reproduced from Meyer et al. 2002:Figure 20 ............................................ 224  Plan view of PTA’s TP 19 showing Features 6 and 7.  Reproduced from  Meyer et al. 2002:Figure 21 .......................................................................... 225  Features identified in Strip Area L and L‐North, 2.70‐2.57 m, Operation  A .................................................................................................................... 226  Features identified in Strip Area L and L‐North, 2.28‐1.94 m, Operation  A .................................................................................................................... 227  Profiles of postmolds and possible postmolds in Strip Area L, Operation  A .................................................................................................................... 228  Features identified in Strip Area Q, Operation A .......................................... 231  Progression of work in Strip Area Y showing expansions and interior  trenches ........................................................................................................ 232  Features identified in Strip Area Y at elevations 3.34‐2.98 m, Operation  A .................................................................................................................... 233  Features identified in Strip Area Y, 2.89‐2.70 m amsl, Operation A ............. 234  Features identified in Strip Area Y, 2.83‐2.70 m and 2.20‐2.16 m amsl,   Operation A ................................................................................................... 235  “West” profile of Feature 137, Strip Area Y, Operation A ............................ 237  South wall profile of Trench 4, Strip area Y, Operation A showing the   relationship between Features 143 and 129/142 ........................................ 238  View to the west‐northwest of disturbed (darker) sediments containing  a concentration of ceramic sherds in Strip Area Y at elevation 3.08 m.  Note evidence of disturbance in the “west” wall of Strip Area Y ................. 239  Plan drawing of F14 and F17(P) in TP47, Building 632, Operation A.  14A   represents a concentration of charcoal and fired clay; 14B and 14C  represent piles of fired clay.  Reproduced from Campbell et al.  (2009:Figure 26) ............................................................................................ 241  PTA’s TP20 exposed in Strip area M, Operation B ........................................ 242  xi 

Figure 6.63.  Figure 6.64.  Figure 6.65.      Figure 6.66.    Figure 6.67.    Figure 6.68.    Figure 6.69.  Figure 6.70.  Figure 6.71.  Figure 6.72.  Figure 6.73.        Figure 6.74.  Figure 6.75.    Figure 6.76.    Figure 6.77.    Figure 6.78.  Figure 6.79.    Figure 6.80. 

Figure 6.81.  Figure 6.82.      xii 

Features identified in Strip Area M, Operation B ......................................... 243  Features identified in Strip Area Pb, Operation B ........................................ 244  View of west wall of the water main trench north of Building 635 showing  the black sandstone feature.  Note the darker soil strata above and below  the sandstone.  Photograph reproduced from Eglin Air Force Base 2003 ... 245  “North” wall profile of Strip Area Wa and TP95 showing modern intrusive  pit intercepting previously deposited fill layers.  Inset shows possible track  marks in the profile of TP95 .......................................................................... 246  “North” wall profile of Strip Area Wa, TP97, and TP95 showing  approximate original location of sandstone “feature” and the extent of a  modern intrusive pit ..................................................................................... 247  Mass spectrum for the cholestanol peak in Feature 69/104 (top) and the  NIST mass spectrum for cholestanol (bottom) ............................................. 250  Size ranges of postmolds and postholes: top) maximum length; bottom)   maximum width ............................................................................................ 252  Size and depth ranges of pit features: top) maximum length; bottom)   maximum depth ............................................................................................ 254  Frequency distribution of features by elevation in meters amsl ................. 256  Relationship of ground surface elevations and elevations of features ........ 256  Vertical relationship of ground surface and features in the midden area  of Operation A: top) ground surface and features; bottom) ground   surface and calibrated median radiocarbon ages before present (BP) as  calculated by CALIB6.01.  Dashed lines indicate the range of elevations  where 50% of the features were first identified .......................................... 257  Distribution of feature types and midden deposits in Operation A ............. 258  Scatterplot of pit volume versus elevation indicating no significant   correlation between the two variables ........................................................ 259  Distribution of pits, possible pits, pits and/or postholes by volume,   midden area, Operation A ............................................................................ 260  Structure 1, Strip Area Jb: a) postmold arrangement showing two   concentric lines of posts; b) same arrangement but with postmolds  containing charcoal highlighted .................................................................... 262  Structure 2, Strip Area L, Operation A .......................................................... 263  Topographic map of Meig’s Pasture site (8OK102) showing semi‐circular  arrangement of midden deposits.  Reproduced from Curren   (1987:Figure 3) .............................................................................................. 264  Comparison of the cumulative percentage distributions of major faunal  classes by MNI for midden samples (Strata IVa and IVb) and analyzed  features ......................................................................................................... 268  Bivariate plot of richness and evenness values for invertebrate taxa in  midden strata (IVa and IVb) and analyzed features ..................................... 271  Comparison of most common invertebrate taxa by weight (≥ 100 g total  for all five features) in .64‐cm fractions with taxa from F4 .......................... 273 

Figure 7.1.  Figure 7.2.    Figure 7.3.    Figure 7.4.  Figure 7.5.  Figure 7.6.  Figure 7.7.    Figure 7.8.  Figure 7.9.  Figure 7.10.    Figure 7.11.  Figure 7.12.    Figure 7.13.  Figure 7.14.  Figure 7.15.    Figure 7.16.    Figure 7.17.  Figure 7.18.    Figure 7.19.  Figure 7.20.  Figure 7.21.  Figure 7.22.  Figure 7.23.      Figure 7.24.  Figure 7.25.   

Formation of ellipses by the intersection of a cone with planes inclined at   various angles ............................................................................................... 294  Idealized examples of a spheriod (left), oblate ellipsoid (center), and  prolate ellipsoid (right) ................................................................................. 295  Examples of grooved (left) and finger‐impressed (right) spheroid   baked‐clay objects (Thomas and Campbell 1991:Figure 6) .......................... 295  Example of an oblate ellipsoid (after Small 1966:74) ................................... 296  Example of a prolate ellipsoid baked‐clay object (Thomas and Campbell  1991:Figure 6) ............................................................................................... 296  Examples of amorphous baked‐clay objects (after Small 1966:76) ............. 297  Example of tabular baked‐clay object (after Campbell et al.   998:Figure 29) ............................................................................................... 297  Summary of baked‐clay object forms ........................................................... 304  Front and back views of grooved spheroid from Feature 22 (FS 115) ......... 304  Examples of oblate ellipsoids: a) Feature 105, FS481; b) TP61, FS76;   c) TP58, FS7 ................................................................................................... 307  Example of prolate ellipsoid, TP6, FS478_126 .............................................. 308  Fragments of cylindrical baked‐clay objects: a) Feature 44, FS261, side  and top views; b) Feature 120, FS617, note central hole ............................. 308  Tabular objects that display one flat surface and one rough surface: top)   Feature 44, FS237; bottom) Feature 105, FS485 .......................................... 307  Amorphous baked‐clay object with finger impressions from TP60, FS46 .... 310  Amorphous baked‐clay object showing palm and fingernail impressions  from TP92, FS533 .......................................................................................... 310  Examples of amorphous baked‐clay objects: a) TP40, FS090028_332; b)  TP7, FS 478_140; c) TP1, FS478_194 ............................................................ 311  Results of principal components analysis of baked‐clay objects and clay   samples ......................................................................................................... 313  Results of principal components analysis showing 95% confidence   ellipses around the data points .................................................................... 314  Comparison of Bayou Park and Choctawhatchee Bay baked‐clay object   samples ......................................................................................................... 314  Spatial distribution of baked‐clay objects by weight in grams ..................... 315  Proveniences containing greater than one percent of the total weight of   baked‐clay objects at 8OK898 ...................................................................... 317  Feature 120, a clay‐lined cooking or roasting pit, with baked‐clay objects   in situ ............................................................................................................. 318  Baked‐clay griddle fragments from AR‐39, Rio Tanamá, Puerto Rico.    Note the flat upper surfaces (left) and irregular bottom surfaces (right).   Compare with Figure 7.14.  Adapted from Pagán Jiménez   2008:Figure 11.1 ........................................................................................... 320  Sherd‐like baked‐clay objects from Feature 60 ............................................ 321  Seriation of baked‐clay objects from three sites in the Choctawhatchee  Bay region ..................................................................................................... 330  xiii 

  Figure 8.1.    Figure 8.2.    Figure 9.1.  Figure 9.2.    Figure 9.3.    Figure 10.1.    Figure 10.2.    Figure 10.3.  Figure 10.4.  Figure 10.5.    Figure 10.6.  Figure 10.7.  Figure 10.8.  Figure 10.9.  Figure 10.10.  Figure 10.11  Figure 10.12.  Figure 10.13.  Figure 10.14.  Figure 10.15.  Figure 10.16.    Figure 10.17.    Figure 10.18.  Figure 10.19.    Figure 10.20. 

xiv 

Phytolith and Starch diagram for a griddle fragment from Bayou Park,   8OK898 .......................................................................................................... 355  Micrographs of phytoliths and starch recovered from a possible griddle   fragment, Bayou Park site (8OK909) ............................................................ 357  Examples of decorated sherds from 8OK898 ............................................... 362  Swift Creek Complicated Stamped sherd with weathered surface and   pencil rubbing showing the stamped design ................................................ 364  Vessel reconstruction of two rim sherds from 8OK898 ............................... 368  Locations of various geological formations containing toolstone in   relation to Bayou Park .................................................................................. 386  Distribution of major ocean currents within the Atlantic and Gulf of  Mexico basins ................................................................................................ 391  Distribution of major ocean currents within the Caribbean and Gulf of   Mexico ........................................................................................................... 391  Westo points from Bayou Park ..................................................................... 394  Left: Biface scavenging and reuse.  Right:  Examples of biface reuse at   Bayou Park .................................................................................................... 394  Little Bear Creek points from Bayou Park ..................................................... 396  Miscellaneous stemmed bifaces ................................................................... 397  Miscellaneous stemmed bifaces ................................................................... 400  Unstemmed bifaces ...................................................................................... 403  Biface fragments ........................................................................................... 405  Recycled biface fragment, CR‐051‐11‐281 ................................................... 406  Microliths from Bayou Park .......................................................................... 408  Examples of utilized flakes ............................................................................ 411  Examples of cores ......................................................................................... 413  Histogram showing the variation in flake weight between Tallahatta  Quartzite and chert ....................................................................................... 417  Cumulative percentage distributions of flake‐form categories, Operations  A and B .......................................................................................................... 419  Comparison of cumulative percentage distributions of flake‐form   categories, Strata IVa‐c and VI with experimentally produced waste flake  assemblages .................................................................................................. 419  Comparison of flake‐size distributions, Operations A and B, with   experimentally produced waste flake assemblages ..................................... 420  Comparison of flake‐size distributions, midden (Strata IVa‐c) and  non‐midden (Stratum VI), with experimentally produced waste flake  assemblages .................................................................................................. 421  Comparison of flake‐size distribution from Bayou Park (midden and non‐  midden combined) with other archaeological assemblages demonstrating   effect of differential access to raw materials on flake size .......................... 422 

Figure 10.21.  Figure 10.22.  Figure 10.23.  Figure 10.24. 

Examples of cobble tools .............................................................................. 425  Examples of cobble tools .............................................................................. 426  Non‐cobble ground stone tools .................................................................... 429  Vertical distribution of lithic artifacts in Stratum VI in relation to feature   depths in Operation A, Strip Area Q, and Operations B and C ..................... 438  Cumulative distribution of artifact classes in unmixed strata and   features ......................................................................................................... 438  Lithic density distribution (artifacts per square meter) in shovel tests and  test pits at Bayou Park .................................................................................. 439  Distribution of lithic artifacts by depth ......................................................... 441  Spatial distribution of lithic tools in Operation A ......................................... 442  Spatial distribution of lithic tool in Operation B ........................................... 443  Distribution of unmodified stone at Bayou Park .......................................... 444  Comparison of richness and evenness values for lithic tool assemblages  from selected sites ........................................................................................ 450  Vertical distribution of stemmed bifaces ..................................................... 452  Hypothetical models of resource extraction for Bayou Park ....................... 456 

Figure 10.25.    Figure 10.26.    Figure 10.27.  Figure 10.28.  Figure 10.29.  Figure 10.30.  Figure 10.31.    Figure 10.32.  Figure 10.33.    Figure 11.1.  Awls and a possible awl or dagger from Bayou Park .................................... 495  Figure 11.2.  Spatula and pins from Bayou Park ................................................................ 497  Figure 11.3.  Pointed fragments and a possible projectile point fragment from Bayou   Park ............................................................................................................... 498  Figure 11.4.  Antler artifacts from Bayou Park .................................................................. 499  Figure 11.5.  Type E cutting‐edged tool (FS 785) from Strip Area Y .................................. 501  Figure 11.6.  Method of hafting a Type E cutting‐edged tool ........................................... 501  Figure 11.7.  Type F cutting‐edged tool (FS 786) from Strip Area Y, front and back    views ............................................................................................................. 502  Figure 11.8.  Columella hammers (a‐b) and a possible awl or dagger (c): a) FS 460,    CS3, Strip Area L, Stratum VI; c) FS 305, Strip Area E, Stratum VI ................ 503  Figure 11.9.  Back and front views of an awl or perforator, FS 247, F48 .......................... 504  Figure 11.10.  Lightning whelk cup, FS 171, F36 .................................................................. 504  Figure 11.11.  Gastropod shell dippers ................................................................................ 505  Figure 11.12.  Shell debitage ................................................................................................ 506    Figure 12.1.  Comparison of proportional representation of biomass by faunal class in   midden column samples and features ......................................................... 548  Figure 12.2.  Bivariate plot of ubiquity versus MNI ranking for edible taxa ...................... 551  Figure 12.3.  Bivariate plot of ubiquity versus biomass ranking for edible taxa ............... 551  Figure 12.4.  Comparison of exploited habitats ................................................................. 555  Figure 12.5.  Comparison of canid tooth sizes ................................................................... 560    Figure 13.1.  Radial cross‐section of a hard clam (Mercenaria spp.) exposing light    (white) and dark (gray) incremental growth structures ............................... 606  xv 

Figure 13.2.  A backlit (A) and direct light (B) radial thin section  of a hard clam shell  (Mercenaria spp.) showing the translucent and opaque shell growth   increments .................................................................................................... 607  Figure 13.3.  Oxygen Isotope δ18O profile (A) showing the annual change in the  concentration of 18O/16O in the shell of a hard clam (Mercenaria  campechiensis) collected live in 1988 from Cedar Key, Florida .................... 608  Figure 13.4.  Six‐part subdivision of translucent and opaque shell growth in hard clam  (Mercenaria spp.) shells ................................................................................ 610  Figure13.5.  Percentage of monthly hard clam (Mercenaria campechiensis)     translucent and opaque shell growth ........................................................... 611  Figure 13.6.  Seasonal and annual growth frequency profiles of hard clams     (Mercenaria campechiensis) collected live 1987‐88 from Cedar Key,     Florida ........................................................................................................... 612  Figure13.7.  Growth frequency profile of hard clams (Mercenaria spp.) in pooled     sample from Bayou Park ............................................................................... 613  Figure13.8.  Growth frequency profile of hard clams (Mercenaria spp.) in Feature 36    from Bayou Park ........................................................................................... 614    Figure 15.1.  Comparison of radiocarbon‐dated occupations at Bayou Park with     Balsillie and Donoghue’s (2004) “Younger B” sea level model..................... 631  Figure 15.2.  Artist’s reconstruction of a closed circular structure (top) and an open  rectangular structure (bottom) similar to those identified at 8OK898 ........ 633  Figure 15.3.  Examples of types of earth ovens ................................................................. 634  Figure 15.4.  Bivariate graph comparing measures of residential stability and    use‐duration for Bayou Park (8OK898) and Montverde (8LA243) ............... 641  Figure 15.5.  Occupational chronology of Bayou Park based on radiocarbon dates and   artifacts recovered from archaeological projects conducted between     1993 and 2012 .............................................................................................. 646  Figure 15.6.  Comparisons of various stemmed bifaces identified as Destin points with   Bullen’s Levy point and Haisten’s Crooked Creek point ............................... 658  Figure 15.7.  Contour map of the Buck Bayou Mound (8WL90) showing the location     of the test pit excavated by New World Research ....................................... 661  Figure 15.8.  Distribution of Late Archaic sites in the Choctawhatchee Bay region ......... 666  Figure 15.9.  Radiocarbon dates in calibrated years BP (2 σ) for sites with Elliott’s Point  components .................................................................................................. 668  Figure 15.10.  Radiocarbon dates in calibrated years BC (2 σ) for sites with Elliott’s Point  components .................................................................................................. 669  Figure 15.11.  Proposed interaction spheres and trade routes 4500‐3000 BP ................... 675  Figure 15.12.  Distribution of steatite vessels and proposed exchange routes in Florida .. 676  Figure 15.13.  Distribution of galena and proposed exchange routes in Florida ................ 676   

xvi 

LIST OF TABLES    Table 1.1.    Table 2.1.  Table 2.2.  Table 2.3.    Table 3.1.  Table 3.2.  Table 3.3.  Table 3.4.  Table 3.5.    Table 4.1.    Table 4.2.  Table 4.3.  Table 4.4.  Table 4.5.    Table 4.6.  Table 4.7.  Table 4.8.    Table 6.1.  Table 6.2.  Table 6.3.  Table 6.4.  Table 6.5.  Table 6.6.  Table 6.7.  Table 6.8.    Table 6.9.    Table 6.10.  Table 6.11.  Table 6.12.   

Summary of radiocarbon dates from 8OK898 ................................................ 23  Native American culture sequence in the Choctawhatchee Bay region ........ 50  Elliott’s Point chronology and diagnostic traits .............................................. 53  Calibrated radiocarbon dates for Elliott’s Point sites ..................................... 54  Characteristic traits of the Elliott’s Point Complex ......................................... 71  Characteristics of Bayou Park site ................................................................... 72  GPR grid sizes and location data ..................................................................... 80  Size and volume data for mechanical strip areas ........................................... 90  Grid coordinates and elevation data for SEARCH shovel tests and test pits .. 91  Strata correlations between representative shovel tests and test pits,  midden area .................................................................................................. 110  Characteristics of soil samples from TP 60 ................................................... 120  Radiocarbon dates from Bayou Park ............................................................ 124  Proveniences of ceramic types from Bayou Park ......................................... 127  Proveniences of ceramic types recovered during PTA surveys and  excavation ..................................................................................................... 130  Vertical distribution of stemmed bifaces in Operation A ............................. 134  Vertical distribution of stemmed bifaces in Operations B and C ................. 135  Archaeological components represented at Bayou Park ............................. 136  Summary of feature types at Bayou Park ..................................................... 174  Cultural materials recovered from sample unit, Feature 60 ........................ 220  Initial and dry weights of feature sediment samples and control samples.. 247  Lipid fractions ................................................................................................ 248  C27 isomer standards for qualitative identification of sample peaks .......... 248  Results of lipid residue analysis .................................................................... 249  Morphological characteristics of pit features ............................................... 253  MNI by faunal classes for midden strata (IVa and IVb) and analyzed  features ......................................................................................................... 268  List of most common invertebrates (≥ 1% of total MNI) in midden strata  (IVa and IVb) and analyzed features.  The two most common taxa in each  sample are indicated by bold type ................................................................ 269  Richness (R) and evenness (E) values for midden strata (IVa and IVb) and   features ......................................................................................................... 271  Vertebrate remains from F4.  Data from Scott 2002:Table 9 ....................... 272  Invertebrate remains from F4.  Data from Meyer et al. 2002: Table 10 ...... 272 

xvii 

Table 7.1.  Table 7.2.  Table 7.3.  Table 7.4.  Table 7.5.  Table 7.6.  Table 7.7.  Table 7.8.  Table 7.9.  Table 7.10.  Table 7.11.    Table 7.12.  Table 7.13.  Table 7.14.    Table 9.1.    Table 10.1.  Table 10.2.  Table 10.3.    Table 10.4.  Table 10.5.  Table 10.6.  Table 10.7.  Table 10.8.  Table 10.9.  Table 10.10.  Table 10.11.  Table 10.13.  Table 10.14.  Table 10.15.    Table 10.16.  Table 10.17. 

Baked‐clay objects recovered from 2011‐2012 Phase III excavations at   8OK898 .......................................................................................................... 299  Baked‐clay objects recovered during previous investigations at 8OK898 ... 301  Primary aplastic inclusions ............................................................................ 302  Secondary aplastic inclusions ....................................................................... 302  Colors represented in the baked‐clay sample .............................................. 303  Baked‐clay object forms and surface treatments ......................................... 304  Provenience data for baked‐clay object forms ............................................. 305  Results of pXRF analysis in parts per million (ppm) ...................................... 312  Baked‐clay objects that display evidence of post‐firing exposure to heat ... 318  Radiocarbon dates associated with baked‐clay objects ............................... 324  Stratigraphic data for baked‐clay objects at 8OK898, 2011‐2012  excavation ..................................................................................................... 326  Stratigraphic data for baked‐clay objects at 8OK898, 1998 PTA block   excavation ..................................................................................................... 327  Stratigraphic data for baked‐clay objects at 8WL1278, all units combined . 328  Stratigraphic data for baked‐clay objects at 8OK877, Block 2, all units   combined ...................................................................................................... 329  Ceramic types from Bayou Park (8OK898) ................................................... 361 

Lithic artifacts recovered during previous projects at 8OK898 .................... 376  Proportional representation of flake types .................................................. 377  Cross‐tabulation of raw material by artifact class for PTA projects at   8OK898 .......................................................................................................... 378  Sample sizes from test units used in 8OK898 debitage analysis .................. 379  Lithic raw materials by artifact type ............................................................. 387  Total artifacts recovered from Bayou Park, 1993‐2012 ............................... 392  Data on stemmed biface, 8OK898 ................................................................ 393  Data on unstemmed bifaces, 8OK898 .......................................................... 403  Data on biface fragments, 8OK898 ............................................................... 404  Data on microliths from 8OK898 .................................................................. 408  Data on utilized flakes from 8OK898 ............................................................ 410  Data on cores from 8OK898 .......................................................................... 413  Summary data for lithic waste flakes ............................................................ 415  Results of Student’s t tests comparing mean flake weight between  Operation A and B and between unmixed strata ......................................... 416  Summary data for lithic waste flakes by raw material type ......................... 416  Results of Student’s t tests comparing mean flake weight between raw  material types in Operations A and B ........................................................... 416  Table 10.18.  Flake‐form data for Operations A and B by strata ........................................ 418  Table 10.19.  Flake‐size data for Operations A and B and for midden (Strata IVa‐c) and    non‐midden (Stratum VI) contexts ............................................................... 420  Table 10.20.  Technological flake types by strata ............................................................... 423 

xviii 

Table 10.21.  Table 10.22.  Table 10.23.  Table 10.24.  Table 10.25.  Table 10.26.  Table 10.27.  Table 10.28.  Table 10.29.  Table 10.30.  Table 10.31.    Table 10.32.    Table 10.33.    Table 10.34.      Table 11.1.  Table 11.2.  Table 11.3.    Table 12.1.  Table 12.2.    Table 12.3.  Table 12.4.  Table 12.5.  Table 12.6.  Table 12.7.  Table 12.8.  Table 12.9.  Table 12.10.    Table 12.11.    Table 12.12.    Table 12.13.    Table 12.14.  Table 12.15.   

Waste flakes of non‐siliceous stone ............................................................. 423  Cross‐tabulation of platform preparation and platform condition .............. 424  Data on cobble tools from 8OK898 .............................................................. 426  Data on non‐cobble ground stone artifacts .................................................. 428  Distribution of unmodified sandstone by strata in Operations A and B ...... 431  Provenience data for unmodified stone (non‐sandstone) ........................... 431  Cross‐tabulation of functional classes by artifact types ............................... 433  Distribution of lithic artifact classes by strata .............................................. 435  Distribution of lithic artifacts by unmixed strata .......................................... 436  Vertical distribution of artifacts in Stratum VI by depth .............................. 437  Comparison of observed and expected frequencies of lithic artifacts by  strata and features, and chi‐square results .................................................. 439  Comparison of the degree of retouch on tool forms from midden (Strata  IVa‐c), non‐midden (Strata VI), features, and other site contexts ............... 446  Comparison of degree of retouch among several sites in chert rich and  chert poor areas of central and south Florida .............................................. 447  Comparison of functional tool types between 8OK898 and a sample of   sites from peninsular Florida ........................................................................ 449  Provenience data for organic artifacts ......................................................... 494  Cross‐tabulation of artifact function by species ........................................... 494  Shell artifacts recovered during PTA 1998 block excavation ........................ 508  Faunal samples from Bayou Park (8OK898) chosen for analysis .................. 518  Subsampled and extrapolated totals for the 1/8‐inch and 1/16‐inch   fractions of shell hash from Features 36 and 37 .......................................... 520  Allometric constants used to calculate biomass (g) from bone weight (g) .. 521  List of taxa identified in analyzed samples ................................................... 523  Summary of faunal data by class for Feature 36 .......................................... 527  Summary of faunal data by class for Feature 37 .......................................... 529  Summary of faunal data by class for Feature 44 .......................................... 530  Summary of faunal data by class for general levels in Feature 44 ............... 531  Summary of faunal data by class for Feature 46 .......................................... 532  Faunal classes represented in bulk samples from four features separated  by screen‐size fractions ................................................................................. 534  Combined faunal data for Features 36, 37, 44, and 46 bulk samples  separated by screen‐size fractions ............................................................... 537  Fauna recovered from Feature 60 separated by provenience and   recovery mesh size ........................................................................................ 538  Summary of faunal data by class for Feature 60, all proveniences  combined ...................................................................................................... 539  Fauna recovered from Feature 139 separated by levels .............................. 540  Summary of faunal data by class for Feature 139, all proveniences   combined ...................................................................................................... 540  xix 

Table 12.16.  Table 12.17.  Table 12.18.    Table 12.19.    Table 12.20. 

Fauna recovered from CS1 and CS2 separated by stratum and level .......... 542  Summary of faunal data by class for Column Samples 1 and 2 combined ... 542  Fauna recovered from TP60, general excavation levels, separated by  stratum and level .......................................................................................... 544  Summary of vertebrate faunal data by class for TP60, all general   excavation levels combined .......................................................................... 544  Comparison of Biomass and MNI between midden column samples and   features ......................................................................................................... 547  Ranking of edible taxa by ubiquity and abundance ...................................... 549  Comparison of burned fauna by class........................................................... 552  Comparison of burned fauna between midden column samples and   features ......................................................................................................... 552  Comparison of exploited habitats based on fauna recovered from Bayou  Park ............................................................................................................... 554  Comparison of fauna and exploited habitats between Bayou Park,   Mitchell River 1, and Meig’s Pasture ............................................................ 562 

Table 12.21.  Table 12.22.  Table 12.23.    Table 12.24.    Table 12.25.      Table 13.1.  Summary data on incremental growth phases in the shells of hard clams  (Mercenaria spp.) from 8OK898 ................................................................... 613    Table 14.1.  Summary of macrobotanical remains from 8OK898 .................................... 621  Table 14.2.  Summary of macrobotanical remains from Feature 26 ............................... 621  Table 14.3.  Summary of macrobotanical remains from Features 36, 37, 44, and 45 ..... 622  Table 14.4.  Summary of macrobotanical remains from Feature 60 ............................... 623  Table 14.5.  Summary of macrobotanical remains from Feature 106 ............................. 623  Table 14.6.  Summary of macrobotanical remains from Feature 22 ............................... 623  Table 14.7.  Summary of macrobotanical remains from Feature 105 ............................. 624  Table 14.8.  Summary of macrobotanical remains from Feature 120 ............................. 624  Table 14.9.  Summary of macrobotanical remains from Features 38 and 51 .................. 625  Table 14.10.  Summary of macrobotanical remains from all samples ............................... 626    Table 15.1.  Seasons of occupation based on various faunal and botanical indicators ... 637  Table 15.2.  Comparison of feature types at Bayou Park, 8OK898, and Montverde,    8LA243........................................................................................................... 641  Table 15.3.  Data used in the calculation of residential stability and use‐duration     indices ........................................................................................................... 642  Table 15.4.  Radiocarbon dates for fiber‐tempered sites east of Choctawhatchee Bay . 670  Table 15.5.  Comparison of the proportional representation of baked‐clay object     varieties by time period ................................................................................ 671   

xx 

LIST OF ACRONYMS    AFB  amsl  ARPA  BP  cal  cmbs  DHR  DoD  EPO  FAC  FS  FR  GIS  GPR  GPS  HPP  ICRMP  MNI  NAGPRA  NEPA  NHPA  NISP  NPS  NRCS  NRHP  NWR  PRI  PTA  SEARCH  SHPO  USDA 

Air Force Base  above mean sea level  Archaeological Resources Protection Act  Before Present  calibrated  centimeters below surface  Division of Historical Resources  Department of Defense  Elliott’s Point Object  Florida Administration Code  Field Sample  Federal Register  Geographic Information Systems  Ground Penetrating Radar  Global Positioning System  Historic Preservation Plan  Integrated Cultural Resources Management Plan  Minimum Number of Individuals  Native American Grave Protection and Repatriation Act  National Environmental Policy Act  National Historic Preservation Act  Number of Individual Specimens  National Park Service  Natural Resource Conservation Service  National Register of Historic Places  New World Research, Inc.  Paleo Research Institute, Inc.  Prentice Thomas & Associates, Inc.  Southeastern Archaeological Research, Inc.  State Historic Preservation Officer  United States Department of Agriculture 

xxi 

This page intentionally left blank.

xxii 

ACKNOWLEDGMENTS    SEARCH  would  like  to  acknowledge  Mark  Stanley,  Lynn  Shreve,  and  Joe  Meyer  from  the  96th  Civil  Engineer  Group  Cultural  Resources  Section,  Eglin  Air  Force  Base,  for  their  assistance  throughout  this  project.    Additionally,  Cat  Nolan  of  the  Cultural  Resources  Section  provided  assistance  with  access  to  previously  excavated  collections  and  George  Cole, also with Eglin’s Cultural Resources Section, provided previous data and maps, as well  as  guidance  regarding  use  of  the  Eglin  artifact  database.      Christopher  Serbe  Eglin  AFB  Contract  Specialist  was  especially  helpful  in  getting  access  for  contractors  and  staff,  sometimes on short notice.  Joel Spaulding with the Real Property Office obtained access for  SEARCH to use Building 584 for a field office and lab.  Paul Steffen, SABER project manager,  was  a  regular  visitor  to  the  site,  and  while  anxious  to  move  forward  with  the  final  demolition  of  Buildings  588  and  591,  was  enthusiastic  about  and  interested  in  the  archaeology of the site.    SEARCH also wishes to thank several professional colleagues who offered assistance, data,  and advice during the project.  Janice Campbell with Prentice Thomas and Associates, Inc.  (PTA) graciously provided copies of PTA reports and answered questions regarding past PTA  excavation  efforts  at  Bayou  Park.    Nancy  White  of  the  University  of  South  Florida  (USF)  Anthropology  Department  assisted  in  the  identification  of  ceramic  sherds  from  the  site,  provided copies of articles, and offered advice and encouragement.  Tom Pluckhahn of USF  also  helped  in  the  identification  of  ceramic  sherds.    Mike  Russo  of  the  National  Park  Service’s Southeastern Archaeological Center provided references and advice on the fauna.   Robert Tykot of USF performed the X‐Ray Fluorescence Spectrometry analysis of a sample  of  baked‐clay  objects  and  USF  doctoral  candidate  Rheta  Lanehart  conducted  the  lipid  residue analysis of sediments from selected features.  Harley Means with the Florida Bureau  of  Geology  answered  questions  about  the  regional  geology.    Specialized  analyses  of  the  sclerochronology of hard clams was performed by Irv Quitmyer of the Florida Museum of  Natural History, archaeobotanical analysis was performed Anna Dixon of USF, and phytolith  and  starch  analysis  of  a  griddle  fragment  was  performed  by  Chad  Yost  of  PaleoResearch,  Inc.    Radiocarbon  dating  was  performed  by  Beta  analytic,  Inc.    Finally,  we  would  like  to  thank Al Miller with Geomill, Inc. for his skillful operation of the hydraulic excavator and for  participating enthusiastically throughout the field work portion of this project.    It  was  a  pleasure  to  work  with  all  of  these  individuals  and  institutions  and  to  be  able  to  share their commitment to and appreciation of the prehistory of Eglin Air Force Base. 

xxiii 

SEARCH PROJECT PARTICIPANTS  Field Team  Robert Austin  Josh Torres  Chris Mickwee  Chris Altes  Kendra Crain  Josh Foster  Linda Geary  Matt Hanks  Rockie Jarvis  Tom Peterson  Jon Simon Suarez  Dennis Wardlaw    Laboratory Analysis  Debra Wells  Jon Simon Suarez  Robert Austin  Travis Colliette  Josh Torres  Brian Worthington 

Report Preparers  Robert Austin   Chris Mickwee  Josh Torres  Dennis Wardlaw  Brian Worthington  GIS Mapping, Photography, Graphics  Curtis Deily  Robert Austin  Linda Geary  Peer Review and Technical Editing  Lisabeth Carlson  Jennifer Salo  Public Outreach  Emily Powlen     Meg Gaillard      Project Management  Anne Stokes  James Pochurek  Debbie Lowry  

       

xxiv 

EXECUTIVE SUMMARY    Cultural Resources Data Recovery was conducted at the Bayou Park archaeological site (aka  Former  Eglin  Prison  Camp  site),  8OK898,  located  on  Eglin  Air  Force  Base  (Eglin  AFB),  Okaloosa  County,  Florida  by  Southeastern  Archaeological  Research,  Inc.  (SEARCH).  The  project was conducted according to the Scope of Work (SOW) included under Contract No.  EAFB‐62221‐008, Task Order No. CR‐11‐0051.  The purpose of this contract was to conduct  archaeological  data  recovery  at  the  Bayou  Park  site,  a  prehistoric  archaeological  site  that  has been determined eligible for listing in the National Register of Historic Places (NRHP).  The  work  was  carried  out  in  fulfillment  of  responsibilities  to  consider  the  effects  of  construction  activities  on  this  NRHP‐eligible  archaeological  site.    The  Mid‐Bay  Bridge  Authority  was  the  contracting  authority  for  this  project  and  Eglin  AFB  provided  technical  oversight.  Mandating legislation and directives include the NHPA of 1966, as amended; the  National  Environmental  Policy  Act  of  1969  (NEPA);  36  CFR  Part  800;  the  Archaeological  Resources  Protection  Act  of  1979  (ARPA);  the  Native  American  Graves  Protection  and  Repatriation Act of 1990 (NAGPRA); Air Force Instruction 32‐7065; and all laws, regulations  and guidelines promulgated by the State of Florida governing cultural resources work.    The background research for this project involved a review of all previous work conducted  at 8OK898 as well as literature regarding the Elliott’s Point Complex in Florida and the Late  Archaic generally.  Copies of previous cultural resource studies at the site, as well as copies  of  original  field  notes  and  maps,  were  provided  by  the  Eglin  AFB’s  Office  of  Cultural  Resources.    These  also  were  used  to  obtain  location  information  for  all  previous  shovel  tests,  test  pits,  block  excavations,  and  features  that  had  been  previously  documented  at  8OK898.  Additional reports were obtained from the Florida Master Site File in Tallahassee,  the SEARCH library in Newberry, and the libraries at the University of Florida and University  of  South  Florida.    In  addition  to  a  review  of  the  literature,  the  artifact  collections  from  previous  surveys  and  excavations  curated  at  Eglin  AFB  were  examined  and  a  sample  of  baked clay objects and lithic artifacts was reanalyzed.     A ground‐penetrating radar (GPR) survey of the site was conducted in an effort to identify  subsurface  midden  deposits,  possible  features,  modern  utility  lines,  and  other  subsurface  anomalies.  The GPR was successful in identifying the larger intact midden deposits as well  as  some  of  the  larger  subsurface  features;  however,  the  majority  of  features  were  not  identifiable.      Once  the  GPR  survey  was  completed,  1‐x‐2‐m  test  units  were  excavated  within  the  footprints  of  two  demolished  buildings,  588  and  591,  followed  by  mechanical  stripping  within and surrounding these buildings.  Work began here because of the need to remove  footers  and  debris  from  the  buildings.    Once  these  buildings  were  cleared  for  debris  removal,  mechanical  stripping  proceeded  to  the  east  across  the  courtyard.    Mechanical  stripping  was  performed  in  30  separate  areas  within  8OK898  for  a  total  volume  of  approximately 4670 cubic meters.  A total of 30 test excavation units were excavated: 11 1‐

x‐1‐m,  10  1‐x‐2‐m,  8  2‐x‐2‐m,  and  1  .5‐x‐1‐m.    The  total  volume  of  excavated  sediments  from  these  units  was  74.86  cubic  meters.    Three  column  samples  (.5‐x‐.5  m)  also  were  excavated and soil samples from features and strata were obtained for laboratory analysis.   Finally, 131 features were identified and mapped, including 112 prehistoric features, and a  representative  sample  was  excavated.    The  excavations  resulted  in  the  recovery  of  over  13,000 artifacts and over 500,000 faunal remains.  The field work was considered complete  when  a  level  of  data  redundancy  was  reached  such  that  additional  excavation  was  considered unnecessary for the goal of mitigating adverse effect.      Artifact  processing  and  initial  sorting  was  accomplished  in  the  field.    Final  sorting  and  detailed  analysis  was  conducted  at  the  SEARCH  laboratory  in  Newberry  and  by  various  specialists.  Analyses continued through January 2013.    In brief, the results indicate that the Bayou Park site was occupied primarily during the Late  Archaic  period  4481‐3766  cal  BP  (2529‐1817  cal  BC),  with  minor  occupations  occurring  during  the  Middle  Archaic,  Middle  to  Late  Woodland,  and  Mississippi  periods.    The  site  consists  of  a  semi‐circular  midden  deposit  situated  on  a  slight  topographic  rise  in  the  southwest  portion  of  the  site  and  a  moderately  dense  lithic  scatter  in  the  far  eastern  portion of the site near Postl Lake.  In between these two areas is a relatively low‐density  scatter of lithic artifacts, primarily waste flakes.      The midden covers an area of approximately 1500 square meters and overlooks a wetland  from  which  a  small  stream  originates.    The  midden  consists  of  dark,  organic‐stained  sediments  containing  animal  bone,  marine  shell,  and  artifacts.    Shell  concentrations  are  evident within the darker midden matrix representing individual episodes of shell disposal  which eventually began to overlap and coalesce in some parts of the site.   The semi‐circular  shape is the result of the structured organization of living space around a central “plaza” or  common area where several very large pit features and a large clay‐lined hearth and earth  oven  were  identified.    Other  large  pits  are  located  around  the  perimeter  of  the  midden.   Smaller pit features are more randomly distributed and are located within or immediately  below the midden deposits.  Some pits were clearly used for refuse disposal while others  may  have  originally  been  used  for  storage.    Lipid  analysis  of  sediment  samples  from  five  features  revealed  the  presence  of  cholestanol  and  epicoprostanol,  which  results  from  the  degradation  of  terrestrial  and  marine  animal  remains.    Plant  sterols  diagnostic  of  hickory  nuts also were identified.      Several post molds and postholes were identified and mapped leading to the recognition of  two possible structures.  One oval‐shaped structure measures no more than 2 or 2.5 m in  diameter,  although  it  was  not  entirely  exposed.    The  second  structure  is  larger  and  rectangular in shape, with a large open front and an associated hearth.  It measures about 7  m  across  and  about  2  m  deep.    The  small  sizes  of  the  structures  and  their  relatively  insubstantial construction suggests they were inhabited by small family groups.      xxvi 

The lithic concentration near Postl Lake consists of at least two high density areas separated  by  a  moderate  to  low‐density  scatter  of  artifacts.    One  high‐density  area  was  excavated  during  previous  investigations  at  8OK898  and  includes  a  large  pit  containing  marine  shell  and lithic tools.  Another high‐density area was identified by SEARCH and contains primarily  lithic waste flakes.  Five other features also were identified in this part of the site.  These are  mostly  refuse  pits  of  various  sizes,  although  the  function  of  two  features  could  not  be  determined.     A total of 16 radiocarbon dates (four obtained during earlier investigations) identified three  temporal occupations while two other possible occupations were identified on the basis of  stratigraphy  and  temporally  diagnostic  artifacts.    The  primary  occupation  was  during  the  Late Archaic period with 14 dates having a maximum two‐sigma range of 4481‐3766 cal BP  (2529‐1817  cal  BC).    Artifacts  associated  with  these  dates  include  a  variety  of  stemmed  bifaces,  microliths,  a  large  number  of  baked‐clay  objects,  and,  quartz,  quartzite,  and  sandstone  abraders.      Along  with  the  radiocarbon  dates,  these  artifacts  place  the  site’s  major  occupation  in  the  earliest  stages  of  the  Elliott’s  Point  phase  of  the  Late  Archaic  Period.     Two other radiocarbon samples are from large wood posts and suggest later occupations.    One sample returned a radiocarbon age of 1990 ± 30, or 1997‐1878 cal BP (48 BC‐AD 72).   This  would  situate  the  feature  within  the  Late  Early  Woodland  period,  which  corresponds  with  the  Deptford  culture  in  northwest  Florida.    The  second  post  returned  a  radiocarbon  age of 250 ± 30 BP.  The intercept crosses the calibration curve in several locations ranging  from modern (‐1 to 11 cal BP) to between 428 and 376 cal BP (cal AD 1522‐1574), with a  75% probability that the true age is between 428 and 270 cal BP, or cal AD 1522‐1680.  If  correct,  this  would  date  the  feature  to  the  Spanish  Contact  period.    No  artifacts  representative  of  either  period  have  been  recovered  from  Bayou  Park.    However  ceramic  sherds  diagnostic  of  the  Weeden  Island  and  Fort  Walton  archaeological  cultures  were  recovered from disturbed contexts.      A possible Middle Archaic occupation is suggested by the presence of lithic waste flakes in  deeply buried (70+ cmbs) contexts.  No temporally diagnostic artifacts were recovered from  these  depths,  although  Middle  Archaic  stemmed  bifaces  (Marion/Newnan,  Putnam,  Levy/Pickwick) were recovered from the upper levels of the site and from disturbed spoil.      Faunal  remains  are  dominated  by  marine  resources,  particularly  oysters,  scallops,  quahog  clams,  and  a  variety  of  fish  species,  particularly  jacks.    Terrestrial  resources  include  deer,  rabbit,  gray  squirrel,  raccoon,  turkey,  crow,  ducks,  alligator,  turtles,  gopher  tortoise,  salamander,  and  lizards.  Eighty‐nine  different  taxa  were  identified  in  the  298,000+  vertebrate  and  invertebrate  specimens  that  were  analyzed.    Plant  remains  were  not  numerous  but  they  provide  important  dietary  and  seasonality  information.    Edible  plant  remains  include  hickory  nut  shells,  acorn  shells,  an  unidentified  seed,  and  an  unidentified  fleshy fruit fragment.  Pine was the primary wood use for fuel along with hickory and oak.   Phytolith and starch grain analysis of a griddle fragment identified bottle gourd, hackberry,  xxvii 

possible coontie, as well as phytoliths derived from wild grasses, sedges, and from the bark  of members of the custard apple family.   The presence of bottle gourd, which is not native  to the New World, suggests the possibility of limited horticulture.  The composition of two  features,  as  well  as  the  presence  of  a  centrally  located  earth  oven,  was  interpreted  to  be  the result of small‐scale feasting for the purposes of reinforcing social identity.    A  number  of  seasonality  indicators  were  identified  in  the  faunal  (quahog  clam,  fish),  botanical  (hickory  nuts  and  acorns),  phytolith  (hackberry),  and  starch  grain  (coontie)  samples.    While  all  seasons  of  the  year  are  represented  by  these  samples,  there  is  a  congruence of all indicators during the early spring through late fall months.      The presence of non‐local stone such as Tallahatta Quartzite and cherts from the Marianna  and  Wrights  Creek  Quarry  Clusters  suggest  that  the  people  who  inhabited  Bayou  Park  traveled north perhaps during the winter months to exploit these toolstone resource areas  in  Southern  Alabama  and  near  the  Florida‐Georgia  state  line.    Local  raw  materials  include  quartz  and quartzite cobbles  for  grinding  stones,  sandstone  for  abraders,  and clay  for  the  manufacture  of  baked‐clay  objects  used  in  dry‐heat  cooking.    Powdered  ferrous  oxide  material  was  found  on  one  piece  of  sandstone  suggesting  that  the  sandstone  fragments,  which contain abundant iron, were being processed for pigments or dyes.    No  intact  burials  were  identified  but  two  human  teeth  were  recovered  from  a  single  feature.    After  consultation  between  Eglin  AFB  and  NAGPRA  tribal  representatives,  the  teeth were reburied on site as requested by tribal representatives.  Given the large area of  the  site  that  has  been  investigated  over  the  years,  the  potential  for  intact  burials  is  considered to be remote.  However, if human remains are encountered during construction  activities, then the provisions outlined in NAGPRA should be followed.    The Bayou Park site has provided a unique opportunity to examine a coastal Late Archaic,  Elliott’s  Point  Complex site  in  some  detail.    It  has  produced  important  information  on  the  age  of  the  deposits,  subsistence,  settlement  layout,  seasonality,  and  mobility  patterns  which should assist in a better understanding of Elliott’s Point in the Choctawhatchee Bay  region.    Based  on  the  results  of  this  Phase  III  data  recovery  project,  it  is  the  opinion  of  SEARCH that adverse effects to 8OK898 resulting from proposed construction activities have  been mitigated.       

xxviii 

1

INTRODUCTION 

This report presents the results of a Cultural Resources Data Recovery project for the Bayou  Park archaeological site (aka Former Eglin Prison Camp site), 8OK898, located on Eglin Air  Force Base (Eglin AFB), Okaloosa County, Florida. The project was conducted according to  the Scope of Work (SOW) included under Contract No. EAFB‐62221‐008, Task Order No. CR‐ 11‐0051.  The purpose of this contract was to conduct archaeological data recovery at the  Bayou  Park  site,  a  prehistoric  archaeological  site  that  has  been  determined  eligible  for  listing in the National Register of Historic Places (NRHP).    The site occupies about 20 acres on the former Eglin Federal Prison Camp. Several previous  archaeological investigations have been performed at the site, each in response to specific  site  development  projects  in  compliance  with  Section  106  of  the  National  Historic  Preservation  Act  (NHPA).    These  investigations  established  that  the  site’s  prehistoric  component dates primarily to the Late Archaic Period and contains archaeological remains  associated  with  the  Elliott’s  Point  Complex.    The  prison  was  officially  closed  in  2006  and  plans  for  the  multi‐purpose  use  of  the  property  involve  demolition  of  some  buildings,  rehabilitation of others, and extensive installation of underground utilities, renovation, and  land modification.  These undertakings pose a cumulative threat to the integrity of 8OK898.   These threats can no longer be met by small‐scale data recovery projects focused only on  areas  of  proposed  impact.  Therefore,  to  reduce  land  use  restrictions  and  free  the  former  prison property from further cultural resources obligations, a program of data recovery was  developed to mitigate any future adverse effects.     The  work  was  carried  out  in  fulfillment  of  responsibilities  to  consider  the  effects  of  construction  activities  on  a  National  Register‐eligible  archaeological  site.    The  Mid‐Bay  Bridge  Authority  was  the  contracting  authority  for  this  project  and  Eglin  AFB  provided  technical  oversight.    Mandating  legislation  and  directives  include  the  NHPA  of  1966,  as  amended;  the  National  Environmental  Policy  Act  of  1969  (NEPA);  36  CFR  Part  800;  the  Archaeological  Resources  Protection  Act  of  1979  (ARPA);  the  Native  American  Graves  Protection  and  Repatriation  Act  of  1990  (NAGPRA);  Air  Force  Instruction  32‐7065;  and  all  laws,  regulations  and  guidelines  promulgated  by  the  State  of  Florida  governing  cultural  resources work.      The  Principal  Investigator  for  this  project  was  Robert  J.  Austin,  PhD.    Josh  Torres,  PhD,  served  as  Project  Archaeologist.    Both  of  these  archaeologists  exceed  the  qualifications  established  in  the  Secretary  of  the  Interior's  Standards  and  Guidelines  (48  FR  44716  [29  September  1983]).    The  field  crew  included  ten  additional  SEARCH  archaeologists:  Chris  Altes,  Kendra  Crain,  Josh  Foster,  Linda  Geary,  Matt  Hanks,  Rockie  Jarvis,  Chris  Mickwee, 

Bayou Park, 8OK898 

Tom Peterson, and Dennis Wardlaw.  Jon Simon Suarez, MA, directed the field lab.  Robert J.  Austin  was  the  primary  author  of  this  report  with  contributions  by  Josh  Torres,  Brian  Worthington, MA, Dennis Wardlaw, MA, Chris Mickwee, MA, Jon Simon Suarez, MA, Anna  Dixon, PhD, Chad Yost, and Irvy Quitmyer.     

PROJECT LOCATION AND PRESENT CONDITION    The  Bayou  Park  site  (8OK898)  is  located  in  Sections  19  and  30,  T1S,  R22W  in  Okaloosa  County,  Florida  (Figure  1.1).    It  is  situated  on  a  point  of  land  where  Boggy  Bayou  and  Choctawhatchee Bay meet near the north‐central edge of the bay.  Weekley Bayou forms  the  northern  boundary  of  the  point.    Site  boundaries  were  established  through  extensive  subsurface  shovel  testing  and  limited  excavations  that  began  in  1993  and  have  been  performed  intermittently  over  the  succeeding  17  years.    Except  for  a  narrow  finger  that  extends to the northeast and separates a trailer park from Postl Lake, the site is contained  primarily within an area bounded by Flagler Road to the northwest, Inverness Road to the  southwest  and  northeast,  and  a  brackish  water  lagoon  known  as  Postl  Lake  to  the  southeast.    This  is  the  location  of  the  former  Eglin  Federal  Prison  Camp  which  contains  a  complex of buildings surrounding a central courtyard (Figure 1.2).  Three of the buildings at  the  west  end  of  the  compound  (581,  588,  and  591)  were  demolished  in  2010,  although  footers,  foundations,  and  surface  rubble  were  still  present  when  field  work  began  in  October  2011.    Vegetation  consists  primarily  of  landscaped  lawns  and  ornamental  plants,  although  a  grove  of  live  oak,  laurel  oak,  hickory,  and  a  few  sand  pine  is  present  at  the  southeastern  end  of  the  site,  between  the  complex  and  the  lake.    Numerous  utility  lines  (gas, water, electric, fiber‐optics, telephone) crisscross the compound resulting in localized  disturbances.    In  addition,  buildings  and  activities  associated  with  the  prison  camp  have  impacted portions of the site.   

SUMMARY OF PREVIOUS INVESTIGATIONS   

Surveys and Excavations, 1993‐2010  The Bayou Park site has been subjected to numerous cultural resource investigations since  it was discovered by Prentice Thomas and Associates, Inc. (PTA) during a cultural resource  assessment  survey  of  the  Eglin  Federal  Prison  Camp  in  1993  (Campbell  and  Meyer  1993).   This section summarizes the major surveys and excavations conducted between 1993 and  2010.    Several  smaller  monitoring  projects  have  been  conducted  by  PTA  and  Eglin  AFB  cultural  resource  personnel  during  this  time,  but  these  are  not  discussed  in  detail  unless  significant discoveries were made.      The 1993 survey consisted of the excavation of 69 50‐x‐50‐cm shovel tests at 20‐m intervals  across a 32‐acre project area (Figure 1.3).  Cultural materials were concentrated in an area  measuring 90 by 120 meters in the tree‐covered southeastern portion of the property.  The  recovery of a Duval  or  Flint Creek  projectile  point  indicated  occupation  during  the  Late  



Introduction 

 

Figure 1.1.  Location of 8OK898 within Eglin Air Force Base, Okaloosa County, Florida.

3

Figure 1.2.  Aerial photograph showing 8OK898 site boundaries within the former Eglin Federal Prison Camp. 

Bayou Park, 8OK898 



Introduction 

 

Figure 1.3.  Map of 1993 shovel testing at 8OK898 (from Campbell and Meyer 1993). 

5

Bayou Park, 8OK898 

Archaic or Early Woodland periods and the site was recommended as potentially eligible for  the NRHP because of the presence of intact subsurface deposits.    The boundary of the site was extended for about 180 m along the northern shore of Postl  Lake  in  1994  during  a  survey  of  several  tracts  of  land  within  Eglin  AFB  (Matthews  et  al.  1994) (Figure 1.4).  Further expansion of the boundaries and initial NRHP evaluation of the  site occurred in 1996 (Mikell et al. 1996) (Figure 1.5).  During this project, nine 50‐x‐50‐cm  shovel tests and three 1‐x‐1‐m test units were excavated in the southeastern portion of the  prison  property.    The  site  boundary  was  expanded  slightly  to  the  north  as  a  result  of  the  shovel  testing.    Apparently,  this  area  had  been  covered  with  fill  since  the  original  1993  survey  and  test  units  found  that  an  intact  deposit  buried  beneath  this  fill.    The  primary  deposit  occurred  at  a  depth  of  30‐40  centimeters  below  surface  (cmbs)  and  extended  to  about 70 cmbs.  Artifacts consisted primarily of lithic waste flakes and a few tools; however,  a burned bone fragment and a small amount of shell also were recovered.      In its 1996 report, PTA concluded that the site is unusual for the area because of evidence  of both horizontal and vertical patterning in the distribution of lithic materials.  Horizontally,  artifacts  were  reported  to  occur  in  discontiguous  concentrations  while  vertically  a  consistent  depth  of  deposition  was  observed  between  about  40  and  70  cmbs.    PTA  also  considered  the  site  to  be  a  relatively  intact  single  component  Late  Archaic  campsite.    For  these  reasons,  PTA  recommended  that  8OK898  be  considered  eligible  for  listing  on  the  NRHP (Mikell et al. 1996:46‐47).    In 1997, PTA excavated 15 50‐x‐50‐cm shovel tests and two 1‐x‐1‐m test units just south of  the Food Services Building (Building 606, formerly Building 50522) in the southern portion  of the prison camp (Campbell and Mathews 1997) (Figure 1.6).  Only 22 waste flakes and a  uniface  fragment  were  recovered  and  these  were  limited  in  their  vertical  and  horizontal  distributions.    The largest block excavation at 8OK898 prior to the current data recovery project occurred  in 1998 when PTA investigated a 30‐x‐30‐m area in the empty lot at the northwest corner of  the  prison  camp  where  a  structure  had  existed  and  two  new  structures  were  planned  (Campbell et al. 1998).  The initial field work effort consisted of the excavation of 21 50‐x‐ 50‐cm shovel tests, one of which was expanded to a 1‐x‐1‐m unit (Test Pit [TP]  6) (Figure  1.7).  These  tests  were  the  first  to  encounter  intact  midden  deposits  and  subsurface  features, and it expanded the site boundaries to the northwest.  This was followed by auger  tests,  mechanical  stripping,  and  block  excavation  of  12  1‐x‐1‐m  units  (Figure  1.8).    The  midden was found to be buried at depths varying from 27 to 53 cmbs.  It consisted of very  dark grayish‐brown sand in some areas and a shell midden in other areas.  The upper 7‐12  cm  consisted  of  recent  fill  underlain  by  Ap  and  AC  horizons  which  developed  over  the  midden  after  the  site  was  abandoned.    The  midden  rests  on  the  preoccupation  ground  surface (Ab3) and underlying C horizon.      6 

Introduction 

Figure 1.4.  Map of 1994 shovel testing at 8OK898 (from Matthews et al. 1994). 

Figure 1.5.  Map of 1996 shovel testing and test pit excavation at 8OK898 (from Mikell et al. 1996).

7

Bayou Park, 8OK898 

Figure 1.6.  Map showing the location of the 1997 shovel testing and  test pit excavation at 8OK898 (from Campbell and Matthews 1997).   Note that Building 50522 was subsequently renumbered as 606. 

Figure 1.7.  Map showing the location of the 1998 shovel testing at 8OK898 (from Campbell et al. 1998).  Note that  Buildings 50501 and 50502 were subsequently renumbered as  591 and 588, respectively.  



Introduction 

 

Figure1.8.  Map showing the location of the 1998 block excavation at 8OK898  (from Campbell et al. 1998). 

9

Bayou Park, 8OK898 

The  features  consisted  of  a  linear  shell  concentration  (Feature  1),  a  possible  cooking  pit  (Feature  2),  a  possible  refuse  pit  (Feature  3),  and  a  postmold.    Shell  from  Feature  1  was  radiocarbon dated to 3560 ± 70 measured  14C years BP (2205‐1860 cal BC) while shell from  Feature  2  was  dated  to  3790  ±  70  measured  14C  years  BP  (2535‐2140  cal  BC)  (Morehead,  Campbell, and Aubuchon 2008:59).    The artifact assemblage included lithic debitage and a small number of tool forms (n=513),  pieces  of  ground  stone  (n=10),  unmodified  sandstone  and  limonite  (n=23),  worked  shell  (n=41), fired clay (n=943), Elliott’s Point Objects (EPOs; n=63), and ceramic pottery sherds  (n=17).    The  EPOs  are  described  as  primarily  amorphous  in  shape,  but  some  tabular  and  melon‐shaped  forms  were  identified.    Temporally  diagnostic  artifacts  include  one  Westo  projectile point, a Carrabelle Punctated sherd, a sherd of Cool Branch Incised, two brushed  sherds,  12  plain  sherds,  and  the  Elliott’s  Point  Objects  (EPOs).    In  addition  to  the  Westo  point,  two  microliths  (identified  as  Jaketown  perforators), a  core,  seven  biface  fragments,  and  a  uniface  were  recovered.    Most  of  the  lithic  material  was  identified  as  Tallahatta  Quartzite  with  a  small,  but  varied  amount  of  cherts  also  present.    The  Westo  point,  Jaketown perforators, and EPOs were considered consistent with a Gulf Formational (Late  Archaic)‐period,  Elliott’s  Point  Complex  site  and  the  radiocarbon  dates  indicated  that  this  part  of  8OK898  was  occupied  primarily  during  the  early  phase  of  Thomas  and  Campbell’s  (1993)  Elliott’s  Point  chronology  (see  Chapter  2).    The  pottery  sherds  also  indicate  occupation,  albeit  non‐intensive,  during  the  Weeden  Island  and  Fort  Walton/Pensacola  periods.    Faunal  analysis  was  limited  to  a  sample  obtained  from  TP12  and  three  features.    The  dominant shellfish was oyster, while quahog clam, bay scallop, and lightening whelk were  common and rangia clams were very scarce.  Most of the shell from Feature 2 (99%) was  burned,  which  led  to  the  conclusion  that  this  may  have  been  a  cooking  pit.    Vertebrate  fauna  included  turtle,  white‐tail  deer,  fish,  opossum,  and  bird.      Analysis  of  charred  botanical remains identified pine as the dominant wood species followed by oak.  Cherry,  elm,  white  cedar,  and  grape  vine  were  present  in  small  amounts.    Hickory  nut  shell  was  found in two features (Feature 1 and Feature 3) and the general midden leading the authors  to conclude that the site was occupied during the fall months.    The  next  significant  cultural  resource  project  occurred  in  2002  when  PTA  conducted  a  survey of a proposed sprinkler and communication cable trench (Meyer et al. 2002) (Figure  1.9).  Forty‐five 50‐x‐50‐cm shovel tests were excavated along the 620‐m‐long trench line at  20‐m intervals.  The trench line began in the northwest corner of the prison camp south of  Building  50501  (currently  591)  and  near  the  block  excavation  conducted  in  1998.    It  then  proceeded north to the compound fence, then southeast along the fence line to a point just  past Building 50523 (currently 635) where it turned southwest and then west, terminating  at  Building  50525  (currently  636).    The  shovel  testing  identified  two  areas  of  possible  occupation: in the northwest part of the compound near Buildings 50501 (591) and 50502  (588), and in the southeast near Building 50525 (636).       10

Introduction 

Figure 1.9.  Map showing the location of the 2002 shovel testing and test pit excavation at 8OK898 (from Meyer et al. 2002). 

In the northwestern area, Shovel Test (ST) 163 encountered shell midden between 35 and  60 cmbs resting on a probable pre‐occupation Ab horizon.  Shovel test 162 encountered a  refuse pit between 48 and 150 cmbs.  A 1‐x‐1‐m test pit (TP19) excavated near these shovel  tests  encountered  the  midden  and  identified  two  additional  features  at  about  150  cmbs.   The strata in TP19 were similar to those encountered during the 1998 block excavation to  the northwest except that the fill was much thicker.  The upper fill zone measured 50‐cm  thick and overlay A and AC horizons, which formed over the top of the buried midden. The  midden  consisted  of  very  dark  gray  sand  from  108‐150+  cmbs.    The  excavation  was  terminated  before  the  base  of  the  midden  was  reached  since  the  proposed  trench  would  not extend that deep. 

11

Bayou Park, 8OK898 

No midden strata were identified in the southeastern portion of the compound; however, a  fourth feature was identified in ST174.  It was subsequently investigated with two 2‐x‐2‐m  excavation units (collectively identified as TP20) and found to be a  bell‐shaped refuse pit.   The top of the feature was encountered at about 70 cmbs and extended to a depth of 145  cmbs.  It contained primarily crushed oyster, quahog clam, and smaller amounts of Rangia,  coquina, and scallop shells, bone fragments, EPOs, and lithics.  The lithic artifacts included a  Ledbetter‐like projectile point.  Three other projectile points were recovered from outside  the features: Six Mile Creek, Destin‐like, and an unidentified “dart point.”    Two  shell  samples  from  the  bell‐shaped  feature  returned  radiocarbon  dates  of  3800  ±  50  measured 14C years BP (2490‐2180 cal BC) and 3770 ± 50 measured 14C years BP (2460‐2130  cal BC) (Beta Analytic, Inc. 2002).  These dates are very similar to the dates obtained from  Features  1  and  2  excavated  in  the  northwest  portion  of  the  site  (described  above),  suggesting that the two areas were relatively contemporaneous.    In addition to the projectile points, other artifacts from this project included lithic debitage  (n=247),  bifaces  (n=7),  microliths  (n=3),  utilized  or  retouched  flakes  (n=3),  pottery  (n=1),  EPOs  (n=23),  and  fired  clay  (n=214).    A  small  sample  of  botanical  remains  was  collected  from  Feature  4,  but  the  only  species  identified  was  a  single  hickory  nut  shell  fragment.   Vertebrate faunal remains included white‐tail deer, fox squirrel, dog or coyote, turkey, box  turtle, mullet, red and black drum, and crevalle jack.  Invertebrates included oyster, quahog  clam, rangia, scallop, banded tulip, moon snail, crown conch, coquina, cross‐barred venus,  cardita, ladder shell, common slipper, and nerite.    During this project, PTA established a permanent grid system with the N1000, E1000 point  located  about  six  meters  to  the  northeast  of  a  concrete  sewer  marker  in  the  northwest  corner  of  the  prison  camp.    The  1998  block  excavation  unit  coordinates  were  revised  to  reflect the new grid and all subsequent work was also tied into this system.   To facilitate  this, secondary datum points were established in various locations throughout the site.    In 2003, Eglin cultural resource personnel monitored excavation of the trench and identified  a  linear  sandstone  feature  just  to  the  northeast  of  Building  635  (formerly  50523).    The  function of this feature is not known, nor has it been determined to be definitely cultural  (Eglin Air Force Base 2003).    In  2007,  work  was  undertaken  to  investigate  areas  that  would  be  impacted  by  the  installation  of  fiber‐optics  cable  within  the  prison  compound  (Morehead,  Campbell,  and  Aubuchon  2008)  (Figure  1.10).    Eight  50‐x‐50‐cm  shovel  tests  were  excavated  in  various  locations  within  the  compound  and  16  similar  shovel  tests  were  excavated  in  a  proposed  parking lot to the northeast of the compound.  In addition to the shovel tests, one 2‐x‐2‐m  unit (TP21) and seven 1‐x‐1‐m units (TPs 22‐28) were excavated in proposed impact areas.   Three of the shovel tests (176, 177, and 183) and four of the test units (21, 23, 24, and 26)  contained cultural materials.      12

Introduction 

Figure 1.10.  Map showing the location of the 2007 shovel testing and test pit excavation at  8OK898 (from Morehead, Campbell, and Aubuchon 2008). 

  While  the  shovel  testing  expanded  the  boundaries  of  the  site  slightly  to  the  north  and  northeast,  no  new  midden  areas  were  identified.    The  2‐x‐2‐m  test  unit  (TP21)  was  excavated near the northern corner of Building 604.  The upper 30 cm were disturbed, but  lithic waste flakes were recovered from undisturbed sediments between 30 and 130 cmbs.   Test  Pit  23,  located  near  the  southwest  corner  of  Building  604  displayed  disturbed  sediments  to  a  depth  of  60  cmbs  with  only  a  single  lithic  waste  flake  recovered  from  between 90 and 100 cmbs.  Test Pit 24, located near the southeast corner of Building 626,  was disturbed to 40 cmbs, but lithic waste flakes were recovered between 60 and 80 cmbs.   In TP26, on the south side of Building 633, lithic and shell fragments were recovered from  between 40 and 50 cmbs beneath an upper disturbed layer; however, no additional artifacts  or shell were recovered below 50 cmbs and excavation ceased at 100 cmbs.  Artifacts were  not abundant; the total number from all test pits consisted of 22 waste flakes, 19 of which  were identified as Tallahatta Quartzite and the remainder as Coastal Plain chert.   

13

Bayou Park, 8OK898 

PTA  provided  a  map  in  this  report  showing  the  depth  of  fill  across  the  prison  compound  based on the results of shovel tests and test units.  This map shows that the depth of fill and  disturbed  sediments  varies  from  25  to  75  cmbs  across  the  compound,  with  the  deepest  occurrences near existing structures.      Subsequent to this investigation, the location of the fiber‐optic line near Buildings 604 and  633 were moved by the contractor and Eglin AFB cultural resource personnel excavated two  test units within the new area of potential effect (APE) (Morehead  et al. 2008: Appendix A).   Test  Pit  29  (1‐x‐.8  m)  was  excavated  next  to  PTA’s  TP26  south  of  Building  633  and  one  chunky fragment of Tallahatta Quartzite and an incised EPO fragment were recovered from  between  50  and  60  cmbs.    Test  Pit  30  (1‐x‐1‐m)  was  excavated  near  PTA’s  ST175  on  the  west side of Building 604.  Three lithic chunks of Tallahatta Quartzite and Coastal Plain chert  and one EPO were recovered from between 60 and 90 cmbs.    In 2008, excavations were conducted to mitigate impact from the installation of a concrete  refrigeration  pad  and  an  electrical  utility  line  associated  with  Building  635  in  the  northeastern portion of the Prison compound (Morehead and Aubuchon 2008).  Five 1‐x‐1‐ m units were judgmentally located where the concrete pad was proposed (Figure 1.11) and  two 1‐x‐1‐m units and two 50‐x‐50‐cm shovel test (STs 199, 200) were excavated where the  utility trench was planned (Figure 1.12).    Both excavation areas exhibited rather extensive disturbance.  In the concrete pad area, fill  and  disturbed  sediments  were  encountered  to  depths  ranging  from  35  to  60  cmbs  and  utility  trenches  were  found  to  have  caused  localized  disturbance  to  depths  of  up  to  one  meter.  Within the proposed utility line, fill was encountered to depths ranging from 50 to  70  cmbs  with  utility  trench  disturbance  to  at  least  one  meter.      Cultural  materials  were  recovered  from  three  test  units  in  the  concrete  pad  area  and  were  extremely  limited  in  number consisting of seven Tallahatta Quartzite waste flakes and two pieces of shell.    In  2009,  PTA  conducted  monitoring  of  the  construction  of  a  parking  lot,  two  driveways,  sidewalks,  and  a  retention  pond  to  the  northeast  of  the  Prison  compound,  as  well  as  improvements in the air conditioner area of Building 635 (Bourgeois et al. 2009).  Portions  of  the  parking  lot  and  retention  pond  areas  were  carefully  graded.    Although  historic  artifacts  were  occasionally  found  and  a  historic  refuse  deposit  was  identified  and  tested  (TP38)  in  the  retention  pond  area,  no  prehistoric  artifacts  or  features  were  documented.   Nor were any artifacts or features identified in the air conditioner area near Building 635.    Also in 2009, PTA conducted mitigative excavations associated with renovations to Buildings  632,  633,  and  634  (Campbell,  Morehead,  et  al.  2009).    Utility  trenches  and  water  line  trenches  were  planned  outside  the  buildings  while  impacts  to  the  interiors  involved  the  removal of the wooden flooring and bases.  Fourteen 50‐x‐50‐m shovel tests, one 1‐x‐1‐m  test pit (TP49), and four 2‐x‐2‐m test pits (TPs 39‐42) were excavated in various proposed  impact areas outside the buildings and 20 50‐x‐50‐cm shovel tests, three 1‐x‐2‐m test units,  one 1.4‐x‐2‐m test unit, and two 2‐x‐2‐m test units were excavated within the interiors of 14

Figure 1.11.  Map showing the location of the 2008  test pit excavation northeast of Building 635 at  8OK898 (from Morehead and Aubuchon 2008). 

Figure 1.12.  Map showing the location of the 2008 test pit  excavation southwest of Building 635 at 8OK898  (from Morehead and Aubuchon 2008). 

Introduction 

 

15

Bayou Park, 8OK898 

the  three  buildings  after  the  flooring  had  been  removed  (Figure  1.13).    Unlike  previous  excavations,  the  2009  project  utilized  a  datum  plane  for  determining  depths  during  unit  excavation.  The datum plane was based on a true elevation of 3.88 m above mean sea level  (amsl)  established  at  a  manhole  cover  near  the  northwest  corner  of  Building  633.   Elevations were transferred to the northwest corners of test units and level measurements  were  recorded  in  cm  below  unit  datum.    Fill  layers  were  removed  by  hand  without  screening and controlled excavation using 10‐cm levels began at the top of the exposed A or  E horizons.    Many  of  the  exterior  shovel  tests  and  test  pits  encountered  heavily  disturbed  sediments;  however,  a  few  encountered  prehistoric  artifacts  in  what  appeared  to  be  undisturbed  contexts  beneath  the  disturbed  upper  strata.    Test  Pit  49  (1‐x‐1‐m),  located  northeast  of  Building 633, penetrated disturbed sediments to 117 centimeters below site datum (cmbd),  but  recovered  lithics,  fired  clay  fragments  and  shell  in  presumably  undisturbed  contexts  between  118  and  187  cmbd.    Shovel  Test  209  was  excavated  on  the  east  side  of  Building  634.    Although  the  upper  66  cm  were  disturbed,  marine  shell  was  recovered  from  what  appeared to be undisturbed sediments between 104 and 114 cmbd.  The authors remark on  the  variety  of  shellfish  remains  in  this  small  sample,  which  included  oyster,  quahog  clam,  Rangia, scallop, and ark.    Test Pit 39  was located on the northwest side of Building 632 and contained shell, lithics,  and fired clay between 60 and 163 cmbd.  Test Pit 40, placed on the northwest side of this  building,  also  recovered  shell,  lithics,  and  a  large  quantity  of  fired  clay  below  60  cm  of  disturbed  sediments.    Prehistoric  artifacts  in  this  unit  continued  to  be  recovered  to  142  cmbd.  One flake was found between 119 and 129 cmbd in TP41, located on the northeast  side of Building 633.  Test Pit 42, placed on the southeast side of Building 634, was the least  disturbed of all the exterior tests.  Fill material was encountered only in the upper 30‐35 cm  which  overlay  undisturbed  sediments  containing  lithic  waste  flakes,  fired  clay,  and  shell.   Excavation was terminated at 199 cmbd.    Excavations within the structures encountered deep fill deposits of up to 65 cm in thickness  and  localized  disturbances  below  this  depth.    In  Building  634,  for  example,  asphalt  slabs  were encountered at 95 cmbd and several historic features and postmolds were identified.   Despite  the  fill  and  disturbance,  three  prehistoric  features  were  identified  in  the  deeper,  undisturbed  strata.    Features  14  and  17  were  identified  in  TP47  within  Building  632.    The  two features blend into one another and are probably functionally related.  Feature 14 was  encountered  at  146  cmbd  and  Feature  17  at  173  cmbd.    Feature  14  is  interpreted  as  a  probable  hearth  and  small  piles  of  charcoal  and  fired  clay  were  found  nearby  suggesting  that these may have come from cleaning out the hearth.     The  third  prehistoric  feature  was  identified  in  TP44  within  Building  634.    Feature  16  was  encountered at a depth of 168 cmbd and extended for about 20 cm.  It was oblong in shape  and contained charcoal flecks, two waste flakes, and a piece of fired clay.  It was interpreted  as a possible hearth.  16

Introduction 

 

Figure 1.13.  Map showing the location of the 2009 shovel testing and test pit excavation at 8OK898 (from Campbell, Morehead, et al. 2009). 

17

Bayou Park, 8OK898 

According  to  Campbell,  Morehead,  et  al.  (2009:84),  the  soils  encountered  during  this  project conform to the Resota series, with an A‐E‐(Bh)‐Bw‐C sequence of horizons beneath  the modern fill episodes.  The sequence typically displayed a dark Ap horizon when present;  a  white  E  horizon;  a  yellowish‐brown  to  brownish‐yellow  Bw  horizon;  and  a  brownish‐ yellow  C  horizon.    No  midden‐stained  sediments  were  encountered  in  any  of  the  shovel  tests or test pits excavated during this project.       Most  of  the  recovered  artifacts  were  historic;  prehistoric  artifacts  consisted  of  96  waste  flakes, 3 EPOs, and 2291 g of fired clay.  One of the flakes is a small bladelet that resembles  a possible blank for a microlithic tool.  It was recovered from between 137 and 147 cmbd in  TP49.    Typical  for  this  site,  most  of  the  debitage  is  Tallahatta  Quartzite  with  smaller  amounts of Coastal Plain chert, miscellaneous chert, metaquartzite, and quartz.  The EPOs  were found in two units, TPs 43 and 44, all below 154 cmbd.      Invertebrate remains included primarily oyster and quahog clam as well as smaller amounts  of scallop, ark, rangia, cross‐barred venus, and horn shell.  Most of the shell is considered to  be from prehistoric contexts.  The vertebrate material, which included some domesticated  species, appears to be primarily from historic contexts, although the authors indicate that at  least some of the non‐domesticated species may be prehistoric.      In 2010 data recovery efforts to mitigate impact from the restoration of Building 636 were  conducted by PTA (Morehead et al. 2011).  The renovation plans involved trenching on the  exterior of the building for electrical and water lines, and the installation of two concrete  pads.  Twenty‐two 50‐x‐50‐cm shovel tests, two 1‐x‐.5‐m test pits, two 1‐x‐1‐m test pits, and  four  1‐x‐2‐m  test  pits  were  excavated  during  the  project  (Figure  1.14).    Most  of  the  units  encountered  fill  and  disturbed  sediments  at  thicknesses  varying  from  20  to  60  cm.   Localized disturbances from utility trenches reached as deep as 90 cmbs.  Artifacts included  mostly  lithic  waste  flakes  and  fragments  of  fired  clay,  although  a  biface  fragment,  microlithic perforator, and an amorphous EPO are reported.  Lithics were recovered as deep  as 110 cm and the EPO came from 70‐80 cm.  Test Pits 54, 56, and 57 contained a dense  concentration of lithics between 10 and 100 cm.  Faunal material included a small amount  of  oyster  shell.    One  hickory  nut  shell  was  recovered  from  80‐90  cmbs  in  TP52.    PTA  also  monitored the removal of interior floor slabs.  Since the maximum depth of impact was only  25  cm,  the  underlying  soil  was  turned  over  with  a  shovel  and  probed  to  determine  if  any  midden  deposits  are  present  beneath  the  building.    None  were  encountered  and  fill  was  identified to a depth of 40 cmbs.     The  most  recent  investigation  at  8OK898  occurred  in  November  2010  when  Eglin  cultural  resource  personnel  monitored  the  demolition  of  Buildings  581,  588,  and  591  (Speal  and  Nolan  2011)  (Figure  1.15).    A  sewer  trench  excavated  to  the  north  of  Building  591  encountered dark midden soils beneath recent fill at a depth of about 75 cm.  The midden  contained a variety of shell species (scallop, oyster, quahog clam, lightning whelk, fighting  conch, moon snail, ark, and rangia) as well as fish and mammal bones.  One quartz flake was  also collected.  Two areas of midden shell were identified within Building 591.  One of these 18

Introduction 

 

Figure 1.14.  Map showing the location of the 2010 shovel testing and test pit  excavation at 8OK898 (from Morehead et al. 2011). 

19

Bayou Park, 8OK898 

Figure 1.15.  Map showing the location of the 2010 monitoring and the area at Building 591 where  midden deposits were identified (from Speal and Nolan 2011). 

(designated Area A) was immediately south of the sewer trench while the second (Area B)  was located near the west end of the building.  The shell was contained within clay fill that  underlay the building and is assumed to have become mixed when the ground surface was  graded and prepared prior to building construction.  These findings extended the northern  and eastern edges of the buried midden in this part of the site.   

Summary  Between 1993 and 2010, 257 50‐x‐50‐cm shovel tests and 57 test units varying in size from  1‐x‐.5‐m  to  2‐x‐2‐m  in  size  were  excavated  at 8OK898,  and  several  small‐scale monitoring  projects  were  carried  out.    These  projects  produced  a  large  body  of  data  on  site  stratification  and  vertical  and  horizontal  distributions  of  artifacts,  faunal  remains,  and  features, as well as chronological data that places the site within the earliest phase of the  Late Archaic Elliott’s Point Complex.  This work laid a strong foundation for the 2011‐2012  data  recovery  project  to  build  on  and  strongly  influenced  the  approach  to  excavating  and  analyzing the site.  Figures 1.16 and 1.17 show the locations of all PTA shovel tests and test  pits excavated between 1993 and 2010.     20

Figure 1.16.  Map of shovel tests excavated by PTA, 1993‐2010. 

Introduction 

 

21

Figure 1.17.  Map of test pits excavated by PTA, 1993‐2010. 

Bayou Park, 8OK898 

22

Introduction 

Four radiocarbon dates on shell indicate that the site was occupied between 4210 and 3970  years BP (recalibrated here using CALIB 6.01 as 2529‐1817 BC) (Table 1.1).  This places the  site well within the Late Archaic period and the “nascent” Elliott’s Point complex as defined  by Thomas and Campbell (1993).   Moreover, there is very close correspondence between  the four dates, particularly those from Feature 4, the bell‐shaped pit identified in southeast  portion of the site, and Feature 2, identified in the midden area in the northwest corner of  the site.  Pairwise comparison using a student’s t test determined that these three dates are  not significantly different from one another and yield a pooled average age of 4196 ± 36 BP.   On the other hand, all three ages are significantly different from the conventional  14C age  obtained from Feature 1.  These dates suggest that 8OK898 may have been occupied over a  period of about 200 years or so.    Sample No. 

Table 1.1.  Summary of radiocarbon dates from 8OK898.  Measured  13C/12C  Conventional  Calibrated Date  Provenience  14 14 C Age  Ratio  C Age  Range (2 sigma) 

Beta‐171507 

Feature 4 

3800±50 

0.0 

4210±60 

2529‐2178 B.C. 

Beta‐129298 

Feature 2 

3790±70 

0.0 

4200±70 

2532‐2135 B.C. 

Beta‐171508 

Feature 4 

3770±50 

0.0 

4180±60 

2468‐2132 B.C. 

Beta‐129297  Feature 1  3560±70  0.0  3970±70  2228‐1817 B.C.  NOTE: Calibrations performed using CALIB 6.01 and MARINE09 calibration dataset (Stuiver  and Reimer 1993). 

  The main areas of artifact concentration appear to be the midden area in the northwestern  corner  of  the  site  and  an  area  of  lithic  artifacts,  baked‐clay  objects,  and  features  to  the  southeast near Postl Lake.  The midden area appears to be unique in having dark, organic‐ stained  sediments,  abundant  faunal  remains,  dense  concentrations  of  baked‐clay  objects  and  fired  clay  fragments,  and  several  features,  including  a  postmold  from  a  possible  structure.        Shovel  testing  and  the  excavation  of  a  1‐x‐1‐m  unit  (TP19)  to  the  east  of  the  1998  block  excavation  indicate  that  the  midden  extends  to  the  north  in  front  of,  and  possibly underneath, Building 591.  Additional shovel tests to the southeast of the sidewalk  bordering  the  block  excavation  area  also  encountered  midden  material  suggesting  an  extension in this direction as well.      A second area of interest is the far eastern portion of the site near Postl Lake.  PTA’s TP20,  which  actually  includes  two  overlapping  2‐x‐2‐m  units,  was  situated  here  and  exposed  a  large  refuse  pit  (Feature  4).    In  addition,  TP20  contained  a  large  assemblage  of  worked  lithics, including four projectile points (one from Feature 4), three bifaces, two microliths,  and  two  utilized  flakes.    Burned  bone  was  recovered  between  20  and  60  cmbs  in  TP3  located to the north of TP20.  This area was the focus of a recent data recovery project by  PTA (Morehead et al. 2011) and it identified a dense concentration of lithics in TPs 54, 56,  and 57.    A  third  area  of  interest  is  located  in  and  around  Buildings  626,  632,  633,  634.    Although  artifact  density  within  this  area  is  very  low,  marine  shell  was  consistently  recovered  from 

23

Bayou Park, 8OK898 

test units placed underneath and near these buildings.  Several pit features and a postmold  were identified in TPs 44, 45, and 47.  In addition, a linear sandstone feature of unknown  age or function was identified during monitoring to the north of Building 635.     Although testing in the rest of the site has been limited, there appears to be only very low  densities of artifacts and minimal amounts of faunal material located in these areas and no  features were identified.     

SUMMARY OF PHASE III RESULTS   

Excavation Strategy  The Phase III data recovery project began on October 17, 2011 with a Ground‐Penetrating  Radar  (GPR)  survey.    Excavation  began  on  November  28,  2011  and  was  completed  on  February 21, 2012.   In order to organize the work effort and to coordinate the location of  buried  utilities,  the  site  was  subdivided  into  three  arbitrary  areas  or  Operations  (Figure  1.18).  Operation A included the midden area identified by PTA and extended about halfway  into  the  central  courtyard.    Operation  B  included  the  area  east  of  Buildings  635  and  636.   Operation C included the courtyard area between Operations A and B.      The  first  task  was  to  conduct  a  GPR  survey  of  the  entire  site  in  an  effort  to  identify  subsurface  midden  deposits,  possible  features,  modern  utility  lines,  and  other  subsurface  anomalies.    Once  the  GPR  survey  was  completed,  1‐x‐2‐m  excavation  units  were  begun  within  the  footprints  of  two  demolished  buildings,  588  and  591,  followed  by  mechanical  stripping within and surrounding these buildings.  Work began here because of the need to  remove footers and debris from the buildings.  Once these buildings were cleared for debris  removal, mechanical stripping proceeded to the east across the courtyard.  Once stripping  in Operation A was completed, work moved to Operation B, then to Operation C.  Finally,  some  additional  stripping  was  conducted  in  Operation  A  to  the  east  of  Building  591  after  the building debris had been removed in order to further explore the extent of the midden.   At the same time that the stripping took place, excavation of 1‐x‐2‐m and 2‐x‐2‐m test pits  was conducted in selected areas.  The test pits were excavated in areas where stripping was  not  feasible  or  to  investigate  areas  of  artifact  or  faunal  concentrations  identified  during  previous work by PTA.      Mechanical stripping was performed in 30 separate areas within 8OK898 ‐‐ 19 in Operation  A, 8 in Operation B, and 3 in Operation C – for a total volume of approximately 4670 cubic  meters.  A total of 30 test excavation units were excavated: 11 1‐x‐1‐m, 10 1‐x‐2‐m, 8 2‐x‐2‐ m,  and  1  .5‐x‐1‐m.    The  total  volume  of  excavated  sediments  from  these  units  was  74.86  cubic meters.  Most of the test units were excavated in Operation A (n=21), including 10 1‐ x‐1‐m  units  in  a  small  block  that  was  excavated  to  investigate  several  large,  overlapping  features.  Four units were excavated in Operation B and five in Operation C.  Three column  samples  (.5‐x‐.5  m)  also  were  excavated  and  soil  samples  from  features  and  strata  were  obtained  for  laboratory  analysis.    Finally,  131  features  were  identified  and  mapped,

24

Figure 1.18. Map showing Operations A, B, and C within site 8OK898 as well as the locations of Strip Areas and Test Units.

Introduction 

25

Bayou Park, 8OK898 

including  112  prehistoric  features,  and  a  representative  sample  was  excavated.    The  excavations  resulted  in  the  recovery  of  nearly  13,000  artifacts  and  over  500,000  faunal  remains.   

Excavation Results  The  details  of  the  excavation  and  various  analyses  are  the  subjects  of  the  following  chapters.    In  brief,  the  results  indicate  that  the  Bayou  Park  site  was  occupied  primarily  during the Late Archaic period 4481‐3766 cal BP (2529‐1817 cal BC), with minor occupations  occurring  during  the  Middle  Archaic,  Middle  to  Late  Woodland,  and  Mississippi  periods.   The site consists of a semi‐circular midden deposit situated on a slight topographic rise in  the  southwest  portion  of  the  site  and  a  moderately  dense  lithic  scatter  in  the  far  eastern  portion of the site near Postl Lake.  In between these two areas is a relatively low‐density  scatter of lithic artifacts, primarily waste flakes.      The  midden  is  estimated  to  cover  an  area  of  approximately  1500  square  meters.    It  is  oriented  with  its  open  side  facing  northwest,  or  upslope.    The  position  of  the  midden  at  some distance from Postl Lake and Choctawhatchee Bay is believed to be due to its location  on high ground overlooking a wetland from which a small stream originates.  The wetland  may  at  one  time  have  contained  a  spring.    The  midden  consists  of  dark,  organic‐stained  sediments  containing  animal  bone,  marine  shell,  and  artifacts.    Shell  concentrations  are  evident  within  the  darker  midden  matrix.    These  represent  individual  episodes  of  shell  disposal  which  eventually  began  to  overlap  and  coalesce  in  some  parts  of  the  site.      The  midden’s semi‐circular shape does not appear to be the result of intentional construction,  but rather the structured organization of living space around a central “plaza” or common  area where several very large pit features and a large clay‐lined hearth were identified.  The  hearth  is  located  at  the  bottom  of  a  large,  deep  pit  and  may  have  been  used  as  an  earth  oven.  Other large pits are located around the inner perimeter of the midden.  Smaller pit  features  are  more  randomly  distributed  and  are  located  within  or  immediately  below  the  midden deposits.    Several post molds and postholes were identified and mapped leading to the recognition of  two possible structures.  One possibly oval‐shaped structure measures no more than 2 or  2.5 m in diameter, although it was not entirely exposed.  The second structure is larger, with  a large open front, and is associated with a hearth.  It measures about 7 m across and about  2 m deep.  The small sizes of the structures and their relatively insubstantial construction  suggests they were inhabited by small family groups.    The site contains several very large pits and numerous smaller ones.  Some pits were clearly  used for refuse disposal although some of the larger ones may have originally been meant  to be used for storage.  Only one of these is clay‐lined, however, which would have helped  to insulate the contents from moisture.  Other pits contain no or very little refuse material  but are distinguished by their dark organic‐stained sandy fill.  Several of these were tested  for coprastanol to determine if they were privies, but the results proved negative.  The lipid 

26

Introduction 

analysis  did  reveal  the  presence  of  cholestanol  and  epicoprostanol  in  several  features,  which  results  from  the  degradation  of  terrestrial  and  marine  animal  remains,  and  plant  sterols diagnostic of hickory nuts.     The lithic concentration near Postl Lake consists of at least two high density areas separated  by a moderate to low‐density scatter of artifacts.  One high‐density area was excavated by  PTA and includes a large pit containing marine shell and lithic tools.  Another high‐density  area was identified by SEARCH and contains primarily lithic waste flakes.  In addition to the  PTA  feature,  five  other  features  were  identified.    These  are  mostly  refuse  pits  of  various  sizes, although the function of two features could not be determined. PTA identified three  small hearths or possible hearths in the northeast part of the site and these were associated  with lithic waste flakes, small amounts of shell, and a few baked‐clay objects.    A total of 16 radiocarbon dates (four obtained by PTA and 12 by SEARCH) identified three  temporal occupations while two other possible occupations were identified on the basis of  stratigraphy  and  temporally  diagnostic  artifacts.    The  primary  occupation  was  during  the  Late Archaic period with 14 dates having a maximum two‐sigma range of 4481‐3766 cal BP  (2529‐1817 cal BC).  Using only the median dates yields a minimum range of 4301‐3971 cal  BP  (2352‐2002  cal  BC),  or  a  little  over  three  centuries.    Most  of  the  Late  Archaic  dates  appear  to  be  in  good  stratigraphic  order.    The  pooled  radiocarbon  age  for  the  midden  is  4078‐3912 cal BP, or 2129‐1963 cal BC.  Dated features cluster into three groups.  Features  60,  69/104,  and  139  range  in  age  between  4150  and  3926  cal  BP  (2201‐1977  cal  BC);  Features 44 and 120 range between 4286 and 4000 cal BP (2337‐2051 cal BC); and two shell  samples from Feature 4 returned nearly identical ages with a pooled average of 4195 ± 42  (4405‐4149  cal  BP,  or  2456‐2200  cal  BC).    This  feature  is  located  near  Postl  Lake  and  indicates  that  this  area  was  being  used  near  the  beginning  of  the  site’s  most  intensive  occupation during the Late Archaic.    Artifacts  associated  with  these  dates  include  a  variety  of  stemmed  bifaces,  microliths,  a  large number of baked‐clay objects, and, quartz, quartzite, and sandstone abraders.   Along  with the radiocarbon dates, these artifacts place the site’s major occupation in the earliest  stages of the Elliott’s Point phase of the Late Archaic, as defined by Thomas and Campbell  (1991).      Two other radiocarbon samples have provided problematic dates.  Both samples are from  large wood posts.  A charcoal sample from Feature 38 returned a radiocarbon age of 250 ±  30 BP.  The intercept crosses the calibration curve in several locations ranging from modern  (‐1 to 11 cal BP) to between 428 and 376 cal BP (cal AD 1522‐1574), with a 75% probability  that the true age is between 428 and 270 cal BP, or cal AD 1522‐1680.  If correct, this would  date  the  feature  to  the  Spanish  Contact  period.    The  second  post,  Feature  51,  returned  a  radiocarbon age of 1990 ± 30, or 1997‐1878 cal BP (48 BC‐AD 72).   This would situate the  feature within the Late Early Woodland period.  No artifacts representative of either period  have been recovered from Bayou Park with the possible exception of a Fort Walton Incised  sherd  recovered  from  a  disturbed  context,  and  a  possible  sherd  of  Lake  Jackson  Plain  27

Bayou Park, 8OK898 

reported  by  PTA.    Both  of  these  could  be  associated  with  the  Spanish  Contact  date,  although they fit more comfortably in a pre‐contact, Mississippi‐period context.  Woodland‐ period  artifacts  include  sherds  of  Carrabelle  Punctated,  Carrabelle  Punctated  or  Ruskin  Linear  Punctated,  and  Swift  Creek  Complicated  Stamped  pottery.    These  are  diagnostic  of  the Middle to Late Woodland period and are believed to have been introduced much later  than the date from Feature 51.  Other ceramic sherds (sand‐tempered plain, indeterminate  check‐stamped,  indeterminate  incised)  are  neither  diagnostic  nor  especially  prevalent.   Whatever  the  age  of  the  ceramic  component  at  Bayou  Park,  it  does  not  appear  to  have  been very intensive.  Except for the two problematic posts, none of the features or midden  dates suggests a post‐Archaic occupation.    A possible Middle Archaic occupation is suggested by the presence of lithic waste flakes in  deeply buried (70+ cmbs) contexts.  No temporally diagnostic artifacts were recovered from  these  depths,  although  Middle  Archaic  stemmed  bifaces  (Marion/Newnan,  Putnam,  Levy/Pickwick) were recovered from the upper levels of the site and from disturbed spoil.   The  small  numbers  of  artifacts  associated  with  this  component,  and  their  scattered  distribution, suggests that it consisted of small, short‐term encampments.    Indeed, except  for Feature 4, which contained abundant faunal remains, most of the non‐midden portions  of  Bayou  Park  were  utilized  in  a  similar  fashion;  that  is,  as  short‐term  encampments  or  specialized activity areas related to the use and maintenance of stone tools. This includes   all  of  the  non‐midden  areas  that  are  believed  to  be  contemporaneous  with  the  midden  based on their stratigraphic position.    One of the defining traits of the Elliott’s Point Complex is the presence of baked‐clay objects  (also  referred  to  as  Elliott’s  Point  Objects  or  EPOs).    Over  11,000  baked‐clay  objects  were  recovered  from  Bayou  Park,  mostly  small  fragments  but  over  700  were  large  enough  for  detailed  analysis.    Amorphous  forms  are  the  most  common  although  oblate,  prolate,  tabular,  and  cylindrical  forms  also  are  present.    Only  a  few  objects  display  surface  modifications in the form of linear grooves; more common are hand, finger, and fingernail  impressions.  Excavation of the large clay‐lined hearth recovered a large number of baked‐ clay objects in and around the feature, verifying their use in cooking.   Three tabular objects  are  believed  to  be  parts  of  griddles  used  in  cooking.    One  of  these  had  starch  grains  and  phytoliths  of  bottle  gourd,  hackberry,  possible  coontie,  and  wild  grasses  embedded  in  its  surface.    Analysis  of  a  small  sample  of  baked‐clay  objects  using  X‐ray  fluorescence  spectrometry suggests that they were made using local clays.    Lithic artifacts are dominated by bifacial implements, primarily stemmed bifaces of various  types.    Use‐wear  analysis  indicates  that  bifaces  were  used  for  a  wide  variety  of  tasks,  including  cutting,  sawing,  and  scraping,  as  well  as  to  tip  dart  points.    Bifaces  were  intensively  used,  maintained,  resharpened,  and  recycled  in  an  attempt  to  conserve  raw  material since no useable tool stone outcrops are located within 100 km of the site.  Other  chipped  stone  tools  include  microliths,  a  few  unifaces,  and  utilized  flakes.    Tallahatta  Quartzite  was  the  primary  raw  material  used  for  stone  tools.    This  material  outcrops  in  southern  Alabama,  150  km  to  the  north.    Cherts  from  the  Marianna  and  Wrights  Creek  28

Introduction 

Quarry  Clusters  also  are  present  but  in  very  low  amounts.    Quartz  and  quartzite  were  obtained  from  secondary  gravel  deposits  in  the  Yellow  River  and  its  tributaries  and  sandstone also was obtained from outcrops exposed by alluvial downcutting.  The gravels  were  used  for  hammerstone/abraders  and  use‐wear  analysis  indicates  that  these  implements were likely used in food processing.  Most of the sandstone contains iron and  one  possible  grinding  stone  or  receptacle  with  powdered  ochre  suggest  processing  of  the  sandstone for its iron content in the making of pigments, dyes, or medicines.    Faunal  remains  are  dominated  by  marine  resources,  particularly  oysters,  scallops,  quahog  clams,  and  a  variety  of  fish  species,  particularly  jacks.    Terrestrial  resources  include  deer,  rabbit,  gray  squirrel,  raccoon,  turkey,  crow,  ducks,  alligator,  turtles,  gopher  tortoise,  salamander,  and  lizards.  Eighty‐nine  different  taxa  were  identified  in  the  298,000+  vertebrate and invertebrate specimens that were analyzed.  This comprised only a sample  of  the  faunal  remains  recovered  from  the  site  and  it  is  estimated  that  a  minimum  of  500,000 remains are available for future research.    Plant  remains  were  not  numerous  but  they  provide  important  dietary  and  seasonality  information.    Edible  plant  remains  include  hickory  nut  shells,  acorn  shells,  an  unidentified  seed,  and  an  unidentified  fleshy  fruit  fragment.    Pine  was  the  primary  wood  use  for  fuel  along  with  hickory  and  oak.    Phytolith  and  starch  grain  analysis  of  a  griddle  fragment  identified bottle gourd, hackberry, possible coontie, as well as phytoliths derived from wild  grasses (Chloridoideae, Panicoideae,  Bambusoideae, and Pooideae), sedges (Cyperaceae),  and  from  the  bark  of  members  of  the  Annonaceae  (custard  apple)  family.    Although  the  seeds  of  wild  grasses  are  edible,  the  grasses  may  have  been  present  in  the  surrounding  environment or have been used to make fiber.   Bottle gourd is a hard‐shelled fruit that can  be  hollowed  out  and  used  as  a  container  and  the  seeds  may  be  consumed  or  used  to  produce oil.  The presence of bottle gourd suggests the possibility of limited horticulture.      A  number  of  seasonality  indicators  were  identified  in  the  faunal  (quahog  clam,  fish),  botanical  (hickory  nuts  and  acorns),  phytolith  (hackberry),  and  starch  grain  (coontie)  samples.    While  all  seasons  of  the  year  are  represented  by  these  samples,  there  is  a  congruence of all indicators during the early spring through late fall months.  This does not  rule  out  the  possibility  that  the  site  was  visited  during  the  winter  months,  but  it  appears  that occupation during that time of the year was not intensive.  The fact that no evidence of  substantial structures was found also suggests that occupation occurred during the warmer  months of the year.    Residential habitation also is indicated by the large pit features and hearths, the presence of  probable food processing tools, and the use of local clays to manufacture the large number  of baked‐clay objects used in cooking.  Residential mobility is indicated by the presence of  lithic  materials  from  distant  locales  and  a  stone  technology  that  emphasized  portability,  flexibility,  and  maintainability.    The  predominance  of  Tallahatta  Quartzite  and  small  amounts of Suwannee Limestone and Ocala Limestone cherts, indicate that the people who  inhabited Bayou Park spent part of the year in the interior where these resources occur.    29

Bayou Park, 8OK898 

Only a few pieces of exotic metamorphic stone were recovered from the site, all fragments  of  ground  stone  tools,  and  no  steatite,  jasper,  or  exotic  cherts  have  been  identified.   However,  it  is  likely  that  the  people  who  occupied  Bayou  Park  interacted  with  their  neighbors  along  Choctawhatchee  Bay.    It  is  noteworthy  that  no  burials  have  been  encountered at Bayou Park.  If people were living at the site for several months of the year,  it  would  be  expected  that  some  deaths  would  occur.    This  suggests  that  the  dead  were  interred elsewhere, perhaps in a communal cemetery or burial mound.    In summary, Bayou Park was occupied primarily during the Elliott’s Point phase of the Late  Archaic  period  sometime  between  4481‐3766  cal  BP  (2529‐1817  cal  BC).    The  site  is  believed  to  represent  primarily  a  seasonal  (spring‐fall)  habitation  site  used  by  small  kin‐ based groups.  Subsistence activities were focused on marine resources, although terrestrial  resources  also  were  exploited.    Structures  were  small  and  of  both  closed  and  open  construction.  Based on the lithic materials from the site, it is hypothesized that people left  Bayou Park during the late fall or early winter and traveled north into the interior for the  winter  months.    There  they  exploited  Tallahatta  Quartzite  and  possibly  cherts  from  the  Marianna and Wrights Creek Quarry Clusters before moving back to the coast again during  the late spring.     

CONCLUSION    The Bayou Park site has provided a unique opportunity to examine a coastal Late Archaic,  Elliott’s  Point  Complex site  in  some  detail.    It  has  produced  important  information  on  the  age  of  the  deposits,  subsistence,  settlement  layout,  seasonality,  and  mobility  patterns  which should assist in a better understanding of Elliott’s Point in the Choctawhatchee Bay  region.  The following chapters detail these efforts and provide the supporting data for our  interpretations of this interesting and significant archaeological site.           

30

2

ENVIRONMENTAL AND   ARCHAEOLOGICAL CONTEXTS    

  

    This  chapter  provides  the  reader  with  environmental  and  archaeological  overviews  of  northwest  Florida  and  particularly  the  area  surrounding  Choctawhatchee  Bay.    These  will  provide  a  context  for  understanding  and  interpreting  the  Bayou  Park  site  (8OK898).    The  focus of the environmental overview is on those natural resources that would have made  the  area  attractive  to  Late  Archaic‐period  populations.    To  the  extent  that  they  are  understood,  temporal  changes  in  climate,  rainfall  patterns,  sea‐levels  and  their  effect  on  surface water, plants, and animal life are discussed since significant changes in these would  have  affected  how  native  peoples  adapted  their  settlement  and  subsistence  strategies  to  local  conditions.    This  is  followed  by  a  review  of  what  is  currently  known  about  the  pre‐ contact Native American occupation of the region.  As extensive and detailed overviews of  the environment, paleoenvironment, and prehistory of the Eglin AFB have been presented  in  previous  reports  on  file  with  the  Cultural  Resources  Section  (see,  in  particular,  Thomas  and Campbell 1993, and most recently, Morehead et al. 2011), this chapter will not attempt  to  repeat  these  summaries.    Instead,  we  focus  on  environmental  conditions  related  specifically to the Bayou Park site and its immediate surroundings followed by a discussion  of the period of time during which the most intensive occupation of Bayou Park occurred,  ca. 4300 B.P. to 3700 BP.    

REGIONAL ENVIRONMENTAL SETTING   

Physiography  The Bayou Park site is located within the Gulf Coastal Lowlands subdivision of the Coastal  Plains  Province  (Brooks  1981;  Puri  and  Vernon  1964).    The  Coastal  Lowlands  consists  of  relatively  undissected,  nearly  level  plains  which  extend  from  the  Gulf  coast  inland  for  approximately 16 to 19 km (10 to 12 miles).  Maximum elevation in the Coastal Lowlands is  typically  less  than  30  m  (100  ft)  above  mean  sea  level  (amsl).    The  Coastal  Lowlands  are  believed  to  have  formed  as  a  result  of  a  transgressive  sea  that  eroded  into  the  Citronelle  hills, leaving a flat terrace after the sea receded (Schmidt 1978).  Subsequent erosion of the  highlands and deposition in the lowlands created subtle surface variations (hills and ridges).   A portion of the Bayou Park site is situated on one of these low hills, approximately 300 m  northwest of Postl Lake (Figure 2.1).    The northern portion of the Coastal Plains in northwest Florida is referred to as the Western  Highlands.    This  is  a  southward  sloping  plateau  that  has  been  dissected  by  numerous  streams  resulting  in  rolling  hills  and  more  substantial  topographic  relief,  with  elevations  reaching as high as 88 m (290 ft) amsl.   Many of the streams have steep‐sided valley walls, 

Bayou Park, 8OK898 

Figure 2.1.  Portion of USGS Destin, FL 7.5’ quadrangle map (top) showing the  location of 8OK898 within Eglin AFB and a topographic map of the site (bottom)  showing the elevated hill system on which much of the Bayou Park site rests. 

32 

 

 

Environmental and Archaeological Contexts 

Figure 2.2.  Physiographic divisions and subdivisions around 8OK898  and Choctawhatchee Bay (Brooks 1981).

a  result  of  downcutting  into  the  clays  that  comprise  the  underlying  Citronelle  Formation.  Brooks (1981) refers to this geomorphic feature as the Eglin Ridge and subdivides it into two  sections ‐‐ the Eastern Clay Hills and Western Sandhills (Figure 2.2).  The Eastern Clay Hills  are  more  deeply  eroded  by  natural  watercourses  and  exhibit  soils  with  large  amounts  of  clay.    The  Western  Sandhills  are  much  less  eroded  by  surface  streams  than  the  eastern  portions of the ridge and therefore exhibit thick sand deposits.    During the Pliocene and Pleistocene Epochs, expansion and recession of glacial ice masses  created episodic fluctuations in sea level, leading to the creation of marine terraces.  These  terraces are of prime geomorphic importance and are defined more as landscape features  than lithologically distinctive stratigraphic or depositional units.  The landforms slope gently  seaward, often terminating landward by an erosionally produced shoreline scarp.  Though  debate continues regarding their ages and exact locations, five general terraces have been  recognized: Silver Bluff Complex, Pamlico, Penholoway, a high terrace complex of multiple,  poorly  expressed  surfaces,  and  an  upland  surface  (Johnson  and  Fredlund  1993).  Each  of  these  terraces  exhibits  some  degree  of  differentiation  in  the  vegetative  communities  present due to changes in elevations and drainage characteristics.    33

Bayou Park, 8OK898 

  Bayou Park is situated on the most recent of these terraces, the Silver Bluff.  Johnson and  Fredlund  (1993:37)  discuss  the  various  interpretations  of  the  Silver  Bluff  Complex  and  conclude that while its base may be primarily Pleistocene in age, its surface has a Holocene  component,  perhaps  due  to  eolian  deposition  during  the  last  5000  years  (Johnson  and  Fredlund 1993:71)    Running east/west through Okaloosa County, approximately 10‐15 km north of Bayou Park,  is a major marine terrace marked by the Cody Scarp.  The Cody Scarp is the northernmost  Pleistocene terrace escarpment, a prominent transition between the clay‐rich Miocene soils  of the northern Florida panhandle and the Pleistocene sands that typify coastal zones (Platt  and Schwartz 1990).    Another major geomorphic feature is Santa Rosa Island and its associated lagoons and bays.   This  large  barrier  island  complex  includes  river‐mouth  swamps  and  marshes,  coastal  terraces, the bay, and the barrier bar/island with its active and relict sand dunes, bay‐mouth  spits, and submerged shell reefs.  Access to the Gulf from Choctawhatchee Bay is through  East  Pass  near  Moreno  Point.    It  is  believed  that  this  pass  narrowed  around  1000  BC,  reducing the salinity of the bay and altering its shellfish ecology (Goldsmith 1966), an event  that is discussed later in greater detail.    

Geology, Clay, and Lithic Resources  Geologically,  the  region  is  covered  by  surficial  sands.  The  sands  are  primarily  Pleistocene  marine deposits that have been reworked and redeposited by aeolian transport and fluvial  erosion  during  the  Holocene.  According  to  the  Natural  Resources  Conservation  Service   (http://websoilsurvey.nrcs.usda.gov/app/WebSoilSurvey.aspx),  the  moderately  well‐ drained  Foxworth  sand,  0‐5%  slopes,  is  the  primary  soil  type  within  the  boundaries  of  8OK898;  however,  archaeological  excavations  have  identified  soil  profiles  in  the  southeastern  part  of  the  site  that  resemble  Kureb  and  Resota  series  (e.g.,  Mikell  et  al.  1996:43‐44).  The  former  is  excessively  drained  and  the  latter  moderately  well‐drained  (http://www.mo15.nrcs.usda.gov/).    Both  soils  are  characterized  by  a  loose,  sandy  composition, with low available water capacity due to very rapid permeability.      These  surface  sands  rest  on  the  underlying  Citronelle  Formation  which  consists  of  non‐ marine quartz sands interspersed with gravel and clay lenses (Clark and Schmidt 1982; T. M.  Scott  et  al.  2001).    The  clay  is  primarily  highly  weathered  kaolinite  containing  heavy  minerals  and  mica  (Means  2009).    Citronelle  sediments  have  been  interpreted  as  representing  the  deposits  of  relict  rivers  that  emptied  into  the  Gulf  of  Mexico  during  the  Pliocene  (Clark  and  Schmidt  1982;  Matson  1916;  Otvos  2004a).    Recent  research  in  Okaloosa  and  Walton  counties  (Means  2009)  has  identified  marine  fossils  in  clayey  sediments  that  previously  have  been  assigned  to  the  Citronelle  Formation  indicating  that  they have a marine origin.  However, since the Citronelle Formation is defined lithologically  by an absence of fossils (Clark and Schmidt 1982:33; Marsh 1966:69), it is unclear whether  these  marine‐derived  sediments  are  part  of  that  formation  or  are  associated  with  the  34 

 

 

Environmental and Archaeological Contexts 

informally  named  Miocene  Coarse  Clastics  (Marsh  1966:69).    The  latter  is  described  as  containing sand, clay, gravel, and shells along with the minerals pyrite, mica, and glauconite  (Clark  and  Schmidt  1982:31).    These  characteristics  almost  exactly  describe  the  clayey  sediments discussed by Means.  Regardless of which formation it belongs to, clay exposures  are found in cutbanks along the region’s more deeply incised streams and rivers, including  the Yellow River which is located along the northern boundary of Eglin AFB (Means 2009:12;  Joe Meyer, personal communication, 2012).      The  Alum  Bluff  Group  is  another  deposit  of  sand  and  clay  with  traces  of  heavy  minerals,  mica,  and  occasional  marine  shells  (Clark  and  Schmidt  1982).    It  grades  laterally  into  the  Miocene  Coarse  Clastics  and  is  known  to  outcrop  in  deeply  incised  streams  west  of  the  Yellow River (Means 2009:11).      Chert‐bearing  limestones  of  early  Miocene,  Oligocene,  and  Eocene  age  are  all  subsurface  formations  with  no  known  exposures  in  the  Coastal  Lowlands.    In  fact,  of  the  numerous  chert outcrops recorded across the state of Florida by Upchurch et al. (1982), only two are  relatively  near  the  Bayou  Park  site:  the  Wright  Creek  and  Marianna  Quarry  Clusters  (QC)  (Figure 2.3).  Cherts in both of these areas formed through the replacement of limestone by  silica which precipitated out of the overlying Hawthorn Formation.      The Wright Creek Quarry Cluster consists of several chert outcrops located along the upper  Choctawhatchee River and its tributary, Wrights Creek.  The Ocala Limestone Formation is  the  parent  rock  within which  most  of  the  chert  from  this region  has  formed.      The  cherts  from  this  formation  tend  to  have  a  grainstone  fabric  and  contain  numerous  foraminifera  casts, particularly those of the Orbitoid family (Upchurch et al. 1982:104).  Some Suwannee  Limestone cherts may also be present.  The closest documented chert outcrop is about 70  km northeast of Bayou Park on the south side of Wrights Creek just north of State Road 79.   Access to chert in this area could have been accomplished easily by canoe travel across the  bay and up the Choctawhatchee River.    The Marianna Quarry Cluster is located farther east and includes the upper reaches of the  Chipola and Apalachicola Rivers.  This group of chert exposures is quite variable with three  different  chert‐bearing  limestone  formations  present:  Tampa/St.  Marks,  Suwannee,  and  Ocala.    Of  these  three,  the  Suwannee  Limestone  cherts  seem  to  be  most  diagnostic  and  these cherts are present at the well‐known Two Egg Quarry located about 145 km northeast  of Bayou Park.  Chert from this quarry is described as “gray, pink, and lavender, with gray‐ colored material predominating” (Sharon and Watson 1971:77).  Upchurch et al. (1982:106)  also describe cherts from his site as light‐colored with low iron content and few observable  fossils.  Light‐colored cherts found on sites located on Eglin AFB and surrounding area often  are referred to as “Two Egg chert.”  Although there is no data to indicate that these were  derived from the Two Egg quarry, they do fit the description of Suwannee Limestone cherts  from the Marianna Quarry Cluster generally.    35

Bayou Park, 8OK898 

Figure 2.3.  Map of geological formations in the southeastern U.S. that contain  knappable toolstone.  Modified from Price (2008). 

Another important source of toolstone for prehistoric people in the Florida panhandle is the  Tallahatta  Formation  (Haywick  and  Carr  2004;  Price  2008).    This  formation  outcrops  in  a  band  that  extends  from  the  Chattahoochee  River  in  southwest  Georgia  west  through  southern  Alabama  and  into  eastern  Mississippi.    It  contains  localized  deposits  of  silicified  sandstone,  referred  to  in  the  archaeological  literature  as  Tallahatta  Quartzite,  which  are  exposed in stream valleys (Lloyd et al. 1983).  The material is typically a dull, opaque tan in  color with numerous quartz grains which provide it with a glittery appearance.  According to  Lloyd  et  al.  (1983:127),  the  outcrops  are  thickest  in  western  Alabama  in  the  counties  of  Clark, Washington, and Choctaw.  At a minimum, this would place these source areas 120 to  130  km  from  Bayou  Park.    Outcrops  are  present  along  the  lower  Chattahoochee  in  southeast  Alabama  and  southwest  Georgia  (Nancy  White,  personal  communication  2012)  and could have been accessed by coastal peoples via the Apalachicola River.  Several sites  along the Apalachicola River that contain artifacts made of Tallahatta Quartzite have been  documented (J. S. Austin 2003).    Finally,  the  Citronelle  Formation  contains  chert  and  quartz  gravels  that  could  have  been  used for stone tool production, although extensive evidence for the use of this material is  lacking at sites around Choctawhatchee Bay.  Means (2009:39) notes that no chert gravels  were identified in any of his sample locations, although quartzite gravels, some up to 5 cm  in diameter, are present in some locations (2009:41).  36 

 

 

Environmental and Archaeological Contexts 

 

Hydrology  Three  major  hydrologic  basins  are  present  within  the  Eglin  AFB:  Choctawhatchee  Bay,  Yellow River Basin, and Pensacola Bay (Science Application International Corporation [SAIC]  2010:4‐6).    The  Bayou  Park  site  is  influenced  primarily  by  the  Choctawhatchee  Bay  basin,  which  in  turn  is  influenced  by  the  western  panhandle  aquifer,  rainfall  patterns,  and  the  watershed of the Choctawhatchee River.     

Aquifer and Rainfall  The  upper  aquifer  of  the  Florida  panhandle  is  dominated  by  a  surficial  sand  and  gravel  aquifer.    The  aquifer  is  divided  into  three  zones  with  alternating  permeabilities  (Miller  1997:75).  The uppermost layer, referred to as the surficial zone, is highly permeable and is  recharged  by  precipitation.    Most  of  the  water  in  this  zone  moves  laterally  along  the  less  permeable middle zone and is discharged as base flow in streams or as steepheads.  Where  they  are  well‐developed  and  unaffected  by  logging  and  subsequent  erosion,  steephead  springs produce a significant amount of potable water and would have provided a valuable  resource to prehistoric populations (Fredlund and Johnson 1993).    Rainfall also percolates through the middle zone and recharges the third, or lowest, zone.   Water  contained  within  this  lower  permeable  zone  is  under  artesian  pressure  due  to  the  overlying confining layer of marl and clay, of which the Citronelle Formation is a part.  This  lower zone is the main water producing zone for the region and it is tapped by numerous  wells.    Natural  discharge  occurs  where  streams  have  dissected  the  landscape  and  intercepted the lower zone; these streams channel water to the Bay and Gulf.      The  upper  sand  and  gravel  aquifer  is  separated  from  the  Floridan  Aquifer  by  another  confining  unit  of  clay,  the  Pensacola  Clay  (SAIC  2010:4‐12).    Unlike  to  the  east  of  Choctawhatchee  Bay  in  the  Marianna  Lowlands,  the  Floridan  Aquifer  is  buried  beneath  several  hundred  feet  of  sand  and  clay  and  is  not  a  major  source  of  water  for  coastal  panhandle residents.  Along the northern reaches of the Choctawhatchee River, however,  the Floridan Aquifer is at or near the surface and is frequently exposed in the river’s channel  as it nears the Alabama border (Barrios 2005).    The saturated thickness of the sand and gravel aquifer, based on current water conditions,  varies  between  15  and  38  meters.    Variation  in  the  thickness  is  heavily  dependent  on  seasonal  rains,  with  saturation  thinning  during  periods  of  extend  drought  or  thickening  during increased precipitation (DeFosset 2004:1).      The  climate  of  coastal  Okaloosa  County  is  characterized  as  subtropical  with  humid,  warm  summers  and  mild  winters.    Yearly  rainfall  averages  157.5  cm,  with  a  monthly  average  of  12.9  cm.    Much  of  the  annual  rainfall  (44%)  occurs  during  the  months  of  July  through  September,  with  most  falling  during  the  months  of  July  and  September  (Overing  et  al.  1995:1; SAIC 2010:4‐1).  The months of October through February show the lowest trends in  37

Bayou Park, 8OK898 

rainfall.    Extended  dry  periods  can  occur  during  any  season  but  are  most  frequent  during  the spring and fall.   In general, dry periods during the fall tend to be longer in duration, but  the  shorter  spring  dry  periods  often  are  more  severe  due  to  rising  average  temperatures  and a greater demand for moisture by ecosystems (Overing et al. 1995:2).   

Choctawhatchee Bay  Choctawhatchee Bay is a large linear estuary stretching east to west for about 50 km across  Walton and Okaloosa Counties.  The bay ranges in width from 2 to 10 km and in depth from  3 to 13 meters.  The primary source of freshwater for the bay is the Choctawhatchee River  which  spills  into  the  bay  via  a  delta  at  its  eastern  end.    The  Choctawhatchee  River  contributes  the  majority  of  the  bay’s  13,856  sq  km  watershed,  which  expands  northwest  through  Florida  and  into  southern  Alabama  (Ruth  and  Handley  2007:143).    The  Choctawhatchee River is fed by 13 identified springs, both the typical fissure‐type vents and  those described as seep springs, where ground water percolates upwards into diffuse pools  (Barrios  2005:6).    The  annual  average  discharge  rate  of  freshwater  into  the  bay  from  the  river is approximately 243 m3/second (Ruth and Handley 2007:143).  Other sources of fresh  water  include  Turkey  Creek,  Rocky  Branch,  Mullet  Creek,  Trout  Creek,  Lafayette  Creek,  Ramsey Branch, Black Creek, Mitchell River, among other smaller streams.    The principle marine access and only direct route to the Gulf of Mexico is East Pass, located  between  Destin  and  Okaloosa  Island,  approximately  9  km  south‐southeast  of  Bayou  Park.   According  to  Goldsmith  (1966),  this  pass  was  open  prior  to  about  1000  BC  but  became  restricted as a result of westward longshore drift that formed a barrier spit known today as  Moreno  Point.      An  1886  coastal  navigational  chart  shows  that  only  a  narrow  pass  connected  the  two  water  bodies,  but  by  1922  this  had  become  little  more  than  an  intermittent stream (Office of Coast Survey 1886, 1922).  When in 1929 major flooding of  the  Choctawhatchee  River  caused  bay  waters  to  rise  precipitously,  local  residents  reportedly  dug  a  trench  along  this  stream  to  enable  bay  waters  to  escape  into  the  Gulf  (Ruth and Handley 2007:145).  The US Army Corps of Engineers widened the pass in 1930  and has maintained it since.  Access to the Gulf also is provided by Santa Rosa Sound just  northwest of East Pass, and by an intracoastal waterway at the Bay’s east end, just south of   the Choctawhatchee River.  According to Hoyer et al. (2013:455), the opening of the pass  increased salinity in the bay resulting in a loss of freshwater marshes and sea grass beds.    

Boggy Bayou  Situated north of Bayou Park, between the cities of Valparaiso and Niceville, Boggy Bayou  opens  into  the  northwest  side  of  Choctawhatchee  Bay  and  extends  to  the  northwest  for  about  5  km.    The  elongated  bayou  varies  in  width  from  approximately  0.38  km  at  its  narrowest and 0.95 km  at its mouth with depths ranging  4 to  6.7 meters (Campbell et al.   1989:1).  The bayou is fed primarily by Tom’s Creek and Turkey Creek.  The overall salinity of  Boggy  Bayou  ranges  from  10  to  15  ‰  at  the  surface  to  20  to  25  ‰  at  the  bottom  (Butts  1997:4).    Near  the  mouth  of  Boggy  Creek  and  just  north  of  Bayou  Park  is  Weekley  Bayou  which is fed by a small stream.  38 

 

 

Environmental and Archaeological Contexts 

Postl Lake  The south and west boundaries of the  Bayou  Park  site  are  formed  by  drainages  which  empty  into  Postl  Lake.    The  lake  is  named  for  Charles  Postl  (Postil),  formerly  of  Chicago,  who  owned  and  operated  a  health  resort in the 1920s and 1930s (Baker  Block  Museum  n.d.).    The  narrow  L‐ shaped lake is actually a small lagoon  extending  northeast  and  then  north,  paralleling Choctawhatchee Bay.  It is  connected to the Bay at its north end  by a narrow drainage and on the east  by a shallow pass.  It is unclear if this  lagoon  was  present  at  the  time  of  prehistoric  occupation  of  Bayou  Park  Figure 2.4.  Portion of 1935 USGS Villa Tasso 15’  or  if  there  was  ever  a  time  when  it  quadrangle map showing Postl Lake.    was  not  connected  to  the  bay.    Early  maps  of  the  area  show  no  lake  or  lagoon  in  this  location  (Campbell,  Morehead,  et  al.  2009:Figures 14‐16; General Land Office 1851; Office of Coast Survey 1886;).   The first map  that shows Postl Lake is a 1935 USGS quadrangle map (Figure 2.4) which indicates that the  lake was landlocked at that time.  A 1955 USDA aerial photograph (Figure 2.5) shows a small  drainage  at  the  lake’s  north  end  and  what  appears  to  be  a  gap  in  the  lake’s  eastern  shoreline connecting it to the bay.  When this gap began to intercept the lake is unknown.   The USDA aerial also shows clearly the prograding nature of the shoreline east of Postl Lake  toward  the  mouth  of  Boggy  Bayou.    The  lake’s  scalloped  eastern  shoreline  also  is  visible.   Scalloped  shorelines  are  features  seen  on  the  lagoonal  sides  of  barrier  islands  and  are  usually  formed  by  washover  fans  during  major  storm  events  (Liu  and  Fearn  2000).    These  observations suggest that Postl Lake may have formed as a result of a prograding spit which  eventually  closed  off  the  lagoon.  Archaeological  site  8OK72  on  the  south  side  of  the  lake  contained fiber‐tempered sherds as well as later artifacts (Lazarus 1964) suggesting that a  lake or a marsh existed at least as early as 1000 BC.    

Flora   The  portion  of  the  western  panhandle  that  contains  Eglin  AFB  and  the  Bayou  Park  site  is  contained within the Sandhill High Pine region (Myers 1990:Figure 6.14).   The sandhill high  pine  region  extends  from  the  Suwannee  River  westward  into  Alabama  and  Mississippi.   Myers  (1990:177‐186)  and  Harper  (1914:233‐238)  provide  general  descriptions  of  the  vegetation which, as the name suggests, is characterized by forests of longleaf pine (Pinus  palustris).  Harper’s description was made prior to modern development, although the area  had already experienced intensive timber and naval stores harvesting during the nineteenth  century  (Harper  1914:239‐240).    This  early  commercial  exploitation  plus  modern  fire  39

Bayou Park, 8OK898 

Figure 2.5.  USDA aerial photograph showing Postl Lake in 1955. 

suppression has led to the general replacement of longleaf pine by xerophytic oak species  and the development of xeric hammocks.     The  Eglin  AFB  Integrated  Natural  Resources  Management  Plan  (INRMP)  divides  the  base  into  four  broad  ecological  associations:  sandhills,  flatwoods,  barrier  islands,  and  wetlands  (SAIC 2010).  The sandhills association is the most extensive, covering over 360,000 acres.  It  is characterized by an open, savanna‐like structure with a canopy of longleaf pine, a sparse  midstory  of  oaks  and  other  hardwoods,  and  a  groundcover  comprised  mainly  of  grasses,  forbs, and low shrubs.   The pine sandhills contain a high diversity of fire‐adapted species.   Other important flora include a variety of xeric‐adapted deciduous oaks, such as turkey oak  (Quercus laevis), live oak (Quercus virginiana), and blackjack oak (Quercus marilandica), and  herbaceous  ground  cover,  of  which  wiregrass  (Aristida  stricta)  and  little  bluestem  (Schizachyrium scoparium) are the most common.  Associated with pine sandhills are xeric  hammock,  scrub,  upland  pine  forest,  and  slope  forests.    Xeric  hammocks  are  typically  characterized as scrubby, low canopy forests with little understory other than saw palmetto  (Serenoa repens).  Typical botanical taxa include live oak, sand live oak (Quercus geminata),  Chapman’s oak (Quercus chapmanii), turkey oak, southern magnolia (Magnolia grandiflora),  sparkleberry  (Vaccinium  arboreum),  beautyberry  (Callicarpa  americana),  and  yaupon  (Ilex  vomitoria)  (Florida  Natural  Areas  Inventory  [FNAI]  and  Florida  Department  of  Natural  Resources  [FDNR]  1990:6‐7).    Scrub  communities  are  characterized  by  sand  pine  (Pinus  clausa),  sand  live  oak,  scrub  live  oak  (Quercus  inopina),  myrtle  oak  (Quercus  myrtifolia), 

40 

 

 

Environmental and Archaeological Contexts 

Chapman’s oak, saw palmetto, and rosemary (Ceratiola ericoides) (FNAI and FDNR 1990:5‐ 6).      Slope forests are unique mesic settings that result from the formation of steepheads, which  are steep and narrow valleys formed by the downcutting action of flowing springs or seeps.   Steephead  springs  typically  produce  well‐developed,  closed  canopy  hardwood  hammocks  on steep, sheltered slopes.  These natural communities are common in the uplands of the  Florida  panhandle  (Means  1991),  and  Thomas  and  Campbell  (1993)  have  observed  that  steepheads are present over a large area of Eglin AFB in Okaloosa County.  In addition to  potable  water,  the  combination  of  a  cool,  moist  microclimate  and  the  densely  shaded  slopes of a steephead results in conditions conducive to the growth of several species more  typical  of  the  Piedmont  and  Southern  Appalachian  Mountains  (FNAI  and  FDNR  1990:15).   Typical  slope  forest  species  include  black  walnut  (Juglans  nigra),  basswood  (Tilia  americana),  strawberry  bush  (Euonymus  americanus),  Florida  yam  (Dioscorea  floridana),  Solomon’s  seal  (Polygonatum  biflorum),  leatherwood  (Dirca  palustris),  American  beech  (Fagus  grandifolia),  white  oak  (Quercus  alba),  laurel  oak  (Quercus  hemisphaerica),  mockernut  hickory  (Carya  tomentosa),  sarsaparilla  vine  (Smilax  pumila),  witchhazel  (Hamamelis  virginiana),  horse  sugar  (Symplocos  tinctoria),  and  Florida  yew  (Taxus  floridana), among many others (FNAI and FDNR 1990:14‐15).    Pine flatwoods occur on flat, moderately well‐drained sandy soils that are often underlain  by a hardpan.  The canopy consists of slash pine (Pinus elliottii) and longleaf pine, with an  understory  that  varies  from  dense  shrubs,  particularly  saw  palmetto,  to  open  grasses  and  herbs.    Soil  moisture  and  fire  history  are  the  main  factors  that  influence  the  types  of  vegetation that occur in pine flatwoods.     Interspersed within the pine flatwoods community along the coast are maritime hammocks,  one  of  which  is  located  within  the  southern  boundary  of  8OK898.    These  are  typically  narrow bands of hardwood forest characterized by live oak, cabbage palm (Sabal palmetto),  redbay  (Persea  borbonia),  southern  magnolia,  American  holly  (Ilex  opaca),  sea  grape  (Coccoloba uvifera), saw palmetto, and beautyberry (FNAI and FDNR 1990:11‐12).      The  natural  communities  associated  with  the  barrier  island  complex  include  primary  and  secondary dunes, interdune swales, maritime forests, and sand pine scrub.  The vegetation  that  occupies  these  communities  is  influenced  by  coastal  (maritime)  processes  such  as  erosion, deposition, salt spray, and storms.      Wetland types on the Eglin AFB are varied and include baygall, seepage slope, dry prairie,  flatwood  lake,  floodplain  forest,  floodplain  swamp,  bottomland  forest,  wet  prairie,  hydric  hammock,  blackwater  stream,  alluvial  stream,  spring  run  stream,  steepheads,  marsh  lake,  slough,  dome  swamp,  strand  swamp,  basin  marsh,  depression  marsh,  floodplain  marsh,  sandhill  upland  lake,  bog,  freshwater  tidal  swamp,  and  salt  marsh  (SAIC  2010).    Artesian  spring‐fed  streams  are  common  along  the  upper  reaches  of  the  Choctawhatchee  River  41

Bayou Park, 8OK898 

(Barrios  2005),  but  most  of  the  streams  near  8OK898  are  seepage  streams  or  blackwater  streams.    Steepheads  are  clear  to  lightly‐colored,  relatively  short,  shallow,  and  narrow  water  courses  that  originate  from  shallow  groundwaters  that  have  percolated  through  deep, sandy uplands.  Most are perennial and flow from north to south towards the coast.   Most  of  the  natural  ponds  and  non‐forested  wetlands  are  usually  relatively  small.    Some  hold  water  permanently,  while  others  are  seasonal.    Some  contain  herbaceous  or  woody  vegetation (SAIC 2010:4‐6, 4‐9).   Figure 2.6 shows the distribution of wetlands on Eglin AFB  in relation to 8OK898.    Marine and marsh environments associated with Choctawhatchee Bay would have been a  critical  source  of  subsistence and  other  resources  for  the  prehistoric  inhabitants  of  Bayou  Park.    Saltwater  marsh  areas  along  the  bay  shores  exhibit  a  diverse  grouping  of  botanical  taxa.    These  taxa  include  black  needlerush  (Juncus  gerardii),  smooth  cordgrass  (Spartina  alterniflora), saltgrass (Distichlis spicata), marsh elder (Iva annua), and cattail (Typha).   

Fauna  The sandhill high pine, pine flatwoods, and Choctawhatchee Bay provided a wide range of  animal species for prehistoric fishers and hunters.  Terrestrial species that would have been  commonly  available  to  native  populations  in  both  the  sandhill  and  the  pine  flatwood  environments  include  white‐tailed  deer  (Odocoileus  virginiana),  Florida  black  bear  (Ursus  americanus  floridanus),  bobcat  (Lynx  rufus),  gray  fox  (Urocyon  cinereoargenteus),  raccoon  (Procyon lotor), opossum (Didelphis virginiana), box turtle (Terrapene carolinia), and eastern  diamondback  rattlesnake  (Crotalus  adamanteus)  (SAIC  2010:Table  5‐2).    The  bayous  branching out of Choctawhatchee Bay and boarding Bayou Park would have also provided  access to alligator (Alligator mississippiensis) and numerous species of aquatic turtles, such  as  alligator  snapping  turtle  (Chelydra  serpentine),  Florida  soft  shell  (Apalone  ferox),  pond  sliders (Trachemys scripta), coastal plain cooter (Pseudemys floridana), and mud and musk  turtles (Kinosternidae).    The  presence  of  highly  productive  seagrass  beds  and  variable  salinities  due  to  the  intermixing  of  freshwater  from  the  Choctawhatchee  River  and  Gulf  tidal  waters  creates  a  diverse  range  of  habitats  for  both  invertebrate  and  vertebrate  resources.    The  benthic  habitat of the bay is dominated by low relief sand and mud flats interspersed with oyster  and  seagrass  beds  (Fox  et  al.  2002:21).    Figure  2.7  reproduces  a  map  of  seagrass  beds  in  Choctawhatchee Bay ca. 1972 presented in Ruth and Handley (2007).  It shows that most  seagrass beds are located in the western bay, near East Pass.  Seagrass beds are noticeably  absent around the mouth of Boggy Bayou, a situation that was also noted by Butts (1997) in  his  survey.    These  researchers  indicated  that  this  is  likely  a  result  of  high  levels  of  heavy  metals probably from a nearby boat yard (Butts 1997:21).      Common  species  that  would  likely  have  been  available  in  the  bay  include  eastern  oysters  (Crassostrea  virginica),  shrimp  (Penaeus  spp.),  blue  crab  (Callinectes  sapidus),  largemouth  bass  (Micropterus  salmoides),  spotted  seatrout  (Cynoscion  nebulosus),  Gulf  menhaden 42 

 

Figure 2.6  Map of Eglin AFB showing the distribution of wetlands in relation to Bayou Park (8OK898).  SOURCE: Science Application International Corporation 2010:Figure 4‐4. 

  Environmental and Archaeological Contexts 

43

Bayou Park, 8OK898 

Figure 2.7.  Seagrass beds in Choctawhatchee bay ca. 1972.  Reproduced from Ruth and Handley (2007). 

(Brevoortia  patronus),  red  drum  (Sciaenops  ocellatus),  Gulf  flounder  (Paralichthys  albigutta), striped mullet (Mugil cephalus), white mullet (Mugil curema), and Gulf sturgeon  (Acipenser oxyrinchus) (Ruth and Handly 2007:143).     

PALEOENVIRONMENTAL CONDITIONS   

Late Pleistocene‐Early Holocene  From  18,000  to  15,000 BP,  immediately  following  the  Pleistocene   glacial   maximum,  Florida’s  climate  and  vegetation  were  influenced  by  the  effects  of  continental  glaciers  located  farther  north  and  by  the  release  of  meltwater  and  ice  rafts  into  the  oceans  as  worldwide climate began to moderate (Alley and Clark 1999; Clark et al. 1999).  Relatively  cool  conditions  are  believed  to  have  prevailed  until  about  13,000  BP  (Watts  and  Hansen  1994).   Gates (1976) has suggested that the temperature may have been as much as 8° to  10° C cooler.  Pollen evidence from Florida lakes combined with data from deep sea cores  (Grimm et al. 1993) suggest variable wet and dry conditions during this time.      At about 13,000 years ago, the climate cooled and extreme aridity may have occurred as a  result  of  the  release  of  massive  amounts  of  glacial  meltwater  into  the  Atlantic  Ocean.   Dunbar  (2002:160‐161)  reports  that  water  levels  in  the  Aucilla  River  were  extremely  low  between 13,000 and 12,700 BP.  He also notes a retreat of spruce in north Florida during  this time, an indication of decreased moisture.  Vegetation in the Southeast lost much of its  diversity.    Dense  mixed  hardwood  forests  expanded,  replacing  the  northern  arboreal  taxa  which  had  proliferated  during  the  preceding  period  (Delcourt  and  Delcourt  1981).    Water  levels  continued  to  be depressed  in  both  north  and  south  Florida,  due  to  reduced  rainfall  and low sea levels, until about 11,700 BP when wetter conditions returned in north Florida  44 

 

 

Environmental and Archaeological Contexts 

(Dunbar 2002).  In south Florida, environmental conditions may have been slightly different.   The region apparently continued to experience a dry climate and a depressed water table  during this time (Grimm et al. 1993; Watts and Hansen 1988, 1994), although some surface  water apparently accumulated in shallow basins (Dunbar 2002:165, 173‐174).  Using pollen  data from Lake Tulane in Highlands County, Grimm et al. (1993:199) show oscillations in wet  and  dry  conditions,  but  the  magnitude  of  the  oscillations  is  not  as  great  as  during  the  preceding 10,000‐year period.    From  10,000  BP  to  8500  BP,  relatively  dry  conditions  are  believe  to  have  been  present  in  central  and  south  Florida  (Watts  et  al.  1996;  Watts  and  Hansen  1988:311‐313).    At  Lake  Tulane,  large  amounts  of  charcoal  are  present  in  sediments  dating  between  about  11,000‐7000  BP  indicating  a  prevalence  of  fires,  which  suggests  that  an  abundance  of  dry  fuel  was  present  (Watts  and  Hansen  1988:311).    The  inference  is  that  precipitation  was  minimal  during  this  time.    This  is  supported  by  the  absence  of  organic  sedimentation  at  other  lakes in  southern  Georgia  and  northern  and  central  Florida  prior to  8500  BP  (Watts  1969, 1971, 1975; Watts and Hansen 1988:311‐313).  In effect these lakes were dry basins.  This pattern is repeated at Little Salt Springs and Warm Mineral Springs in Sarasota County.   Clausen et al. (1979:610‐611) report no aquatic or marsh plants in pollen dated to 9920 BP  at Little Salt Springs and dry conditions also are indicated at Warm Mineral Springs between  9870‐8520 BP (Clausen et al. 1975:197).  In north Florida, Dunbar (2002:175‐176) indicates  that  dry  conditions  prevailed  there  between  10,000  and  9000  B.P.,  but  that  water  levels  gradually  began  to  rise  shortly  thereafter,  reaching  near‐modern  levels  by  8500‐8000  B.P.   (Watts 1969, 1971, 1983; Watts and Stuiver 1980).    Fredlund  and  Johnson  (1993)  conducted  fossil  pollen  analyses  at  four  sites  on  Eglin  AFB,  providing  local  paleoenvironmental  data  and  allowing  for  the  reconstruction  of  the  Holocene  history  of  the  pine‐oak  forests  of  the  region  around  8OK898.    The  evidence  suggests  that,  as  with  patterns  observed  regionally,  the  local  climate  shifted  from  one  of  less  annual  rainfall  to  one  of  more  mesic  conditions  with  a  more  seasonally  variable  moisture  regime  around  8400  BP.    The  changes  in  pollen  percentages  and  accumulation  rates  which  were  observed  for  tree  and  shrub  taxa  document  a  1200‐year  period  of  vegetative readjustment following the onset of these climatic changes.  Prior to this period  an open oak‐pine forest containing thickets of xeric shrubs were present.  With the onset of  a more mesic climate, pines began to increase in abundance.   Fires associated with the new  climatic regime selected for fire‐tolerant longleaf pine seedlings rather than those of other  tree species.  These selective processes continued to shift the proportions of arboreal taxa  until  about  7200  BP,  when  the  longleaf  pine  forests  reached  the  state  of  dynamic  equilibrium  that  was  observed  historically  within  the  region  (Fredlund  and  Johnson  1993:120‐121).   

Sea‐Level Rise  Associated with these climatic shifts, was a major global rise in sea level as glaciers melted  (summarized for Florida by Balsillie and Donoghue 2004; Dorsey 1997).  At 18,000 BP sea‐ 45

Bayou Park, 8OK898 

level stood at ‐120 meters.  The subsequent rise was slow while glacial conditions prevailed  at high latitudes but became very rapid during the latest Pleistocene and earliest Holocene,  rising nearly 100 m between 15,000 and 8000 BP (Dorsey 1997:Figure 5).    The major changes in Pleistocene versus Holocene sea levels affected Florida in three major  ways.  First, because of the very broad, gently sloping continental shelf on the Gulf Coast,  higher sea levels greatly reduced the land area of the Florida Platform.  Second, given the  region’s  porous  limestone  bedrock,  the  higher  sea  levels  of  post‐glacial  times  resulted  in  much  higher  groundwater  levels  in  the  artesian  aquifer.    This,  coupled  with  increasing  rainfall and warmer climate during the Holocene, had a dramatic effect on water resources  for  animals  and  humans.    Third,  more  abundant  water  and  a  warmer,  wetter  climate  combined to create more modern native plant communities.        As  eustatic  sea  level  rise  slowed  and  began  to  fluctuate  after  about  8000  B.P.,  stream  systems  cut  multiple  steepheads  and  then  stabilized;  their  valley  bottoms  filled  with  primarily pine detritus from the surrounding forest vegetation.   Santa Rosa Island formed  during this time as well (Johnson and Fredlund 1993).     

Mid‐to‐Late Holocene  Modern  types  of  native  plant  communities,  dominated  by  coniferous  forests,  mixed  coniferous‐deciduous  forest,  deciduous  forest,  and  various  types  of  wetland  vegetation,  began  to  develop  around  8500‐8000  BP  (Fredlund  and  Johnson  1993;  Watts  et  al.  1996;  Watts and Hansen 1988).  These conditions have been fairly stable since that time, although  periodic, short‐term fluctuations in climate, precipitation, and sea level have occurred as a  result  of  glacial  expansion  and  contraction.    At  the  local  scale,  these  climatic  fluctuations  resulted in conditions that were both wetter and drier than today.  For example, Gleason et  al. (1974:311) report evidence of ancient fires in a peat bog in Marion County, and from this  infer four separate periods of severe drought between 4030 and 2900 B.P.   Fredlund and  Johnson (1993:118) report that cores from Eglin AFB display perturbations in the ratios of  pine  to  oak  pollen  after  7200  BP  and  they  suggest  that  these  are  related  to  short‐term  climatic fluctuations.    Sea  level  also  was  higher  during  this  time,  although  estimates  of  how  high  vary  among  researchers.  Fairbridge (1974) has estimated that it was 3‐4 m higher than present.  Using  data from nearby St. Vincent Island, Stapor and Tanner (1977) suggest a mid‐Holocene high  of  1.5  m  above  present  levels  (see  also  Stapor  et  al.  1991;  Tanner  1991).    Scholl  et  al.  (1969:Figure  2)  offer  a  more  conservative  estimate  of  4  m  below  present  levels  while  Faught and Donoghue (1994) placed sea level at +2.45 m at 5500 BP.       By 4000 BP the climate, water levels, and plant communities of Florida attained essentially  modern  conditions,  although  multidecadal‐  and  century‐scale  variability  did  occur  (Mayewski et al. 2004; Poore 2008; Soto 2005).  Of particular interest for this study is the  evidence for a cool, dry event between 4200 to 3800 BP followed by a period of extreme  aridity between 3500 and 2500 BP.  These climatic events are found in the global climate  46 

 

 

Environmental and Archaeological Contexts 

record at temporally equivalent periods (Mayewski et al. 2004), but their local effects were  variable and, for Florida, not well understood (Poore 2008).  The rate of sea‐level rise also  slowed and many of the barrier islands around the Gulf began to form near their present  position (Williams et al. 1999).  The slowing of sea‐level rise allowed progradation of several  river  deltas  around  the  Gulf  of  Mexico  that  currently  support  coastal  swamp  forest.    For  example,  swamp  and  bottomland  forest  near  the  Apalachicola  River  occupy  a  delta  that  appears to have expanded since 4000 BP (Donoghue and White 1994).      Several studies have provided evidence of higher and lower sea‐level stands during the past  3000 years (e.g., Balsillie and Donoghue 2004; Fairbridge 1974; Stapor et al. 1991; Tanner  1991, 1992; Walker et al. 1994, 1995).   Dating of shells in beach ridges on St. Vincent Island  and on barrier islands in southwest Florida indicate that the ridges were deposited during  discrete periods which are interpreted as resulting from rises in sea level at ca. 2000 B.P.,  1100 B.P., and 150 B.P. (Stapor et al. 1991; Tanner 1991, 1992).  These periods of higher sea  level  correspond  with  periods  of  warmer  temperatures,  the  Roman  Warm  Period,  the  Medieval Warm Period, and modern climatic conditions, respectively.  Around Apalachicola  Bay,  evidence  of  human  habitation  alternates  with  strata  of  estuarine  muds,  suggesting  fluctuations  in  sea‐level  with  associated  movements  in  the  position  of  the  habitable  shoreline (Donoghue and White 1994).      Johnson  and  Fredlund  (1993:52)  note  the  similarities  in  sea  level  data  from  St.  Vincent  Island with data from Eglin AFB, with the latter including the proximity of shell middens to  the  present  shoreline,  submergence  of  middens  below  present  sea  level,  and  the  vertical  position of paleosols and tree stumps relative to modern sea level.  Using these combined  date, Johnson and Fredlund present a model of Holocene sea level fluctuation for Eglin AFB  that  is  reproduced  here  as  Figure  2.8.    According  to  their  reconstruction,  sea  level  during  the Late Archaic period, when Bayou Park was most intensively occupied, would have been  approximately  two  meters  above  present  levels  but  falling  at  approximately  4000  BP,  reaching a low of ‐2 meters at about 3000 BP.  Sea level then began another slow rise to  about  +1  meter  at  around  1000  BP  before  falling  again  to  –  1  m  during  the  Fort  Walton  period.   

Moreno Point  Presently, Moreno Point is the end of a barrier spit located in Destin on East Pass, almost  directly  across  Choctawhatchee  Bay  from  8OK898.    It  has  been  hypothesized  that  the  formation  of  Moreno  Point  had  a  significant  environmental  impact  on  local  coastal  areas,  specifically  the  salinity  of  the  Bay  and  the  effect  this  had  on  local  shellfish  populations  (Goldsmith 1966; Thomas and Campbell 1991:114‐115, 1993:539‐540).  Based on the results  of  his  sedimentary  study  of  Choctawhatchee  Bay,  Goldsmith  (1966)  hypothesized  that  sometime  between  7000  and  3000  BP,  Moreno  Point  began  to  form  as  the  result  of  westward  longshore  drift,  eventually  becoming  a  barrier  spit.    By  about  3000  BP,  the  spit  had reduced the opening of Choctawhatchee Bay to the Gulf of Mexico sufficiently enough  to  restrict  water  exchange  between  the  two  bodies  of  water,  resulting  in  natural  47

Bayou Park, 8OK898 

Figure 2.8.  Reconstructed Holocene sea level fluctuations for the Eglin Air  Force Base/Choctawhatchee Bay region.  Reproduced from Johnson and  Fredlund (1993:Figure 25). 

desalination  of  the  bay.    The  bay’s  reduced  salinity  ultimately  resulted  in  environmental  conditions  unsuitable  for  halophilic  (favoring  high  salinity)  species  of  shellfish  while  also  providing niches for species that are more freshwater tolerant.    The slow accretion of Moreno Point and related morphological events are cited as having  influenced  the  prehistoric  populations  that  occupied  the  shores  of  Choctawhatchee  Bay  throughout  the  Late  Holocene  (Thomas  and  Campbell  1991,  1993).      The  lower  salinity  in  the bay resulting from the formation of Moreno Point is suggested as the factor responsible  for  the  differences  in  shellfish  composition  between  Elliott’s  Point  (Late  Archaic)  and  Deptford  (Early  Woodland)  middens.    The  presence  of  bay  scallop  (Argopecten  irradians)  and quahog (Mercenaria spp.) in earlier Elliott’s Point middens, two species that favor high  salinity habitats, is seen as evidence that Choctawhatchee Bay was open to the Gulf during  this  time  period.    The  subsequent  absence  of  scallop  and  quahog  in  the  middens  of  later  Deptford populations is offered as support of the hypothesis that by 2500 BP the salinity of  the bay had decreased substantially as a result of the accretion of Moreno Point.      It appears that the slow accretion of the barrier spit at Moreno point continued through the  Middle  Woodland  period,  reducing  the  salinity  levels  to  a  point  that  even  eastern  oyster  (Crassostrea  virginica),  a  species  that  can  tolerate  varying  salinity  levels,  could  no  longer  survive.    Rangia  cuneata,  a  species  of  marsh  clam  that  thrives  in  waters  with  very  low  salinity, is the dominant species represented in Middle Woodland (Santa Rosa/Swift Creek)  48 

 

 

Environmental and Archaeological Contexts 

middens  throughout  the  Choctawhatchee  Bay  system  (Thomas  and  Campbell  1993:564‐ 565).    A  return  to  oyster  utilization  is  recorded  archaeologically  for  Weeden  Island  (Late  Woodland)  and  Fort  Walton  (Mississippian)  populations,  and  Thomas  and  Campbell  (1993:565)  posit  that  sometime  between  the  Middle  and  Late  Woodland  periods,  a  significant  event,  such  as  a  major  hurricane,  reopened  the  pass,  increasing  bay  water  salinity to levels suitable for oyster.    

PREHISTORIC OVERVIEW    The prehistory of Choctawhatchee Bay and Eglin AFB has been summarized by Thomas and  Campbell  (1993:489‐637)  with  periodic  updates  presented  in  various  reports  since  that  time.    Table  2.1  is  their  most  recent  update  of  the  region’s  cultural  chronology  and  is  reproduced  from  Morehead  et  al.  (2011:Figure  4).    As  the  Bayou  Park  site  was  occupied  primarily  during  the  Late  Archaic  period,  the  following  sections  provide  a  review  of  Late  Archaic/Elliott’s  Point  developments  followed  by  a  brief  summary  post‐Archaic  developments in the panhandle.    Before proceeding, however, we must first note our use of the term Late Archaic in place of  Thomas and Campbell’s Gulf Formational when referring to the period of time from about  5000 BP to 2500 BP (3000 BC to 500 BC).  As discussed in detail by Thomas and Campbell  (1993:518‐522), the Gulf Formational concept was first put forward by Walthall and Jenkins  (1976) to refer to the period of time when fiber‐tempered pottery first began to appear in  the  Southeastern  US.    While  it  is  formally  used  in  Alabama  and  the  western  Florida  panhandle,  it  has  not  seen  consistent  usage  elsewhere  nor  is  it  referenced  in  the  most  recent regional syntheses (e.g., Bense 1994; Milanich 1994; Sassaman 2010).     While arguing in favor of its usefulness at the local level, Thomas and Campbell (1993:521)  acknowledge  that  the  absence  of  reliable  dates  for  the  appearance  of  fiber‐tempered  pottery in the panhandle may have contributed to the reluctance of some archaeologist to  use  it  in  their  temporal  reconstructions.  As  they  note,  there  are  no  dates  from  the  Choctawhatchee  Bay  region  which would  place  the  appearance  of  fiber‐tempered pottery  any earlier than 3135 BP (1675‐1025 cal BC), and they conclude that fiber‐tempered pottery  was  a  late  addition  to  the  already  existing  Late  Archaic/Elliott’s  Point  material  culture  (Campbell et al. 2004:132; Thomas and Campbell 1993:528, 538, 541).  On the other hand,  White (2003) reports fiber‐tempered pottery as early as 3970 radiocarbon years (rcy) BP in  the  Apalachicola  River  region,  or  2900‐1980  cal  BC.    Further,  Thomas  and  Campbell  (1993:540) admit that the characteristics of Elliott’s Point “do not fit neatly within the Gulf  Formational framework.”     While  we  do  not  intend  to  enter  the  debate  on  whether  the  Gulf  Formational  concept  is  applicable  or  useful,  we  have  decided  to  use  the  more  general  Late  Archaic  designation  when  referring  to  the  period  of  time  that  encompassed  the  Elliott’s  Point  Complex.    This  decision is based on its more uniform usage throughout Florida and the SEUS.  Its use also  49

Bayou Park, 8OK898 

  Table 2.1.  Native American culture sequence in the Choctawhatchee Bay region  (modified from Morehead et al. 2011:Figure 4).  STAGE 

CALENDAR  YEARS 

PERIOD 

CULTURE 

  Historic 

Mississippian 

Woodland 

Gulf  Formational 

Archaic 

Lithic 

 

50 

AD 1800— AD 1700— AD 1600— AD 1500— AD 1400— AD 1300— AD 1200— AD 1100— AD 1000— AD 900—  AD 800—  AD 700—  AD 600—  AD 500—  AD 400—  AD 300—  AD 200—  AD 100—  —  100 BC —  200 BC —  300 BC —  400 BC —  500 BC —  600 BC —  700 BC —  800 BC —  900 BC —  1000 BC —  2000 BC —  3000 BC —  4000 BC —  5000 BC —  6000 BC —  7000 BC —  8000 BC —  9000 BC —  10000 BC —  11000 BC — 

Historic    Late Mississippian  Fort Walton/  Pensacola 

Middle Mississippian  Early Mississippian  Late Woodland 

Weeden Island 

Middle Woodland 

Santa Rosa/  Swift Creek 

Early Woodland 

Deptford 

Gulf Formational 

Elliott’s Point/  Norwood 

Early to Middle Archaic 

Paleoindian/Early Archaic   

PHASE/COMPLEX 

          Four Mile Point    Indian Bayou            Horseshoe Bayou    Lassiter        Okaloosa    Alligator Lake          Elliott’s Point 

 

 

Environmental and Archaeological Contexts 

avoids the issue of the timing of the appearance of fiber‐tempered pottery discussed above  as a factor in the definition of the Gulf Formational.  If the primary reason for using the Gulf  Formational  term  is  to  highlight  the  relationships  between  Late  Archaic  societies  that  inhabited  the  Florida  coastal  panhandle  with  contemporaneous  Late  Archaic  groups,  then  the  Elliott’s  Point  designation  satisfactorily  achieves  this  goal.    Indeed,  Thomas  and  Campbell (1993:528) have stated that since fiber‐tempered ceramics in the region tend to  be  found  in  settings  that  also  contain  Elliott’s  Point  sites,  “we  are  considering  all  Gulf  Formational  sites  to  be  possibly  related  to  Elliott’s  Point”  (see  also  Morehead  et  al.  2011:20).    

The Elliott’s Point Complex  The  peoples  who  lived  in  the  Southeastern  United  States  during  the  Late  Archaic  period  (approximately  5000‐3000  BP,  or  3000‐1000  BC)  exhibited  a  variety  of  cultural  characteristics and organizational systems (Anderson 1996).  Traits such as the production  of  fiber‐tempered  ceramics,  degree  of  residential  mobility,  investment  in  fixed  territories,  and degree of organizational complexity were manifested in different ways among different  groups regardless of intergroup distance (Anderson 1996:157, 166‐167).  Many Late Archaic  cultures  participated  in  long‐distance  interaction  within  the  Poverty  Point  Exchange  Network, a vast yet discontinuous system of exchange centered on the Poverty Point site in  northeastern Louisiana (Figure 2.9) (Gibson 2000). The cultural diversity that existed within  the  Late  Archaic  Southeast  largely  resulted  from  differences  in  physiography,  resource  structure,  and  degree  of  interaction  between  neighboring  social  groups  (Anderson  1996:176).     The  Elliott’s  Point  Complex  is  a  local  manifestation  of  Late  Archaic  cultural  traditions  that  was  centered  around  Choctawhatchee  Bay.    It  was  first  defined  by  Lazarus  (1958)  as  a  geographically  bounded  collective  of  sites  exhibiting  a  combination  of  Late  Archaic  assemblage  characteristics,  including  an  absence  of  pottery,  the  presence  of  stemmed  projectile points, and more importantly, the presence of baked‐clay objects and microlithic  “drills” and “perforators.”    Lazarus  (1958:26)  noted  the  similarities  between Elliott’s  Point  and Poverty Point, where microliths and baked‐clay objects both were utilized, although he  also  noted  differences  between  the  two  in  terms  of  their  baked‐clay‐object  assemblages.   Lazarus named the local complex Elliott’s Point after the location of the sites where he first  identified  these  traits.    Since  its  initial  taxonomic  designation,  the  Elliott’s  Point  Complex  has  come  to  include  a  developmental  culture  sequence  based  on  artifact  assemblage  characteristics  and  presumed  participation  within  the  Poverty  Point  Exchange  Network,  evinced  by  the  material  correlates  of  long‐distance  economic  interaction  such  as  artifacts  fashioned  from  exotic  materials  (Campbell,  Morehead,  et  al.  2009;  Thomas  and  Campbell  1991).  Much of the recent discussion and refinement of the Elliott’s Point Complex is due  to the work of Thomas, Campbell, and colleagues based on CRM‐related projects on Eglin  AFB.       51

Bayou Park, 8OK898 

Figure 2.9.  Location of the Bayou Park site in relation to Poverty Point  and other related sites. 

Chronology  Elliott’s  Point  has  been  subdivided  into  a  three‐sequence  chronology  based  on  temporal  changes in artifact assemblages (Campbell, Morehead, et al. 2009:17; Thomas and Campbell  1993:527‐528).    The  radiocarbon  chronology  and  diagnostic  artifacts  or  traits  are  listed  in  Table  2.2  and  a  list  of  relevant  calibrated  radiocarbon  dates  is  presented  in  Table  2.3.    A  formative or “nascent” expression that began around 2500 BC is represented by sites such  as  8OK898  and  Meig’s  Pasture  (8OK102),  located  on  the  east  side  of  Rocky  Bayou  (Campbell, Morehead, et al. 2009:17).  This early phase of Elliott’s Point is characterized by  smaller shell midden and accretional mound sites, a predominance  of  amorphous  baked‐ clay objects, and minor evidence of trade represented by artifacts such as steatite vessels  (Thomas  and  Campbell  1991).    Radiocarbon  dates  from  8OK898  suggest  the  site  was  occupied from 2535 to 1860 BC, though the majority of ranges fall within 2500 to 2200 BC  (Beta Analytic, Inc. 2002; Meyer et al. 2002:64).  Dates from Meig’s Pasture place the site’s  Elliott’s Point component at around 2425 to 1955 BC (Curren 1987:71‐73).      52 

 

 

Environmental and Archaeological Contexts 

Table 2.2.  Elliott’s Point chronology and diagnostic traits.  Age  Diagnostic Traits 2500 ‐ ~2000 BC  Small shell middens and accretional mound sites; amorphous baked‐clay  objects; evidence of extra‐local trade (steatite).  Classic  ~2000 – 1000 BC  Small campsites and larger “aggregation and redistribution” sites; “well‐ formed” baked‐clay objects, microliths, stemmed Archaic projectile  points; exotic trade items; first appearance of fiber‐tempered pottery  (late)  Terminal  1000 – 600 BC  Fiber‐tempered pottery; shift in shellfish consumption from quahog clam  and scallop to Rangia  SOURCE:  Thomas and Campbell 1993; Campbell et al. 2004. Phase  Nascent 

  A  “classic”  or  fluorescent  stage  of  the  Elliott’s  Point  Complex  is  said  to  have  followed  sometime between 2000 BC and 1100 BC, typified by a greater number of exotic items, an  increase  in  the  number  of  “well‐formed”  baked‐clay  objects  (Thomas  and  Campbell  1993:528),  and  evidence  of  long‐distance  trade  as  well  as  trade  redistribution  through  periodic  population  aggregation  (Thomas  and  Campbell  1993:527‐528).    A  component  of  the  Alligator  Lake  site  (8WL29)  on  Fourmile  Peninsula  is  considered  a  fluorescent‐age  Elliott’s  Point  component,  dating  from  1675  to  1025  BC.    Campbell  et  al.  (2004:137‐139)  also include a second occupation at Meig’s Pasture and the Graveyard Point site (8SR44) as  fluorescent‐age  sites.    The  former  is  said  to  date  to  1700‐1455  cal  BC  while  the  latter,  located  at  East  Bay,  is  reported  as dating  to  1585‐1245  cal  BC.    However,  the  samples  on  which  both  of  these  dates  are  based  are  marine  shell  and  it  does  not  appear  that  any  13 12 C/ C adjustment was made to the measured radiocarbon ages of 3630 ± 50 BP and 3490  ±  BP,  respectively.    When  the  corrected  conventional  radiocarbon  ages  are  used,  the  calibrated  dates  are  2270‐1956  BC  for  Meig’s  Pasture  and  2124‐1729  BC  for  Graveyard  Point.     Several  sites  located  on  Fourmile  Peninsula,  including  the  Buck  Bayou  Mound  (8WL90),  which is  believed  to  be  a  population  aggregation  site,  are  thought  to  have  been  most  heavily utilized during Elliott’s Point fluorescence as well (Thomas, P. M. 1989; Thomas and  Campbell  1991:104‐108,  1993:530‐532).  The  full  range  of  the  Elliott’s  Point  artifact  assemblage  developed  during  this  period,  including  an  increase  in  exotic  materials  presumably obtained through trade (Campbell et al. 2004:137‐140;  Thomas and Campbell  1991:108‐113).  The appearance of fiber‐tempered ceramics also occurred during the latter  part of this phase, although they do not seem to have been widely accepted (Campbell et al.  2004).       The  final  phase  of  the  Elliott’s  Point  Complex  is  characterized  by  a  series  of  events  that  occurred  sometime  after  1000  BC,  including  an  increased  acceptance  of  fiber‐tempered  ceramics,  a  decrease  in  exotic  trade  items,  and  a  shift  in  settlement  patterns  marked  by  greater use of interior, non‐coastal occupations (Campbell et al. 2004:139‐149; Thomas and  Campbell  1993:528,  537‐540).  It  appears  that  this  aspect  of  Elliott’s  Point  lasted  until  around 600 BC, after which the definitive attributes of the complex are replaced by those of  the  Deptford  tradition  (Campbell  et  al.  2004:140;  Thomas  and  Campbell  1993:527‐528).   53

Table 2.3.  Calibrated radiocarbon dates for Elliott’s Point sites.  Site # 

Site Name 

Sample # 

Material 

8WL01278  8WL01278  8WL01278  8WL01278  8OK00898 

Mitchell River 1  Mitchell River 1  Mitchell River 1  Mitchell River 1  Bayou Park 

WK‐9644  WK‐9647  WK‐9651  WK‐9648  Beta‐171507 

shell  shell  shell  charcoal  shell 

8OK00898 

Bayou Park 

Beta‐129298 

shell 

8SR00017  8OK00898 

Aden Bayou  Bayou Park 

Beta‐191432  Beta‐171508 

8WL01005  8OK00102  8OK00102  8OK00102  8OK00102  8OK00102  8SR00017 

NN  Meig’s Pasture  Meig’s Pasture  Meig’s Pasture  Meig’s Pasture  Meig’s Pasture  Aden Bayou 

Beta‐81709  Beta‐21253  DIC‐3294  Beta‐21254  Beta‐21255  DIC‐b  Beta‐203566 

charcoal  shell  soot from  steatite sherd  shell  shell  shell  shell  shell  charcoal 

Mikell and Saunders 2007  Mikell and Saunders 2007 Mikell and Saunders 2007 Mikell and Saunders 2007  Meyer et al. 2002  Morehead, Campbell, and  3790±70  Aubuchon 2008  Thomas, Campbell, and  b  Mathews1994  3770±50  Meyer et al. 2002 

‐1.4  ‐1.8  ‐0.9  ‐25.3  0.0 

4280±50  4190±55  4145±50  3880±50  4210±60 

Calibrated  Years BCa  (Confidence Level)  2581‐2285 (1.0)  2470‐2156 (1.0)  2432‐2121 (1.0)  2473‐2204 (1.0)  2529‐2178 (1.0) 

0.0 

4200±70 

2532‐2135 (1.0) 

b  0.0 

b  4180±60 

2470‐2210b  2468‐2132 (1.0) 

3740±50  3700±80  3690±50  3670±80  3630±90  3630±50  b 

‐25.3  ‐1.0  0.0d  ‐1.0  ‐1.0  0.0d  b 

3735±50  4100±80  4097±50  4070±80  4030±90  4037±50  b 

2291‐2014 (.98)  2437‐1976 (1.0)  2361‐2026 (1.0)  2402‐1933 (1.0)  2383‐1867 (1.0)  2270‐1956 (1.0)  2230‐1870b

0.0  0.0d  ‐25.9  ‐25.0  b  ‐25.0c  ‐25.0c  ‐25.0c  ‐25.0c 

3970±70  3897±70  3520±50  3390±80  3440±40  3135±125  3085±130  2670±150  2575±80 

2228‐1817 (1.0)  2124‐1729 (1.0)  1976‐1737 (.98)  1888‐1502 (1.0)  1881‐1663 (.98)  1688‐1048 (1.0)  1631‐976 (.99)  1134‐406 (.97)  849‐481 (.93) 

Measured 14C  Years BP 

13

Reference 

b  b b b 3800±50 

Yates 2000  Curren 1987  Curren 1987  Curren 1987  Curren 1987  Curren 1987  Campbell, Morehead, et al.  2009  Morehead, Campbell, and  8OK00898  Bayou Park  Beta‐129297  shell  3560±70  Aubuchon 2008  8SR00044  Graveyard Point  Beta‐39718  shell  3490±70  Thomas & Campbell 1993  8WL01278  Mitchell River 1  WK‐9689  charcoal  b Mikell and Saunders 2007  8WL01278  Mitchell River 1  Beta‐139437  charcoal  b Mikell and Saunders 2007  8WL02233  NN  not identified  charcoal  b Campbell, Bourgeois, et al. 2009  8WL00029  Alligator Lake  FSU‐32  charcoal  3135±125  Lazarus 1965  8WL00029  Alligator Lake  FSU‐64  charcoal  3085±130  Lazarus 1965  8WL00035  Four Mile Village  b  charcoal  2670±150  Lazarus 1965  8WL00029  Alligator Lake  GX‐0155  charcoal  2575±80  Lazarus 1965  a  Calibrations performed using CALIB 6.01 and INTCAL09, MARINE09 calibration datasets (Stuiver and Reimer 1993).  b  Data not reported.  c 13 12 C/ C ratio not reported; used standard estimate of  ‐25.0‰ for charcoal (Stuiver and Polach 1977).   d 13 12   C/ C ratio not reported; used standard estimate of 0.0‰ for marine shell (Stuiver and Polach 1977). 

C/12C  o/oo 

Conventional  14 d  C Years BP 

 

 

Environmental and Archaeological Contexts 

Sites  occupied  during  the  Deptford  period  include  Alligator  Lake  (8WL29),  with  an  occupation radiocarbon dated to between 840 and 415 BC (Campbell et al. 2004:137‐139).    The factors contributing to the decline of the Elliott’s Point Complex, and the emergence of  the  subsequent  Deptford  culture,  are  not  well understood.    Morehead  et  al. (2011:24‐25)  suggest  that  it  may  have  been  tied  to  the  roughly  contemporaneous  demise  of  Poverty  Point, which has itself been hypothesized as resulting from massive flooding of the Lower  Mississippi Valley as a result of global cooling and increased precipitation (Kidder 2006; but  see  Kidder  2010  for  a  more  nuanced  interpretation  of  Poverty  Point’s  demise).      An  alternative  explanation,  offered  by  Ken  Sassaman  (1993:225‐227),  is  that  trade  in  raw  materials,  so  important  to  the  rise  of  Poverty  Point,  was  disrupted  by  the  widespread  adoption of ceramic pottery from the east, while Gibson (2010) suggests that the people of  Poverty Point may have returned to a mobile lifestyle in order to facilitate communication  and social networking between related groups.  At present, the question of how and why  Elliott’s  Point  (and  Poverty  Point)  disappeared  from  the  archaeological  record  approximately 2500 years ago remains unresolved.   

Material Culture  According  to  Thomas  and  Campbell  (1991:108‐113;  1993:532‐537),  the  artifacts  that  identify  Elliott’s  Point  archaeologically  include  baked‐clay  objects,  microlithic  tools  (often  referred  to  as  Jaketown  Perforators),  exotic  trade  items  (primarily  steatite),  and  Florida  Archaic  Stemmed  projectile  points,  particularly  a  type  known  as  Destin  (Thomas  and  Campbell  1991:110‐111;  Campbell,  Morehead,  et  al.  2009:17).  Baked‐clay  objects  around  Choctawhatchee  Bay  are  found  in  a  variety  of  sizes  and  may  be  “well‐formed”  or  amorphous  in  shape  and  can  be  plain  or  exhibit  surface  treatments  and  decorations.  Amorphous  objects  are  the  most  numerous  (McGee  1995:104;  Thomas  and  Campbell  1993:533).  Steatite, in the form of stone vessels, pipes, boatstones, and ornaments, is also  common  in  both  early  and  late  Elliott’s  Point  assemblages  (Campbell,  Morehead,  et  al.  2009:17‐18;  Thomas  and  Campbell  1993:534).    Thomas  and  Campbell  (1991:108)  note  an  absence  of  the  lapidary  artifacts  so  common  to  Poverty  Point‐related  assemblages  elsewhere (e.g., hematite plummets); they surmise that the few specimens that have been  recovered,  such  as  a  jasper  gorget  from  8WL87  and  a  jasper  figurine  from  the  shore  of  Choctawhatchee Bay, were obtained through trade.  A “moderately active” bone and shell  industry is also considered a characteristic of Elliott’s Point  material culture, including the  manufacture and use of bone tools as well as bone and shell beads (Thomas and Campbell  1991:108).  Although  microliths  are  common,  not  all  resemble  Jaketown  Perforators,  as  Fairbanks (1959:97) pointed out.     Fiber‐tempered  pottery  of  the  Norwood  series  is  considered  a  late  addition  to  an  already  well‐established  Elliott’s  Point  cultural  tradition  (Campbell  et  al.  2004;  Thomas  and  Campbell  1991,  1993),  and  its  arrival  had  no  observable  impact  on  the  economic  and  technological  characteristics  that  defined  the  Elliott’s  Point  Complex  (Campbell  et  al.  2004:149).    Fiber‐tempered  ceramics  are  estimated  to  have  been  incorporated  sometime  55

Bayou Park, 8OK898 

between  1675  and  1025  BC  based  on  data  from  Alligator  Lake,  though  the  possibility  of  earlier incorporation is acknowledged (Bourgeois et al. 2009:19; Campbell et al. 2004:139).   Campbell, Bourgeois, et al. (2009:96) report a much earlier date of 2500 to 2300 BC from a  feature at 8WL2233 where two sherds of fiber‐tempered pottery also was found; however,  the  sherds  were  not  recovered  from  the  feature  but  from  approximately  40‐50  cm  above  the  top  of  the  feature  (Campbell,  Bourgeois,  et  al.  2009:81,  87).    Thus  the  association  of  fiber‐tempered ceramics with the radiocarbon date is debatable.  There also appears to be  an issue with the reported calibration (cf. Table 2.3). The earliest dates for fiber‐tempered  pottery  then  are  from  Alligator  Lake  and  these  are  at  least  1,500  years  later  than  the  acceptance of pottery elsewhere in Florida and the rest of the region (White 2003a, 2004).     Assuming  that  fiber‐tempered  pottery  did  not  appear  in  the  Choctawhatchee  Bay  region  until relatively late, the logical question is why ceramic technology was resisted here long  after it was accepted elsewhere?  The answer may be related to the socioeconomic forces  influencing Elliott’s Point society. Sassaman (1993) suggests that ceramic vessel technology  diffused  slowly  and  erratically  westward  after  its  introduction  on  the  east  coast.    He  suggests that part of the reason that ceramic technology was so slowly accepted, was that it  undermined  the  value  of  steatite  vessels.  The  groups  and  individuals  who  controlled  the  Late  Archaic  trade  networks  likely  enjoyed  prestige  and  power,  as  well  as  the  ability  to  influence the pace and direction of technological innovation.  A technology that allows for  the  creation  of  vessels  from  local  clay  sources  would  have  rendered  the  steatite  vessels  obsolete,  and  as  steatite  was  likely  a  valuable  market  within  the  trade  network,  the  magnates  who  benefited  most  from  the  movement  of  steatite  may  have  actively  campaigned against the inception of ceramic vessel technology.  Similar examinations of the  socioeconomic  relationships  between  steatite  and  ceramic  vessels  in  Georgia  have  also  been  undertaken  by  Wagoneer  (2009).    Campbell  et  al.  (2004:148)  suggest  that  trade  in  steatite was a significant part of the local Elliott’s Point economy, resulting in an economic  aversion  to  ceramic  vessels  and  its  rather  late  and  limited  acceptance.  Considering  the  commonality  of  steatite  in  Elliott’s  Point  assemblages  (Thomas  and  Campbell  1993:534),  this interpretation seems valid if untested.   

Settlement Patterns  Most  Elliott’s  Point  sites  are  located  in  coastal  settings,  particularly  around  bays,  with  a  minor  occurrence  of  sites  beyond  1.5  km  from  the  bay  (Thomas  and  Campbell  1991:104,  1993:528).    Calvin  Jones  (1993)  documented  clusters  of  Elliott’s  Point  sites  around  Choctawhatchee,  Apalachicola,  St.  Andrews,  Apalachee,  and  Pensacola  bays,  with  Choctawhatchee  Bay  having  the  greatest  number  of  sites.    Sites  there  tend  to  cluster  around  virtually  every bayou  on  all  sides  of  the  bay  and  around  specific  locations  such  as  East Bay, Cinco Bayou and the Narrows, Boggy and Rocky Bayous, Fourmile Peninsula, and  Basin Bayou (Figure 2.10).  Of the interior sites that have been recorded, favored locations  are on the terraces and ridges overlooking rivers or at the heads of tributary seeps (Thomas  and Campbell 1993:256).  These distributions suggest a heavy reliance on ecotonal coastal  locations  with  potable  water  that  primarily  occur  along  the  shores  of  the  bay  proper  and  along the bayous and creeks that drain into the bay.  56 

 

Figure 2.10.  Distribution of Late Archaic sites in the Choctawhatchee Bay region.  Data from Florida Master Site file (2012) and Campbell et al. (2004).  Base Map: SAIC (2010) 

  Environmental and Archaeological Contexts 

   

57

Bayou Park, 8OK898 

The types of Elliott’s Point sites that are present around Choctawhatchee Bay are not well  understood.  Among those that have been identified are smaller sites referred to as stations   (e.g.,  8OK2694,  Mathews  et  al.  2009),  exploitation  sites  (e.g.,  8WL173,  Campbell  et  al.  2007),  or  resource  extraction  camps  (e.g.,  8WL2270,  Mallory  and  Campbell  2008).    These  smaller sites are typically located near interior drainages and have no associated middens or  features.    Lithic  workshops  and  specialty  craft  sites,  defined  by  a  concentration  of  lithic  tools, debitage, and microdrills, also have been identified (Thomas and Campbell 1993:531‐ 532).    Other  sites  include  occupational  loci  referred  to  as  encampments  (e.g.,  8WL36,  Thomas  et  al.  2001),  seasonal  camps  (e.g.,  8OK427,  Campbell  et  al.  2008),  or  short‐term  camps (e.g., 8WL2126, Mallory and Campbell 2006).  These sites typically have middens and  features and are often located in clusters, perhaps to take advantage of nearby aggregation  or “redistribution” centers (Thomas and Campbell 1993:528).  Shell middens/mounds that  represent  more  extensive  occupation  sites,  regional  centers,  or  potential  population  aggregation  sites  are  also  present.  These  sites  are  usually  characterized  by  circular  or  horseshoe‐shaped (arcuate) shell deposits.  Alligator Lake (8WL29) is thought to be one of  these  prominent  habitation  sites,  while  Buck  Bayou  Mound  (8WL90)  may  potentially  represent  an  arcuate  feasting  monument  associated  with  trade  redistribution  (Thomas  1989;  Thomas  and  Campbell  1993:529‐532).    The  Buck  Bayou  Mound  is  unique  among  Elliott’s Point sites in part because of its size, nearly two meters high with a basal dimension  of 125 by 110 m (Thomas 1989:4).  While there is no evidence that the mound was used for  human interments, it contains exotic materials such as steatite and quartz tools, as well as  bone tools, cut shell beads, and fragments of baked‐clay objects.      Meig’s Pasture (8OK102), an arcuate shell midden, is interpreted as having been a seasonal  camp rather than a redistribution center due to the lack of “exotic” artifacts and its small  size  (Thomas  and  Campbell  1993:528‐529).    However,  Campbell  et  al.  (2009a:19)  acknowledge  the  possibility  that  Meig’s  Pasture  may  be  a  predecessor  of  Buck  Bayou  Mound for early Elliott’s Point population aggregation.      Thomas  and  Campbell  (1991:104‐105,  1993:528)  assert  that  Elliott’s  Point  settlement  patterns  are  very  similar  to  those  of  Poverty  Point‐related  sites  in  the  Mississippi  Valley,  where smaller occupation and specialty sites tend to cluster around larger sites thought to  have been regional centers of distribution and influence.  This pattern is represented most  directly at Fourmile Peninsula, where occupation sites (e.g., 8WL36) and specialty lithic sites  (e.g.,  8WL192)  are  located  in  close  proximity  to  the  Buck  Bayou  Mound.    However,  the  majority of Elliott’s Point sites identified around Choctawhatchee Bay have been identified  as camps rather than “extensive” habitations (Thomas and Campbell 1993:528).      It  appears  that  a  shift  in  settlement  systems  may  have  occurred  toward  the  end  of  the  Elliott’s  Point  Complex  that  coincided  with  both  a  greater  utilization  of  fiber‐tempered  ceramics and a drop in the salinity levels of Choctawhatchee Bay (Campbell et al. 2004:140;  Thomas  and  Campbell  1991:114‐115,  1993:539‐549).    Fiber‐tempered‐using  groups,  thought to have become prominent by at least 1100 BC (Thomas and Campbell 1993:527‐ 528), exhibited new settlement preferences for the shores of brackish water bayous and for  58 

 

 

Environmental and Archaeological Contexts 

interior  occupations  along  transportation  routes  following  major  and  minor  streams  (Campbell  et  al.  2004:140).    In  addition,  sometime  around 1000  BC,  the  slow  accretion  of  the barrier island forming Moreno Point likely restricted water exchange between the Gulf  and the bay.  As discussed above, this event may have resulted in lower salinity levels and a  reduction in suitable habitat for quahog and bay scallop, which formed a significant part of  the Elliott’s Point diet.  This desalination of the bay may have contributed to the settlement  shifts noted for this time and perhaps contributed to the onset of the succeeding Deptford  period (Campbell et al. 2004).    

Subsistence and Seasonality  Although  the  triad  of  oyster,  quahog,  and  bay  scallop  are  sometimes  referred  to  as  diagnostic  of  Elliott’s  Point  subsistence  (e.g.,  Thomas  and  Campbell  1993:541),  there  are  few  detailed  data  on  subsistence  and  seasonality  practices.    Elliott’s  Point  subsistence  discussions are typically limited to gross shellfish proportions and descriptions of vertebrate  fauna recovered from large‐mesh (1/4‐inch) screens (e.g., Campbell et al. 1998; Meyer et al.  2002;  Thomas  1989;  Thomas,  Campbell,  Mathews,  and  Morehead  1994).    An  exception  is  Meig’s Pasture where a large array of invertebrate and vertebrate taxa was recovered using  small‐mesh screens.  The analysis of these samples indicates extensive use of aquatic faunal  resources (Russo 1987).  In addition to bay scallop, quahog clam, and oyster the remains of  Rangia, marsh periwinkle, ray, shad, sea catfish, jack, white‐tailed deer, turtle, rabbit, and  river  otter  were  also  recovered.    Another  well‐documented  Late  Archaic  site  is  Mitchell  River  1  (8WL1278)  on  the  east  side  of  Choctawhatchee  Bay  (Mikell  and  Saunders  2007).   This site contains an Elliott’s Point component; however, the faunal data for all of the site’s  Late Archaic components were collapsed into a single table so it is difficult to characterize  Elliott’s  Point  subsistence  on  its  own.    It  is  worth  noting,  however,  that  a  variety  of  mammals, reptiles, and a few freshwater fish were recovered from excavations at the site,  in addition to marine fish and shellfish remains (Mikell and Saunders 2007:Table 6).    Bayou Park has been interpreted as having been an early Elliott’s Point seasonal camp with  evidence  of  varied  activities  (Campbell,  Morehead,  et  al.  2009:60).    Campbell  et  al.  (1998:70) conclude that, typical of early Elliott’s Point sites, the dominant shellfish species  are oyster, quahog clam, and bay scallop.  Feature 3 is recorded as having been dominated  by oyster and quahog with “appreciable” amounts of conch and only a minor incidence of  bay scallop (Campbell et al. 1998:70).  The midden at 8OK898 consisted of primarily oyster,  bay scallop, and quahog with some Rangia, Strombus, and moon snail.  Limited vertebrate  remains  indicate  that  shellfish  was  augmented  with  opossum  and  deer,  fish  (all  were  unidentified),  turtle,  and  bird;  however,  the  vertebrate  fauna  is  biased  towards  larger  elements since the sample was derived from large‐mesh recovery screens.  Consequently,  fish remains are underrepresented.     Seasonality studies at Elliott’s Point sites are limited.  The most detailed studies have been  conducted  for  the  Meig’s  Pasture  site.    Claasen  (1987)  analyzed  shellfish  size  and  internal  shellfish  growth  ring  data  at  Meig’s  Pasture,  but  the  results  of  the  analyses  were  59

Bayou Park, 8OK898 

contradictory (Claassen 1987).  The size of analyzed specimens suggested collection during  warm  months,  while  internal  growth  ring  data  implied  collection  during  periods  of  cooler  water  temperatures.    Claassen  (1987:43)  suggested  that,  given  differences  in  the  sample  sizes,  more  validity  should  be  given  to  the  internal  growth  ring  data,  which  suggest  cool  weather  occupation.    Hickory  nut  shells  were  recovered  in  abundance,  and  a  single  grape  seed and a cabbage palm seed were also recovered (Newsom 1987:45‐49).  It is likely that  the hickory nuts were utilized for subsistence, but the grape and cabbage palm occurred in  such low numbers that it is difficult to arrive at any conclusion regarding their use as food.   The presence of hickory mast further suggests a fall or winter occupation.     Russo  (1987:56‐57)  observed  that  most  of  the  fish  species  in  the  sample  are  found  in   shallow‐water  estuarine  environments  and,  in  general,  are  most  common  during  warm  months.    Based  on  these  data  (of  which  the  sample  size  was  much  larger  than  the  plant  remains),  he  concludes  that  Meig’s  Pasture  was  occupied  during  a  warm‐weather  season  (Russo 1987:57).  In retrospect, however, the shellfish, botanical, and faunal data seem to  contradict one another in terms of a single season of occupation.  The most parsimonious  explanation  is  that  Meig’s  Pasture  was  occupied  during  several  seasons  of  the  year,  although a more thorough seasonality study is probably warranted.     Hickory  nut  shells  were recovered  from  Bayou  Park  during  PTA’s  Phase  III  investigation  in  1998,  indicating  that  the  inhabitants  were  engaged  in  gathering  activities  at  least  during  portions of the fall and perhaps winter (Campbell et al. 1998:75).  This may suggest that the  site was occupied either during fall or winter months, or that hickory mast was stored and  processed at 8OK898 out of the season of harvest.     Additional  information  on  Elliott’s  Point  seasonality  can  be  inferred  from  the  Buck  Bayou  Mound  (Thomas  1989;  Thomas  and  Campbell  1993).    The  depositional  structure  of  the  mound  strata  suggest  that  the  site  was  not  occupied  year‐round,  but  rather  represents  occasional,  high  intensity  visits  punctuated  by  periods  of  abandonment  (Thomas  and  Campbell 1993:531).  In addition, the presence of bay scallops is interpreted as representing  a  summer  occupation,  as  bay  scallops  enter  the  shallow  waters  of  Choctawhatchee  Bay  primarily during the warm months, though Thomas and Campbell (1993:531) admit that this  evidence  is  not  conclusive.    However,  if  Buck  Bayou  Mound  represents  a  periodic  aggregation  site,  as  they  claim  (Thomas  and  Campbell  1993:529‐531),  rather  than  a  seasonal  camp  or  village,  then  these  data  may  reflect  seasonal  patterns  of  population  aggregation  rather  than  the  seasonal  settlement  movements  of  one  or  a  few  groups.   Indirect evidence of seasonal movements by Elliott’s Point people has also been recorded at  8WL1005  on  Alaqua  Creek,  a  watercourse  that  feeds  into  Choctawhatchee  Bay,  where  a  cache  containing  two  steatite  bowls  was  found  (Hemphill  et  al.  1995).    The  caching  of  heavier  site  furniture  may  denote  the  desire  to  travel  unencumbered,  as  well  as  the  intention of returning to the cache area to retrieve the items later.    In  sum,  very  little  of  a  detailed  nature  is  known  about  Elliott’s  Point  subsistence  around  Choctawhatchee  Bay,  and  by  extension,  about  the  scheduling  of  Late  Archaic  seasonal  60 

 

 

Environmental and Archaeological Contexts 

movements.    The  majority  of  sites  around  the  bay  have  been  identified  as  seasonal  “camps,” and subsistence appears to have been centered on marine invertebrate species.   However,  this  latter  interpretation  is  based  on  limited  samples  collected  with  large‐mesh  screens and do not adequately reflect the full range of maritime subsistence, specifically the  importance  of  fish,  whose  small  bones  escape  detection  without  the  use  of  small‐mesh  screens.  Where detailed analyses have been undertaken it is apparent that shellfish were  augmented  with  a  variety  of  other  food  sources,  including  fish  and  several  terrestrial  species.     

Trade in the Choctawhatchee Bay Region  From  Choctawhatchee  Bay,  Elliott’s  Point  populations  would  have  had  access  to  both  bountiful  resources  and  avenues  of  transportation  to  distant  groups  and  their  resources.  The evidence usually cited for participation in long‐distance exchange includes the presence  of  items  fashioned  from  foreign  materials  such  as  exotic  cherts,  steatite,  copper,  jasper,  galena,  hematite,  and  even  pumice  (Campbell,  Morehead,  et  al.  2009:18‐20;  Thomas  and  Campbell 1991:112‐113, 1993:531‐533).  Fragments of steatite vessels have been found at  the  Buck  Bayou  Mound,  Horseshoe  Bayou  (8WL36),  Pirate’s  Bay  (8OK183)  and  caches  of  steatite    vessels  have  been  recovered  from  8WL87  and  8WL1005  (Gagliano  and  Webb  1970:71; Thomas and Campbell 1993:534). Diagnostic artifacts common in the assemblages  of  contemporary  groups  throughout  the  region,  such  as  baked‐clay  objects,  are  also  abundant  in  many  Elliott’s  Point  sites,  presumably  connecting  local  populations  to  groups  much further away.     Other lines of evidence include similarities in projectile point styles.  Seven projectile points  with straight‐sided stems were recovered from the Alligator Lake site and are described as  Delhi points, a style that was popular at Poverty Point (Lazarus 1965:95, 98).  Delhi points  are  corner‐notched  with  drooping  barbs  (Ford  and  Webb  1956:59‐60).    Only  five  of  the  Alligator Lake specimens are pictured but only two resemble the type description for Delhi  points.  The other three points display straight basal blade corners.  Lazarus also describes  two basally notched points as Marshall points, another projectile point type that was found  at  Poverty  Point.    One  of  the  Alligator  Lake  specimens  reportedly  is  made  of  clear  quartz  crystal (Lazarus 1965:98).      Gagliano  and  Webb  (1970)  state  that  aspects  of  the  Buck  Bayou  Mound  assemblage  compare closely with items found at the adjacent Cedarland and Claiborne sites, located at  the mouth of the Pearl River on the Mississippi Gulf Coast (see Figure 2.9).  Both of these  sites  are  believed  to  have  been  directly  associated  with  the  Poverty  Point  Exchange  Network  (Gagliano  and  Webb  1970;  Russo  1996).    A  cache  of  10  steatite  vessels  was  recovered  from  the  Claiborne  site  (Gagliano  and  Webb  (1970:58‐59)  and  projectile  points  similar to the Florida Archaic Stemmed Levy subtype were found at both sites (Gagliano and  Webb 1970:59).  The construction of arcuate shell middens at Elliott’s Point sites is similar  to the arcuate middens at Cedarland and Claiborne.    61

Bayou Park, 8OK898 

Connections with Poverty Point‐related sites presumably occurred primarily along the coast.   In this interpretation, Alligator Lake was the initial contact point for long‐distance Poverty  Point  traders  because  of  its  proximity  to  the  Gulf  of  Mexico  and  inland‐reaching  watercourses.  The Buck Bayou Mound on Fourmile Peninsula served as an aggregation site  for  Elliott’s  Point  villages  in  the  surrounding  area  (Campbell  et  al.  2009a:19;  Thomas  and  Campbell  1993:532),  facilitating  large  feasting  events  and  inter‐group  trading  that  ultimately resulted in the redistribution of goods as well as the accretion of a mound that  was likely recognized as a monument to fellowship and the prestige of the hosting parties  (Russo 2004).  This interpretation is based partly on the stratigraphic interpretation of the  Buck  Bayou  Mound  itself,  which  is  interpreted  as  representing  large,  episodic  feasting  events punctuated by periods of abandonment (Thomas and Campbell 1993:532).    With  the  “decline”  of  the  Elliott’s  Point  Complex,  involvement  in  long‐distance  exchange  appears  to  have  been  reduced  significantly.    The  late  acceptance  of  fiber‐tempered  ceramics  coupled  with  a  decrease  in  exotic  items  suggests  both  a  reduction  in  exchange  activities and a waning of the influences of the trading elite that favored steatite.  Indeed,  the  Deptford  culture  that  followed  Elliott’s  Point  exhibits  few  signs  of  long‐distance  interaction and has been described as “conservative” in comparison to the cultures of the  Late  Archaic  period  (Campbell  et  al.  2004:149).    Further,  it  may  be  no  coincidence  that  a  reduction in Elliott’s Point trade and the influx of ceramic vessels roughly coincide with the  drop  in  bay  salinity  that  resulted  in  changes  in  shellfish  species  composition  and  possible  shifts  in  settlement  patterns  (Campbell  et  al.  2004:148‐149).    Although  there  are  several  scenarios that could explain the correlations between the acceptance of ceramics, changes  in  subsistence  availability,  shifts  in  settlement  patterns,  and  shifts  in  the  local  power  structure, the manner in which these events are connected remains uncertain.   

Mounds, Ceremonialism, and Social Interaction  Ceremonialism,  mortuary  practices,  and  mound  function  are  poorly  understood  in  the  Choctawhatchee  Bay  area  as  they  are  across  the  Late  Archaic  Southeast  (Gibson  2000;  Thomas and Campbell 1993).  Elliott’s Point mounds are nearly exclusively shell and are only  found in coastal settings.  These include Buck Bayou Mound, Graveyard Point, and Meig’s  Pasture.  Russo  (2004:40)  suggests  that  shell  found  in  piles,  lenses,  and  other  features  associated  with  shell  mounds  represent  the  epiphenomena  of  interactions,  particularly  of  feasting  events.    He  argues  that  with  its  open  plaza  and  raised  circular  boundary,  the  arcuate  shell  structure  provided  a  public  forum  for  all  aspects  of  food  preparation,  distribution,  and  disposal  (Russo  2004:41).  These  public  rituals  facilitated  formal  social  interactions  and  affirmed  social  connections.  The  structured,  conspicuous  disposal  of  the  shell produced by these events served as the ceremonial seat of rituals associated with the  feast activities and memorialized the host’s generosity and prestige.  Thus, it is possible to  interpret  Elliott’s  Point  shell  mounds  as  representing  feasting  monuments  resulting  from  population  aggregation  whereby  ritual  gathering  was  intended  to  reaffirm  economic  and  social  ties  while  simultaneously  redistributing  wealth  obtained  through  exchange  interactions within the Poverty Point Exchange Network.     62 

 

 

Environmental and Archaeological Contexts 

The  identification  of  Meig’s  Pasture  as  an  intentionally  constructed  or  accretional  mound  has been questioned, however, by Caleb Curren who conducted excavations at this site in  1987  (Curren  1987:11‐15).    On  the  basis  of  several  trenches  excavated  through  the  “mound”  Curren  concluded  that  the  site  consists  of  numerous  shell  deposits  and  pit  features that were situated on a low sandy rise surrounding a shallow depression that may  have  held  water  at  one  time  (Curren  1987:78).    His  trench  profiles  clearly  show  the  discontinuous  and  overlapping  nature  of  the  deposits  and  features,  and  support  his  interpretation that is not an accretional midden or intentionally constructed tumulus, but is  the result of periodic refuse disposal events.  This does not invalidate the claim that the site  may have served as a locus for feasting, however, nor does it contradict the suggestion of  Thomas  and  Campbell  (1993:528‐530)  that  Meig’s  Pasture  may  have  been  a  precursor  of   the  accretional  mound  building  and  population  aggregation  events  that  culminated  in  the  Buck  Bayou  Mound.    Unfortunately,  there  are  no  radiocarbon  dates  from  Buck  Bayou  Mound or the surrounding midden with which to establish temporal relationships between  the two sites.    The only sand mound that is of potentially associated with Elliott’s Point is 8WL10 (known  as Mound near Mack Bayou), located on the south shore of Choctawhatchee Bay just east  of  Fourmile  Peninsula.    Moore  (1918:541)  reported  a  stone  pendant  in  association  with  a  soil discoloration that may have been a badly decomposed burial.   An “arrowhead or knife,  of  flint,”  a  sandstone  hone,  and  a  “bar  amulet”  were  also  recovered  from  the  mound.    Willey (1949:221) suggested the possibility that 8WL10 may have been a preceramic burial  mound but the evidence was inconclusive. Thomas and Campbell (1993:529) conclude that,  as mound burial is not a definitive attribute of Poverty Point‐related cultures, the mound’s  relationship to the Elliott’s Point Complex is questionable.  Recent research, however, has  revealed that there is a great deal of variability in the burial practices of contemporary Late  Archaic populations, including instances of mound burial (Russo 1994).  Additionally, sand  mounds have been associated with Late Archaic populations across the region (Peacock et  al.  2010)  and  8WL10’s  association  with  the  Elliott’s  Point  Complex  may  need  to  be  reconsidered.       At Mitchell River 1 (8WL1278) near the mouth of the Choctawhatchee River, 10 burials were  discovered in two locations on the western end of a low (ca. 40 cm) rise that may be the  remains of a small mound (Mikell and Saunders 2007:178‐179).  The burials were interred in  shell‐filled pits or shell midden deposits.  The pits intruded into the underlying sand with the  tops of the burials varying from 62 to 92 cmbs.  Two radiocarbon dates were obtained for  the  burial  area:  4840‐4570  cal  BP  from  Level  8  (70‐80  cmbs)  and  5900‐5660  cal  BP  from  Level 9 (80‐90 cmbs).  Since the burials are intrusive into earlier deposits, at least some of  the shallower burials post‐date this time frame.  Other portions of the site have been dated  to the Elliott’s Point period (see Table 2.3) and amorphous baked‐clay objects and a steatite  vessel rim sherd also were recovered.   

63

Bayou Park, 8OK898 

With  the  possible  exceptions  of  8WL10  and  8WL1278,  no  definitive  Elliott’s  Point  burials  have  been discovered  to  date.    Considering  the  level  of  work  that  has  been  conducted  at  Elliott’s  Point  sites,  the  absence  of  burial  contexts  is  potentially  significant  in  terms  of  associations  with  contemporaneous  groups.  The  absence  of  a  shared  mortuary  complex  may  denote  a  lack  of  ideological  associations  between  Elliott’s  Point  populations  and  contemporaneous  Late  Archaic  groups,  or  at  least  the  lack  of  evidence  for  such  a  connection.   

Post‐Archaic in the Panhandle   

Woodland Period  The  widespread  use  of  ceramics,  the  interment  of  the  dead  in  mounds,  and  increased  regionalization characterize the Woodland period, which began at around 2600 BP (600 BC)  in the Choctawhatchee Bay region and continued until about 1000 BP (AD 1000) (Campbell  et al. 2004:140; Thomas et al. 1996).     The first archaeological manifestation that is recognized as a Woodland culture is Deptford.   Stephenson  et  al.  (2002:325‐333)  define  three  regional  variations  in  the  Southeast:  Gulf,  Atlantic, and Coastal Plain‐Interior Riverine; Florida’s panhandle falls within the Gulf region.   Florida Deptford is described as primarily a “coastal dwelling culture” which relied heavily  on  maritime  subsistence  strategies  (Milanich  and  Fairbanks  1980:66).    Deptford  villages  located  on  the  coast  are  usually  found  in  conjunction  with  live  oak,  magnolia,  and  palm  hammocks  located  near  salt  marshes.  Interior  Deptford  is  found  along  lakes  and  streams  where  hickory  and  oak  are  present.    These  are  believed  to  represent  seasonal  camps,  presumably occupied during the late fall and winter.  Deptford sites contain plain pottery or  ceramics  with  checked  patterns  stamped  on  the  exterior  of  the  pot,  and  many  have  distinctive podal supports (Milanich 1994).    Temporally,  two  phases  have  been  identified  in  the  Choctawhatchee  Bay  region:  Alligator  Lake (ca. 600‐50 BC) and Okaloosa  (50 BC‐AD 150) (Thomas and Campbell (1993:547‐554;  Thomas et al. 1996).  A possible third phase was identified at the Early Fish Fry site (8OK126)  overlying Alligator Lake phase deposits and producing radiocarbon dates of 330 and 320 BC  (Thomas and Campbell 1993:257).      Excavations at Pirate’s Bay (8OK183) have produced important information about Deptford  subsistence and settlement (Thomas et al. 1991).  This coastal site has produced evidence  that Deptford people exploited a wide range of local marine and terrestrial food resources.  In  addition,  trade  items  from  the  Lower  Mississippi  Valley  and  southern  Georgia  were  recovered,  providing  evidence  of  participation  in  a  far‐flung  exchange  network  with  neighboring  cultures.    The  lithic  assemblage  is  distinguished  by  the  presence  of  small,  backed white quartz pebbles that appear to have been specialized tools.     

64 

 

 

Environmental and Archaeological Contexts 

An important component of Deptford culture in northwest Florida is a mortuary‐ceremonial  complex  referred  to  as  Yent  and  originally  defined  by  Sears  (1962).    The  Yent  Complex  included  the  inclusion  of  exotic  goods  obtained  through  exchange  in  mortuary  contexts.   These exotic items, which included galena, mica, artifacts made of metamorphic rock, and  similarities in ceramic vessel design, may have come to Florida via exchange with Hopewell  cultures farther north (Milanich 1994).  The Yent Complex appears strongest in the Big Bend  region of northwest Florida.  Farther west, burial mounds and characteristic Yent Complex  artifacts are rare (Milanich 1994).  Instead, Deptford people disposed of their dead in graves  within or adjacent to their villages (Thomas and Campbell 1993).      The  Santa  Rosa‐Swift  Creek  culture  replaced  Deptford  throughout  northwest  Florida,  beginning  about  1850  BP  (AD  150).    It  is  marked  by  two  ceramic  series:  Swift  Creek  and  Santa Rosa.  It seems that Swift Creek ceramic designs originated in southern Georgia and  were  subsequently  adopted  by  Deptford  people  in  Florida.    Swift  Creek  pottery  exhibits  complicated stamped designs consisting of scrolls, concentric circles, teardrops, and spirals.   Check  stamping  also  was  used  by  Swift  Creek  potters.    Swift  Creek  vessel  forms  include  squat bowls and deep cylindrical pots. The replacement of Deptford ceramics by Swift Creek  in  the  Florida  panhandle  took  place  over  several  centuries.    Santa  Rosa  ceramics  contain  incised, punctated, and rocker‐stamped designs and appear to be a continuation of ceramic  traditions that originated in the Lower Mississippi Valley.      Santa  Rosa‐Swift  Creek  villages  were  located  on  the  coast  and  in  the  interior  forests  and  river valleys throughout the panhandle.  Excavations at Horseshoe Bayou (8WL36) and the  Old  Homestead  site  (8WL58)  have  provided  information  about  Santa  Rosa‐Swift  Creek  subsistence, settlement, and socio‐political and religious organization (Thomas et al. 1996,  1998,  2001).    8WL58  consists  of  a  circular  shell  midden  surrounding  a  plaza.    Rangia  accounts for 99% of the shell from the site.  The plaza contains no shell midden, but a dark  earth  midden  was  identified  and  excavation  revealed  numerous  postmolds. At  8WL36  a  horseshoe‐shaped shell midden composed of Rangia is present.  Analysis of faunal remains  indicated  that  a  wide  range  of  local  marine  and  terrestrial  food  resources  were  exploited  (DeFrance 2001)    Based on their work at these sites, Thomas et al. (1996) defined two phases for Santa Rosa‐ Swift  Creek  in  the  Choctawhatchee  Bay  region:  Lassiter  (ca.  AD  150‐450)  and  Horseshoe  Bayou (ca. AD 450‐650).  The former is characterized by Swift Creek Complicated Stamped  pottery,  St.  Andrews  Complicated  Stamped,  West  Florida  Cord  Marked,  Crooked  River  Complicated Stamped, Alligator Bayou Stamped, Santa Rosa Stamped, Basin Bayou Incised,  Gulf Check stamped, and Franklin Plain.  The Horseshoe Bayou phase is distinguished by a  distinctive  ceramic  type  –  Horseshoe  Bayou  Complicated  Stamped  –  that  exhibits  a  bold  check stamp with a raised dot in the center of the check.    The  socio‐religious  aspect  of  this  culture  has  been  defined  as  the  Green  Point  complex  (Sears  1962),  which  had  associations  with  the  Hopewell  interaction  sphere  and  may  have  65

Bayou Park, 8OK898 

developed out of the preceding Yent Complex.  Through this exchange network, Santa Rosa‐ Swift  Creek  people  gained  access  to  exotic  items,  such  as  copper,  mica,  ear  spools,  and  ceramics (Bense 1998).     The  emergence  of  Weeden  Island  cultural  attributes  in  the  Choctawhatchee  Bay  region  began  at  about  1350  B.P.  (AD  650).    Weeden  Island  has  been  defined  as  a  religious‐ ceremonial  complex  that  was  adopted  by  local  regional  cultures  in  southern  Georgia  and  Alabama  and  along  the  west  coast  of  Florida  (Milanich  1994).    Early  Weeden  Island  is  characterized  by  the  appearance  of  complicated  stamped  pottery  along  with  the  characteristic  Weeden  Island  pottery  decorated  with  incised  and  punctated  lines  (e.g.,  Carrabelle Incised, Carrabelle Punctated, Keith Incised, and Weeden Island Incised).  There  appears  to  be  some  continuity  between  Santa  Rosa‐Swift  Creek  and  Weeden  Island  occupations.  Not  only  are  both  cultural  expressions  found  in  the  same  coastal  environmental settings, but these cultures exploited similar marine resources.  Fish remains  include  herring,  saltwater  catfish,  sea  catfish,  jack,  porgies,  sheepshead,  mullet,  flounder,  bowfin,  drum,  and  gar.    Shell  middens  indicate  a  preference  for  oysters,  although  conch,  Rangia  and  other  species  also  are  present.    Vertebrate  faunal  remains  include  white  tail  deer, freshwater turtle, and birds.  Acorns and hickory nuts were collected as were various  plant species, such as yaupon, wild grape, palmetto shoots, and gallberry.      Mortuary  ceremonialism  reached  its  peak  during  early  Weeden  Island  times.    Ornately  decorated ceramics and those shaped as stylized designs or animal effigies were interred in  burial  mounds,  often  on  the  mounds’  east  side  (Milanich  and  Fairbanks  1980).    Early  Weeden Island villages also appear to have been arranged in circular patterns as evidence  by several “ring” sites that were identified on the Tyndall Air Force Base (Russo et al. 2009).      Late Weeden Island (AD 750‐1000) is identified by the presence of check‐stamped and cob‐ marked pottery, and is referred to as Wakulla Weeden Island by Milanich (1994:194‐204).    Wakulla  sites  are  located  on  the  coast  and  in  the  interior  of  the  panhandle,  as  well  as  in  southwest  Georgia  and  southeast  Alabama.    According  to  Milanich  (1994:194),  maize  agriculture was adopted in the panhandle during the Wakulla period; however, no evidence  of maize has been documented for the Choctawhatchee Bay region.  Interior Weeden Island  sites in this region tend to be smaller and distributed around springheads, a trend suggested  by Milanich and Fairbanks (1980) as distinctive of the culture.       

Mississippi Period  For this report, the term “Mississippi” is used for the chronological period beginning about  1000 years ago (ca. AD 1000) and continuing until about 450 BP (AD 1550), while the term  “Mississippian”  is  reserved  for  the  broader  late  prehistoric/protohistoric  cultural  tradition  that  emanated  from  the  Mississippi  Bottoms  and  influenced  regional  cultures  such  as  Pensacola  and  Ft.  Walton.    These  influences  included  maize  agriculture,  shell‐tempered  pottery,  institutionalized  social  inequality,  a  chiefdom  level  of  political  organization,  and  participation  in  long‐distance  exchange  relations  that  involved  the  movement  of  exotic  items and religious iconography throughout the Southeast (Ashley and White 2012).  66 

 

 

Environmental and Archaeological Contexts 

The Fort Walton culture was centered in the Red Hills surrounding modern Tallahassee and  extended west to Choctawhatchee Bay.  It was only in the Red Hills, however, that most of  the  trappings  of  the  wider  Mississippian  phenomenon,  including  maize  agriculture,  were  adopted.      Farther  west  around  Pensacola  Bay,  the  Pensacola  variant  of  the  Fort  Walton  culture evolved in place out of the preceding Weeden Island culture and was influenced by  cultural developments in Alabama (Harris 2012; Milanich 1994).  The Choctawhatchee Bay  region  is  interesting  because  ceramic  series  identified  with  both  of  these  variants  are  present.  The Fort Walton series is characterized by grit‐ and/or sand‐tempered pottery with  distinctive  incised  and  punctated  surface  designs  as  well  as  undecorated  vessels  (Willey  1949:452‐488).    The  Pensacola  series  is  distinguished  from  the  Fort  Walton  series  by  the  presence of shell tempering associated with both decorated and plain vessels (Willey 1949).      Two  Mississippi‐period  phases  have  been  identified  for  the  Choctawhatchee  Bay  region  based  on  the  increasing  frequencies  of  Pensacola  series  pottery  in  Late  Fort  Walton  sites  (Mikell  1990,  1992).    The  Indian  Bayou  phase  (ca.  AD  1000‐1350)  is  dominated  by  Fort  Walton  series  pottery  with  small  frequencies  of  Pensacola  series  sherds.    The  Four  Mile  Point phase (ca. AD 1350‐1600) is characterized by relative frequencies of Pensacola pottery  that range from between 30 and 40 percent to as much as 70 percent of the collections.      Mississippi‐period villages are located almost exclusively in coastal locations.  Few interior  village sites are known and most of these are along the Yellow River, at the headwaters of  south‐flowing tributaries, or on creeks at settings inland from Choctawhatchee Bay (Thomas  et al. 1996).  Major villages were likely occupied year‐round by at least limited populations,  while  the  smaller  hunting,  gathering,  and  horticultural  loci  were  occupied  seasonally  by  small groups.      The principal village pattern is represented by 8WL51, an off‐Eglin site on the west side of  Hogtown Bayou, which Lazarus (1971:45) describes as “... six or seven small midden piles of  shell  ...  arranged  in  a  pattern.”      Smaller  Mississippian  coastal  sites  were  less  intensively  utilized  and  non‐nucleated,  and  may  represent  dispersed  households  and  resource  exploitation or special function campsites.      Although there were fewer mounds than during the preceding Weeden Island period, there  is clear evidence of ceremonialism in Fort Walton/Pensacola culture.  Thomas et al. (1996)  indicate  that  at  least  six  Mississippian  mounds  exist  in  the  Choctawhatchee  Bay  area  and  that  these  contain  a  variety  of  Fort  Walton  and  Pensacola  ceramics.    The  Fort  Walton  Temple  Mound  (8OK6)  is  the  most  impressive  of  these.    This  large  platform  mound  measures 12 ft in height, 223 ft by 220 ft at its base, 90 ft by 150 ft at its summit.  Over 80  burials  are  reported  to  have  been  interred  in  the  mound  and  it  may  have  served  as  a  regional center of Fort Walton/Pensacola activity.  The site has been the subject of several  investigations (Fairbanks 1965; Lazarus and Fornaro 1975) that have produced evidence of  multiple burials, shell and bone tools, shellfish, and vertebrate fauna, lithics, and mica.     67

Bayou Park, 8OK898 

CONCLUSION    In  this  chapter,  we  have  attempted  to  summarize  what  is  known  about  the  environment  and archaeology of the Choctawhatchee Bay region in order to provide a context for what  follows.    The  focus  has  been  on  the  Late  Archaic  period  and  the  Elliott’s  Point  Complex  because  the  principal  occupation  of  the  Bayou  Park  site  was  during  this  archaeological  period.  Special emphasis was placed on describing what is known (and what is not) about  the  people  who  inhabited  this  part  of  the  Florida  panhandle  approximately  4500  to  2500  years ago.  This information, and particularly the gaps in our knowledge about the Elliott’s  Point complex, was used to develop a research design that is the subject of Chapter 3.     

68 

3

RESEARCH DESIGN AND METHODS 

Archaeological  work  conducted  over  an  18‐year  period  at  Bayou  Park  (8OK898)  has  demonstrated that the site’s primary occupation was during the Elliott’s Point phase of the Late  Archaic period, with only minor evidence for later Weeden Island and Fort Walton components  (see Chapter 1).   A historic‐period (mid‐twentieth century) component also is present.  Unlike  other sites where overlying occupations are extensive and separation of the Elliott’s Point  component is sometimes difficult, Bayou Park  represents a nearly “pure” Elliott’s Point site  with subsurface features, a midden containing preserved faunal and botanical remains, and  lithic,  ceramic,  and  bone  artifacts.    The  site  is  eligible  for  the  NRHP  because  these  intact  prehistoric deposits have potential to contribute to a better understanding of the Elliott’s Point  Complex.  Consequently, the primary focus of the data recovery effort was on gathering data  related to the Late Archaic, Elliott’s Point component at 8OK898 in order to mitigate impact  from future site development to this NRHP‐eligible component.    The summary of previous work in Chapter 1 indicates that much valuable information has been  obtained  from  the  site,  particularly  in  terms  of  general  site  layout,  documentation  of  the  stratigraphic  and  spatial  distributions  of  major  artifact  classes,  and  basic chronology.  The  approach taken for this data recovery effort was to utilize as much existing information as  possible  to  avoid  duplication  of  effort  and  to  focus  the  field  and  analysis  efforts  on  those  research questions that have not been addressed or for which adequate data have not yet been  collected from the site.  To achieve this goal, SEARCH developed a field strategy that utilized  ground‐penetrating radar (GPR) and mechanical stripping to identify midden areas and features  (trash  pits,  hearths,  postmolds,  etc.),  the  excavation  of  exposed  features,  a  targeted  investigation  of  midden  and  non‐midden  areas  using  hand‐excavated  test  units  to  obtain  stratigraphic data, the collection of faunal and botanical data from column samples, and the  collection of radiocarbon samples from features and controlled contexts in test units.    In the following sections, the research questions that guided the Bayou Park data recovery and  analysis are described, along with the field methods used to recover the necessary data classes  for addressing these questions.  Laboratory methods related to the processing of artifacts and  fauna  also  are  discussed.  Methods  related  to  specific  types  of  analyses  are  presented  in  individual chapters that discuss results of these analyses.   

RESEARCH DESIGN    In numerous reports and several published articles, Thomas and Campbell have laid out their  argument for Elliott’s Point as representing a local manifestation of the larger Poverty Point  phenomenon. More specifically, they consider the Bayou Park site to have been a particularly 

Bayou Park, 8OK898 

good example of a typical Elliott’s Point site.  Chapter 2 provides a complete discussion of their  argument as well as the relevant references.   In Tables 3.1 and 3.2 we summarize some of  their salient points, particularly those that can be evaluated with data from Bayou Park.      While Thomas and Campbell present a persuasive case regarding the connection between  Elliott’s Point and Poverty Point, we note that Bayou Park is one of the earliest recorded Late  Archaic  sites  in  the  Choctawhatchee  Bay  area,  with  four  radiocarbon  dates  indicating  occupation during the initial phase of the Elliott’s Point Complex, or sometime between 4481‐ 3766 cal BP (2532 and 1817 cal BC).  This span falls slightly before the earliest of dates recorded  for the Poverty Point site in Louisiana (Connolly 2006), predating the onset of the significant  mound building and exchange activities that would later come to define that regional center.   This begs the question – if the Elliott’s Point Complex is defined in large part by its association  with Poverty Point and the Poverty Point Exchange Network, yet one of its earliest sites pre‐ dates the fluorescence of Poverty Point, what then constitutes Elliott’s Point? How does the  Elliott’s  Point  Complex relate to Poverty Point? And how does Bayou Park fit into existing  interpretations of Elliott’s Point that were discussed in Chapter 2?     To answer these questions, we first needed to develop or expand on existing baseline data  regarding site chronology, site structure, subsistence and seasonality, technology, and site  function.  The specific questions related to these topics are outlined below.     

Dating and Site Chronology  Four C14 dates had been reported for the site prior to the 2011‐2012 excavation, but additional  dates were desirable to refine the site’s chronology, particularly since we expected structural  remains and features to be identified and excavated.   Additional radiocarbon dates also would  aid in both intra‐ and inter‐site analyses of subsistence practices, technology, and season of  occupation, and would provide data that could be used to answer question regarding any  changes in site use through time.  Obtaining radiocarbon dates from well‐controlled contexts  was thus a major goal of the project.      To guard against the possibility of dating non‐cultural charcoal or roots, we planned to obtain  samples for radiocarbon dating from charred wood and marine shells obtained from controlled  stratigraphic contexts and features.  Charred wood fragments can migrate vertically as a result  of bioturbation (root action, burrowing animals and insects) and we intended to choose these  samples carefully from sealed contexts that we were certain represented cultural events.  In  terms of shells, we intended to select robust shells, such as quahog clam (Mercenaria sp.), and  if possible use only a single species for dating to limit between‐species variation in carbon  uptake and fractionation (George Luer, personal communication, 2011).   

  Site Structure  Morehead et al. (2011:24) have observed that “Although rich deposits have been encountered  [at 8OK898], there is no observable pattern to the distribution.  It is possible to excavate in one  

70

Research Design and Methods 

  Table 3.1.  Characteristic traits of the Elliott’s Point Complex.  Traits  References  Elliott’s Point dates from approximately 2500‐600 BC  Campbell et al. 2004:132; Campbell et al.  2009:17; Morehead et al. 2008:81  Represents a continuation of the well‐developed Late  Thomas and Campbell 1993:528  Archaic adaptive pattern  Represents a local manifestation of the Poverty Point  Campbell et al. 2004:132‐133; Thomas and  culture  Campbell 1991, 1993:531;   Involvement in a well‐developed Poverty Point trade  Campbell et al. 2004:137; Thomas and  network a “distinguishing attribute”  Campbell 1991:108, 115, 1993:528, 531  Settlement pattern mirrors that found in the Lower  Thomas and Campbell 1991:105‐106, 1993:528  Mississippi Valley  Settlement is strongly coastal oriented  Campbell et al. 2004:134; Campbell et al.  2009:20; Thomas and Campbell 1991:104  Oyster, quahog clam, and scallops are a diagnostic  Thomas and Campbell 1993:541; Campbell et  feature of Elliott’s Point sites  al. 1998:77; Campbell et al. 2004:134; Campbell  et al. 2009:20  Most Elliott’s Point sites represent camps clustered  Thomas and Campbell 1993:528  around regional centers  Regional centers arranged in oval pattern, smaller sites  Thomas and Campbell 1991:105‐106,  in semi‐circular pattern  1993:528‐531  Accretional mounds are characteristic, e.g., Meig’s  Campbell et al. 2009:18  Pasture, Buck Bayou  Lithic workshops separated from mound or living areas  Thomas and Campbell 1991:106‐107,  1993:531‐532  Regional centers served as loci of community feasts,  Thomas and Campbell 1993:531‐532  redistribution centers for exotic, non‐local goods  Earliest evidence of Elliott’s Point Complex consists of  Thomas and Campbell 1993:533, 541  “crude, amorphous baked‐clay objects” in association  with Late Archaic projectile points  Well‐formed baked‐clay objects, particularly spheroids,   Thomas and Campbell 1991:108, 1993:533  are “typical” of “classic” Elliott’s Point Complex  Baked‐clay objects (Elliott’s Point Objects) “clearly”  Thomas and Campbell 1993:532  linked with Poverty Point  Microlith industry similar to Poverty Point (Jaketown  Campbell et al. 2004: 133; Thomas and  perforators)  Campbell 1991:108, 1993:533  Stemmed projectile points are “directly comparable” to  Campbell et al. 2004:133; Thomas and  projectile points found at Poverty Point, Claiborne, and  Campbell 1993:535  Cedarland  A particular style of hafted biface, referred to as a Destin  Thomas and Campbell 1991:110‐111, 1993:536  point, is characteristic of Elliott’s Point  No well‐developed lapidary industry  Thomas and Campbell 1991:108, 1993:534  A “moderately active” bone and shell industry” is  Thomas and Campbell 1991:108, 1993:535  characteristic  Bone and shell beads may have been manufactured for  Thomas and Campbell 1993:535  export in exchange for non‐local goods  Fiber‐tempered pottery introduced late in the Elliott’s  Campbell et al. 2004:139; Thomas and  Point chronology as an addition to well‐established  Campbell 1993:538‐539  complex. 

 

71

Bayou Park, 8OK898 

  Table 3.2.  Characteristics of Bayou Park site.  Characteristics  References  Contains all the elements necessary for Elliott’s Point:  Campbell et al.:1998:77  EPOs, Jaketown perforators, exotic items, “unusual” shell  composition (oyster, clam, scallop)  Portion of excavated midden area was the focus of food  Campbell et al. 1998:78‐79  preparation activities  Possible seasonal (fall, winter) occupation (hickory nuts)  Campbell et al.:1998:86  Possible late summer through fall occupation (scallops,  Morehead et al. 2011:114  hickory nuts)  “At least semi‐permanent” habitation, but year round  Meyer et al. 2002:64; Morehead et al.  occupation cannot be ruled out  2011:114  Northwestern area investigated in 2002 posited as a  Meyer et al. 2002:72  special activity area due to pieces esquillees, a rod, and a  double‐backed flake; possibly bead manufacturing.   Compared with 8WL36  Feature 4 in southeastern area interpreted as part of a  Meyer et al. 2002:64  feasting event  Lithic technology oriented toward biface production and  Morehead et al. 2008:79  maintenance  Possible earlier introduction and subdivision of EPC based  Campbell et al. 2004:132  on C14 dates from 8OK898  Possible living surface between 40 and 70 cm in non‐ Morehead and Aubuchon 2008:71‐72  midden area  No observable pattern to the distribution of  deposits  Morehead et al. 2011:24 

  area and uncover dense shell midden and features, and then place excavation units less than   10 m away and find no evidence of Elliott’s Point remains whatsoever….  Similar situations have  been encountered at a series of Elliott’s Point sites along Rocky Bayou and Rocky Creek where  multiple investigations failed to reconstruct intra‐site residential versus activity patterning….”     Although features (including a postmold) have been identified during previous investigations,  there  has  been  no  large‐scale  exposure  of  subsurface  features  sufficient  to  identify  the  structural layout of the site.  Nor has there been sufficient exposure of midden deposits to  determine the true nature and extent of these deposits.  If overlapping features are present,  this would suggest reuse and (perhaps) reoccupation of areas over time.  PTA’s 1998 block  excavation of a small portion of the site’s midden indicated that shell and non‐shell areas are  present.  Some of the questions that we wished to answer concerned the nature of these shell  and non‐shell deposits.  Do the shell areas represent discrete deposits of one or a few episodes  of refuse disposal?  Or are they lenses representing more continuous disposal over a lengthy  period of time?  Are the non‐shell areas devoid of fauna or do they contain vertebrate remains?   How  large  is  the  midden?    If  structures  could  be  identified  through  the  exposure  and  arrangement of postmolds or postholes, another set of questions arise.  It might be possible to  estimate structure size and thus infer population density at the site.  The size of structures, and  the size and numbers of posts associated with them, can provide information on the degree of  architectural  investment  by  site  occupants  which  can  then  be  used  to  infer  intensity  of  occupation.  How do these data compare with what is observed locally and regionally?  What  72

Research Design and Methods 

might  the  degree  of  occupation  intensity  mean  in  terms  of  interaction,  association,  or  environmental adaptation?    Several spatially distinct areas of artifact, midden, and feature concentration within 8OK898  suggest that there may be functional or temporal differentiation across the site in terms of  activities performed.  For example, lithic reduction areas are separated from living areas at  some Elliott’s Point sites.  Is there evidence for a similar separation of activity areas at Bayou  Park?  Additional data collection could determine if these spatially distinct areas represent  functionally different activity areas within a single site complex, or separate campsite locations  that were utilized at different points in time.  Finally, we wished to determine if any other  evidence of organized site structure exists.  For example, does the midden conform to the  arcuate structure identified at other Late Archaic sites in the region?   Thomas and Campbell  (1991:105) suggest that this is a trait shared with Poverty Point sites in the Lower Mississippi  Valley,  while  Russo  (2004)  suggests  that  arcuate  and  circular  shell  “rings”  are  the  loci  of  communal  feasting  events.     If a so‐called plaza exists, what were the activities that were  conducted within this social space?  What does the presence or absence of formal site structure  suggest about the earliest Late Archaic expressions on the Choctawhatchee Bay?    

Subsistence  A wide range of subsistence remains have been identified at 8OK898, including shellfish, fish,  deer, turkey, turtle, squirrel, and bird.  However, these previous excavations utilized a 1/4‐in  (6.4‐mm) mesh screen for recovery.  Numerous studies have demonstrated the necessity of  using fine‐mesh screens to obtain a more representative sample of vertebrate faunal remains,  particularly those associated with fish (e.g., Quitmyer 2004; Wing and Quitmyer 1985).  That  only four species of fish have been reported for 8OK898 (S. L. Scott 2002), underscores this fact  since coastal sites typically contain many more fish species.   In addition, quahog, scallop, and  oyster shell have been referred to as a “hallmark” of the Elliott’s Point Complex (Campbell et al.  1998:77), but no detailed faunal analysis of an Elliott’s Point site has ever been conducted.  Moreover, the only botanical remains reported from the site include a few hickory nut shells,  five  unidentified  plant  specimens,  and  one  unidentified  seed  from  a  bell‐shaped  feature  (Campbell et al. 198:75‐76; Roberts 2002).  Accordingly, a primary concern of the data recovery  project was the recovery of bulk samples for detailed faunal and botanical analysis.   Not only  were  we  interested  in  characterizing  the  diet  of  the  site’s  inhabitants,  but  we  wanted  to  compare these data to similar data from other Late Archaic sites in the Choctawhatchee Bay  region.      

Seasonality and Settlement Patterns  Bayou Park has been tentatively interpreted as a seasonal (fall‐winter) habitation based on the  recovery of hickory nut shells (Campbell et al. 1998:76, 86; Meyer et al. 2002:40).   Morehead  et al. (2011:114) expanded this to include late summer (July‐August) based on the presumed  seasonal availability of scallops in the bay during those months.  The seasonal occupation of the  site needed to be confirmed with additional data, specifically the analysis of clam shell growth  rings  and  detailed  botanical  analysis.    Was  the  site  truly  occupied  during  only  one  or  two 

73

Bayou Park, 8OK898 

seasons, or could it have been a permanent habitation?  If the site was occupied seasonally,  was there a consistency over time in the season of use, or could it have been occupied during  multiple seasons?  What does seasonality or multi‐seasonality suggest about early Late Archaic  society?    How  does  Bayou  Park  compare  with the diversity of subsistence and settlement  strategies observed both along the Gulf coast and regionally?     Seasonal occupation implies settlement mobility and the most likely scenario is movement  between the coast and the interior.  Because there are no significant toolstone resources along  the northwest Florida coast, movement to and from interior locales should have resulted in the  deposition at Bayou Park of lithic tools and manufacturing debris of raw materials obtained  from interior source areas (Lloyd et al. 1983; Upchurch et al. 1982).  Determining if these  materials were obtained directly, implying movement of people, or indirectly via exchange, is  difficult; however, the size and form of lithic and materials might provide clues as to whether  they were procured directly or indirectly via exchange (e.g., Austin 1997:150‐155, 158‐160).   

Site Function  Site  function  does  not  appear  to  have  been  addressed  in  previous  studies  beyond  the  designation of the site as a seasonal habitation.  A formal analysis of site function taking into  account seasonality data, number and types of features, evidence of structures and their sizes,  the presence of interior or exterior hearths, and their locations is necessary to adequately  address this issue.  Changes in site function through time may be determined depending on the  number of features and radiocarbon dating of various features.    An  important  aspect  of  site  function  is  establishing  the  intensity  of  site  use.      Differential  patterning in site intensity is a reflection of different types of types of settlements, for example  a nucleated village versus a dispersed campsite.  This is based on the assumption that the  amount of material remains deposited at a site is a function of the span of time that is spent at  a specific location during the annual cycle and the cumulative time over which the site has been  used (Lightfoot 1984:89).  By plotting these variables on X and Y axes, it is possible to identify  four idealized types of archaeological sites based on intensity of use (Figure 3.1).      To operationalize this model, we planned to use a method devised by Gallivan (2002).  Using  archaeological data that had been shown to be highly correlated with one another, Gallivan  developed  two  indices  that  could  be  plotted  according  to  Lightfoot’s  heuristic  model.   Residential  stability  is  determined  using  the  following  measures:  feature  richness,  lithic  evenness, mean house floor area, mean post diameter, mean number of structural wall posts,  mean number of interior features, and mean pit volume.  The average of the resulting values is  the “residential stability index” (Gallivan 2002:542).  A similar index attempts to identify the use  duration of a site by averaging the densities of features, postmolds, burials, and artifacts, in  addition to the ratio of burials to houses (Gallivan 2002:543).  The resulting measures are  standardized so that values range between 0 and 1 and these are then plotted on a bivariate  graph, allowing placement of individual sites within the four idealized categories shown in  Figure 3.1.    74

Research Design and Methods 

1

Multiple, year-long occupations

Use Duration

Multiple, brief occupations

Single, brief occupations

Single, year-long occupations

0 0

1

Residential Stability

Figure 3.1.  Model of occupational intensity (after Lightfoot 1984). 

Except for burials, Bayou Park contains all of the data classes used by Gallivan in his study and  has the added benefit of containing faunal remains.  We hoped to be able to develop similar  indices with these data and compare the results to more conventional analyses of assemblage  diversity and organizational structure to arrive at an interpretation of site function. 

  Technology   Two major classes of artifacts have been identified at Bayou Park: baked‐clay objects (aka  Elliott’s Point Objects) and stone tools and manufacturing debris.  Neither has received detailed  analysis, although subjective comparisons between sites have been made in several reports.     

Baked‐Clay Objects   Baked‐clay objects are one of the distinguishing characteristics of Elliott’s Point Complex sites  and one of the chief reasons for positing a relationship with Poverty Point; but there are many  unanswered  questions  about  their  origins  and  use.    What  types  of  baked‐clay  objects  are  present at Bayou Park? What might the variety (or lack thereof) of baked‐clay objects suggest  about the development of the technology or the utilization of specific cooking technologies  during the Late Archaic? Are there diachronic or synchronic patterns of baked‐clay object types  present?    What  might  the  presence  or  absence  of  these  patterns  mean  in  terms  of  site  organization or site utilization through time?   How do these data compare to what is observed  both locally and regionally throughout the Late Archaic Period?   How do they compare to other  sites with baked‐clay objects, including Poverty Point? Were the better formed, deco, rated  objects acquired via trade, or were all baked‐clay objects manufactured locally?  We hoped to  be able to address some of these questions through detailed analysis of paste and temper 

75

Bayou Park, 8OK898 

characteristics,  between  site  comparisons  of  assemblage  composition,  and  the  use  of  instrumental analyses to compare selected objects with local clay sources.    Microliths   Microliths are part of the material culture package that is used to infer a relationship between  Elliott’s Points sites and Poverty Point‐related sites.   While microlithic stone tools are relatively  common at Elliott’s Point sites, they are not abundant at Bayou Park.  Is this due to sampling  bias or is it a true reflection of the stone technology of the site’s inhabitants?  If microliths are  indeed rare at Bayou Park, but common at other contemporaneous sites, what could be the  reason?  There has been no detailed analysis of lithic technology and raw material acquisition  strategies, nor any functional analysis of stone tools.  What were Elliott’s Point microliths used  for and how closely do they resemble the Jaketown Perforators and other microlithic tools  found at Poverty Point‐related sites elsewhere?      Projectile Points  Elliott’s Point sites appear to contain a wide variety of stemmed bifaces, some of which are  claimed to resemble types found at Poverty Point and other related sites (Morehead et al.  2011:111‐114; Thomas and Campbell 1991:108‐111, 1993:535‐536).  This includes a unique  type reportedly identified by Clarence Webb and referred to locally as a Destin point (Thomas  and Campbell 1991:110‐111).  Morehead et al. (2011:112) have noted the basic similarity in the  haft areas of the various stemmed points identified at Bayou Park; that is, they tend to have  sloping shoulders, a straight to slightly contracted stem with concave edges, and an essentially  straight base.  This distinguishes them from the Destin type, which tends to have a concave  stem base.  Is the Destin point a true diagnostic that can be consistently separated from other  Late Archaic stemmed points?  Or is it simply a “variation on a theme”?  Alternatively, what do  the straight‐based stemmed points at Bayou Park represent?  Do they reflect a different social  group  or  are  they  simply  Destin  variants?    Finally,  how  does  the  overall  lithic  assemblage  compare  with  other  Late  Archaic  sites  in  the  panhandle  and  what  might  similarities  or  differences in these assemblages suggest about settlement and interaction?   

Poverty Point Relationships  As previously discussed, the Elliott’s Point complex is believed to be “a localized expression of  the Poverty Point culture” (Thomas and Campbell 1991).  Similarities in settlement patterns and  artifact assemblages, along with the presence of exotic artifacts at some sites, form the basis  for  this  interpretation.    The  review  of  previous  research  in  Chapter  2  demonstrates  that  similarities between Elliott’s Point and Poverty Point sites do exist; however, the interpretation  of these similarities and what they represent has been contested, most notably by Nancy White  (2003b, 2004), who suggests that similarities in material culture may simply be the result of  similar site functions and subsistence practices by coastal‐oriented groups.  While it is unlikely  that the debate over Elliott’s Point and Poverty Point relationships can be resolved with data  from a single site, there are some questions that can profitably be investigated at Bayou Park.   Certainly material culture comparisons were a major goal of the project.  Do the baked‐clay  objects from Bayou Park closely resemble those from Poverty Point‐related sites, or do they, as  White has suggested, represent the use of a similar technology to address a similar problem,  76

Research Design and Methods 

that  is,  the  absence  of  stone  suitable  for  dry  or  wet  heat  cooking?    Does  the  microlith  assemblage  at  Bayou  Park  and  other  Elliott’s  Point  sites  truly  resemble  the  microlith  assemblages from Poverty Point, Jaketown, and other related sites, or is it a local manifestation  of a more widely used technology?  Is the settlement pattern of small habitation sites and  campsites surrounding a regional center a valid interpretation of the archaeological data?  Is  the wide‐spread appearance of arcuate shell middens a cultural or social construct (Russo  2004), or is it simply an artifact of landscape and site formation processes (Marquardt 2010a,  2010b)?      Is  there  any  evidence  of  trade  at  Bayou  Park?    What  types  of  trade  items  are  represented?  What might the presence or absence of exotic items suggest about this site  relative to other early Choctawhatchee Bay Late Archaic sites in terms of significance and  interaction?   What are the overall implications of the data recorded from 8OK898 regarding  site  utilization,  structure,  and  seasonality  on  a  local  and  regional  scale?  How  does  the  development of the Late Archaic populations around the Choctawhatchee Bay compare to the  greater development of Late Archaic traditions and what might this means in terms of long‐ distance interaction and association or identity?     

Site Formation  A  final  area  of  interest  is  the  archaeological  deposits  themselves.    In  particular,  we  were  interested in examining the efficacy of excavating only where artifacts or midden deposits are  abundant.  Most investigations of archaeological sites in Florida follow this approach under the  assumption  that  areas  of  high  artifact  density  represent  “activity  loci”  while  areas  of  low  artifact density have tended to be overlooked (Austin 2002; Mathis 1993).   Since much of the  archaeological record is the result of human discard practices, the possibility that areas of low  artifact density might contain evidence of behaviors other than those related to ceramic and  stone tool use and discard is a serious issue from both the research and cultural resource  management perspectives.  Because we intended to expose large areas using heavy equipment  to look for features, this seemed to offer an opportunity to test this assumption by comparing  the spatial distributions of archaeological features with the spatial distributions of midden  deposits or other concentrations of artifacts that had been identified previously through broad‐ scale testing.  If the correlation between features and middens or artifact concentrations is  positive,  then  our  strategies  of  investigating  most  intensively  those  areas  where  cultural  materials are abundant are justified.  However, if the correlation is negative, then greater  efforts should be made to investigate low‐density areas during testing and excavation projects.   

FIELD METHODS    Prior to fieldwork, the project area was divided into three arbitrary areas – Operations A, B, and  C  –  as  shown  on  Figure  3.2.    These  areas  served  to  delineate  portions  of  the  site  for  the  purposes  of  obtaining  AF‐103  Civil  Engineering  Work  Clearance  for  buried  utilities.    An  extension of the site to the east along Postl Lake (labeled B‐extension on Figure 3.2) was not   included in the AF‐103 permit because no excavation was planned in this low‐density part of  the site, although ground‐penetrating radar work was conducted there.    

77

Figure 3.2.  GPR grids within Operations A, B, and C. 

Bayou Park, 8OK898 

78

Research Design and Methods 

Before discussing the field methods, we should explain the use of directions when referring to  the orientation of survey areas, mechanical strip areas, and test units.  The SEARCH grid system  established for the site is based on magnetic north.  While most test units also are oriented to  magnetic  north,  the  ground‐penetrating  survey  areas  and  the  mechanical  strip  areas  are  oriented parallel to existing buildings and sidewalks, which are oriented parallel to existing  roads,  or  at  angle  of  approximately  35°  east  of  north.    To  avoid  confusion  and  facilitate  communication in the field, we established an operational “north” for the survey and strip  areas which corresponded to northeast on our grid system.  Throughout the report, when  referring to these operational directions, we will place them in quotes to distinguish them from  magnetic north.     

Ground‐Penetrating Radar Survey  The first task of the data recovery effort was a ground‐penetrating radar (GPR) survey of the  site.    The  survey  was  conducted  in  an  effort  to  identify  subsurface  midden  deposits  and  possible  features,  modern  utility  lines,  and  other  subsurface  anomalies.    GPR  uses  electromagnetic  waves  generated  from  an  antenna  as  the  energy  source  from  which  measurements are collected and interpreted.  The radar energy that is propagated into the  ground is reflected off buried objects, features, and changes in soil types (Conyers 2004).  The  lapse  in  travel  time  and  varying  amplitude  of  the  electromagnetic  waves  collected  along  transects are interpolated to create a two‐dimensional image of the subsurface that are used to  identify soil disturbances that could signal the presence of archaeological features, buried  objects, grave shafts, and changes in sediment type.      Because of the presence of buildings, sidewalks, and other obstacles, Operations A, B, and C  were subdivided into sections using convenient landmarks (see Figure 3.2).  To ensure accurate  georeferencing of located anomalies, a datum stake was placed in a selected corner of each  grid and the location of the stake was determined using a handheld Trimble XH Series 6000  GPS.  Once the site grid system was established (see below), each grid datum was recorded in  relation to the site datum (1000N, 1000E) with a Sokia Model 50X total station (Table 3.3).   Geophysical data were collected using a GSSI/SIR‐3000 Integrated Radar Control Unit, mounted  on an UtilityScan wheel‐cart with a 400‐MHz shielded antenna (Figure 3.3).  The wheel cart  containing the GPR unit was pushed along transects spaced 50 cm apart, and transect start and  stop locations were documented on field data sheets to cross‐reference accuracy during post‐ processing.  The GPR data were processed and analyzed with GPR‐SLICE (Version 7.0) imaging  software, and GSSI’s RADAN (Version 7) GPR processing software.    

Grid System and Site Mapping  A previous grid system was established in 2002 by PTA (Meyer et al. 2002).  Several permanent  datum points also were established and elevations in meters above mean sea level (amsl) were  obtained for these points.  Additional datum points were established during a 2009 testing  project by PTA (Campbell et al. 2009).  Unfortunately, except for the building corners and a  manhole cover, none of these datum points could be relocated.  This required that the original  grid datum be reestablished from existing field notes.  According to the 2002 PTA report and 

79

Bayou Park, 8OK898  Table 3.3.  GPR grid sizes and location data. 

  GPR  Grid # 

1  2  3  4  5  6  7  8  9  10  11  12  15  16  17  18  19  24  41  42  43  Subtotal    25  26  27  28  29  30  31  32  33  34  35  36  37   Subtotal   14  20  21  22  23  38  39  40  44  Subtotal   13  Total 

80

A  A  A  A  A  A  A  A  A  A  A  A  A  A  A  A  A  A  A  A  A 

Northing,  Easting  995N, 966E  979N, 990E  960N, 1020E  967N, 1043E  1012N, 1022E  1050N, 1047E  1004N, 1014E  996N, 1046E  1010N, 1083E  1036N, 985E  1049N, 1008E  1030N, 1088E  945N, 1042E  1062N, 1001E  1079N, 1013E  1110N, 1032E  1101N, 1042E  1056N, 1054E  1124N, 1102E  1097N, 1079E  1157N, 1086E 

Grid Datum Corners  UTMs  (Northing, Easting)  3371629N, 549017E  3371613N, 549041E  3371594N, 549070E  3371601N, 549095E  3371646N, 549073E  3371682N, 549098E  3371638N, 549065E  3371629N, 549098E  3371642N, 549134E  3371670N, 549036E  3371683N, 549058E  3371700N, 549165E  3371577N, 549093E  3376940N, 549052E  3371711N, 549063E  3371741N, 549082E  3371733N, 549093E  3371690N, 549105E  3371756N, 549153E  3371729N, 549130E  3371789N, 549137E 

B  B  B  B  B  B  B‐EXT  B‐EXT  B‐EXT  B‐EXT  B‐EXT  B‐EXT  B‐EXT 

910N, 1204E  919N, 1249E  964N, 1268E  944N, 1283E  946N, 1260E  986N, 1269E  948N, 1328E  985N, 1351E  1001N, 1391E  1030N, 1446E  1038N, 1445E  1002N, 1298E  987N, 1291E 

3371545N, 549256E  3371554N, 549301E  3371596N, 549319E  3371576N, 549332E  3371578N, 549311E  3371618N, 549320E  3371580N, 549379E  3371617N, 549402E  3371633N, 549442E  3371662N, 549497E  3371670N, 549496E  3371636N, 549349E  3371619N, 549342E 

C  C  C  C  C  C  C  C  C 

987N, 1120E  951N, 1110E  939N, 1143E  987N, 1162E  997N, 1156E  1063N, 1194E  1049N, 1181E  1056N, 1177E  1050 N, 1216 E 

3371621N, 549171E  3371583N, 549161E  3371573N, 549194E  3371620N, 549213E  3371629N, 549207E  3371695N, 549245E  3371681N, 549232E  3371688N, 549228E  3371682N, 549268E 

970N, 1084E 

3371605N, 549136E 

Operation 

A&C 

Size (Acres)  0.301  0.099  0.094  0.012  0.315  0.030  0.089  0.334  0.062  0.038  0.010  0.473  0.019  0.040  0.158  0.059  0.077  0.037  0.040  0.067  0.053  2.429  0.055  0.036  0.086  0.047  0.030  0.033  0.086  0.100  0.052  0.098  0.040  0.069  0.034  0.900  0.225  0.133  0.145  0.064  0.033  0.064  0.015  0.016  0.010  0.752  0.202  4.081 

Research Design and Methods 

Figure 3.3.  GSSI GPR unit used for data collection at 8OK898. 

  field maps on file at Eglin AFB, the 1000N, 1000E point was located about six meters to the  northeast of a concrete sewer marker in the northwest corner of the prison camp.  This location  corresponded  roughly  with  the  common  corner  of  TPs  7,  11,  and  14  in  PTA’s  1998  block  excavation.  The concrete marker was recorded as 996N, 964E on field maps for the 2002  project,  with  angles  and  distances  recorded  to  the  corners  of  Buildings  588  and  591.    In  contrast, the 2009 report records the concrete marker as being located at 999.1N, 964.4E.  It  does not appear that a permanent hard point, such as steel rebar, was used to mark the 1000N,  1000E datum originally, and conversations with Jan Campbell of PTA and Joe Meyer, currently  with Eglin AFB, failed to resolve the discrepancy or positively identify either the concrete sewer  marker or the exact location of 1000N, 1000E.  Therefore, it was necessary to use the total  station and the locational data recorded on the 2002 field map to relocate and reestablish both  of these points.  Using this method it was determined that a square concrete sewer marker  located near the intersection of Inverness Road and Flagler Road (Figure 3.4), was the probable  concrete sewer marker referenced in both reports.  It should be noted, that the angles and  distances on PTA’s field map do not match exactly the location of this concrete marker.  In  addition, as there was no way to know for certain which of the two coordinates for this location  is correct, we assumed the original coordinates of 996N, 994E for the concrete sewer marker  and using this as a starting point, PTA’s 1000N, 1000E datum was reestablished.  Subsequent  stripping,  which  revealed  the  outline  of  the  1998  excavation  block,  indicated  that  the  reestablished datum location is approximately ~ 1.24 m southeast of the original datum (Figure  3.5).  A two‐foot long piece of steel rebar was set at this new location as a permanent marker 

81

Bayou Park, 8OK898 

Figure 3.4.  View to the northeast of Operation A showing locations of the concrete sewer  marker, PTA’s 1998 block excavation, and SEARCH’s grid datum.  (UTM E549051 N3371632, Zone 16, WGS84).  A second permanent datum stake (two‐foot long 

steel rebar) also was set at 1000N, 1200E (UTM E550251 N3372632).    The elevation of the concrete sewer marker is recorded in the 2009 report as 3.30 m amsl and  the elevation of a second datum (a manhole cover located north of Building 633, which was  identified  in  the  field)  is  recorded  as  3.88  m  amsl.    Using  these  two  elevations  and  the  reestablished  1000N,  1000E  datum  location,  SEARCH  established  the  grid  locations  and  elevations of 60 hard points across the site including all of the lamp poles, several building  corners, and miscellaneous electrical boxes and fire hydrants.  These included several of the  building corners that were used as hard points by PTA.  These data are provided in Appendix  3.1.  Our coordinates and elevations for their hard points were generally within tolerable limits  (± 1 m) of the original documented data such that we felt confident that PTA’s grid system and  datum  plane  were  reestablished  as  accurately  as  possible.  The 60 hard points were used  throughout the project to establish locations and elevations for strip areas, test excavation  units, and most importantly, features exposed during mechanical stripping.  In addition, all of  the GPR datum stakes were tied into the grid system and datum plane.  Finally, the locations of   all hard points, including the 1000N, 1000E and 1000N, 1200E datum stakes, were recorded  with the Trimble GPS.     

82

Research Design and Methods 

TP60

Figure 3.5.  Relative locations of SEARCH and PTA 1000n, 1000E  datum points within Operation A. 

Aligning PTA Excavations with SEARCH Grid  When available, we used the grid coordinates recorded in PTA field notes or maps included  with PTA reports to determine the locations of previously excavated shovel tests and test units.  Unfortunately, PTA excavations conducted prior to 1998, and some excavated after this date,  have no grid coordinates associated with them and it was necessary to estimate their locations  using maps available in PTA reports and from field notes.   Moreover, the compass and tape  method used to record many of the shovel test and test unit locations originally (Joe Meyer,  personal communication, 2012) are less accurate than the total station resulting in random  measurement errors.  In general, the procedure that was used to reconcile the two systems  involved using the UTM coordinates for the 60 SEARCH hard points that were obtained with the  Trimble GPS, then plotting these on a georeferenced aerial photograph of the site in GIS (Zone 

83

Bayou Park, 8OK898 

16, WGS84).  Using the SEARCH 1000N, 1000E datum as a reference, all of the PTA shovel tests  and test pits with coordinates also were converted to UTMs and plotted on the same aerial.   All  of the PTA tests without coordinates were manually added to the aerial based on their relation  to  one  another  and  known  landmarks  (e.g.,  building  corners,  sidewalks,  etc.).  Their  UTM  coordinates were then calculated in GIS and these were then converted mathematically to the  SEARCH grid system.      Because the two grid systems do not align perfectly, we knew that adjustments would need to  be made so that previous excavations could be matched as closely as possible with the SEARCH  grid system.  When comparing the locations of PTA tests on the GIS map to PTA report or field  maps, it was clear that some of these tests were not plotted correctly on the GIS.  However, we  were fortunate in being able to identify PTA’s 1998 excavation block during stripping as well as  their TPs 19 and 20, which enabled us to georectify some locations.  We have discussed the  excavation  block  above,  but  will  describe  the  two  test  unit  locations  here  because  they  illustrate the difficulty in matching up the two grid systems and provide justification for our  final decisions regarding grid alignment.    In Strip Area K, a 1‐x‐1‐m square area of mottled soil was identified which we believe to be  PTA’s  TP19.    Two  smaller,  square  stains  also  were  identified  in  this  strip  area  and  were  identified as PTA’s STs 167 and 168.  These three excavations were conducted during a 2002  survey of a sprinkler and communication trench (Meyer et al. 2002).  The grid location of TP19  recorded in PTA field notes is 1026N, 1041E; on the SEARCH grid the coordinates are 1024.55N,  1040.55E  or  a  difference  of  ‐1.45N,  ‐.45E.    Similarly,  the  grid  coordinates  for  ST  167  are  reported in field notes as 1009N, 1035E and those for ST168 as 1027N, 1048E.  Using the  SEARCH  grid,  the  coordinates  for  these  shovel  tests  become:  ST167,  1013.31N,  1032.92E;  ST168, 1013.20N, 1028.10E.  The differences from the recorded coordinates are much greater  for the shovel tests than for TP19 and this is likely due to the methods used to record their  locations  originally  in  the  field;  i.e.,  compass  and  tape.    Considering  the  close  agreement  between the two coordinate systems for TP19, as well as the fact that no other large, square  excavation is recorded as being close to the 1‐x‐1‐m stain revealed by stripping, and assuming  the relational accuracy of the shovel tests and test units on the PTA maps, we feel confident in  stating that the two smaller square stains are indeed STs 167 and 168.  Using these three units  as reference points, it was then possible to visually shift other shovel test locations along the  2002 sprinkler and communication trench alignment to more closely match their locations in  relation to Buildings 591, 626, and 632 as shown on PTA maps.     PTA’s TP20 was more difficult to locate.  The coordinates for this unit, which actually consisted  of two overlapping 2‐x‐2‐m units excavated to expose a feature, are recorded in field notes as  927N, 1252.7E.  Using the total station, we reestablished this location and used the hydraulic  excavator to strip the area in hopes of seeing the characteristic straight‐sided, mottled soil stain  indicative of a back‐filled excavation unit.  None was identified.  We then used measurements  obtained from PTA’s field maps to measure the TP’s location in relation to Building 636.  This  area was then stripped and the southern edge of TP20 was identified.  The SEARCH coordinates  for the unit are 933.63N, 1238.30E.  Using this location as a reference point, we then adjusted  84

Research Design and Methods 

the shovel tests excavated along the sprinkler line to more closely match their relationship to  TP20 and existing buildings.    Final manual adjustments to other PTA tests were made as necessary to match the locations  shown on PTA maps as closely as possible.  These were converted to the SEARCH grid system  using the UTM/grid conversion method described above.  The new coordinates were then used  for all subsequent mapping and distributional analysis.      This  lengthy  discussion  has  been  necessary  to  explain  the  differences  in  grid  coordinates  between the two grid systems.  Certainly one potentially complicating problem is the reporting  of two different sets of coordinates for the original concrete sewer marker, from which all  other  coordinates  (including  both  the  PTA  and  SEARCH  1000N,  1000E  data  points)  were  measured.  Our decision to use the original set of coordinates rather than the second, revised  set of coordinates can be questioned, but at the time we had no good reason to choose one set  of coordinates over the other.  We chose the original set because we felt that these gave us the  best opportunity for relocating PTA’s original 1000N, 1000E datum.  In hindsight, using the  second set of coordinates might have lowered the margin of error between the two systems,  particularly for later excavations.  On the other hand, the fact that we could determine no  constant  mathematical  correction  regardless  of  which  system  was  used  suggests  that  the  location errors were random and adjustments would have needed to be made regardless of  which system was used.  For those who wish to use them, we have included the original and  revised PTA coordinates along with the new SEARCH coordinates in Appendix 3.2.      

Mechanical Stripping   The large number of utility lines that crisscross the site, as well as the presence of numerous  structures and sidewalks, made site‐wide stripping with heavy equipment impractical.   In  addition, as a condition of the AF‐103 process, only hand‐digging was allowed within three feet  of any buried cable lines.  Consequently, mechanical stripping of selected areas was performed  instead.  Based on the previous work conducted at the site, the areas that seem to offer the  best potential for stripping included the area in and around Buildings 588 and 591 (both of  which had been recently demolished), and the open field near PTA’s 1998 block excavation  both located within Operation A.  The goal was to continue to expose, excavate, and analyze  the midden and feature deposits in this portion of the site.  Operation A also contained midden  deposits with abundant faunal remains which we planned to collect for analysis with column  samples.  Operation B, in the southeastern portion of the site nearest Postl Lake, included the  location where PTA excavated a large pit feature and recovered a large number of finished  tools (TP20).  Stripping of selected areas here was planned in order to determine if other  features are present and if so to gather data and radiocarbon dates for comparison with the  midden in Operation A.   Operation C included the central courtyard and several buildings.   Limited stripping was planned in this area between the existing sidewalks where few artifacts  had  been  found  previously.    Finally,  the  location  of  a  sandstone  feature  identified  during  monitoring to the north of Building 635 in 2003 also was planned for stripping.      85

Bayou Park, 8OK898 

Figure 3.6.  View to the “north” of Strip Area K with hydraulic excavator.  Note midden deposits on the left. 

In all, mechanical stripping was performed in 30 separate areas and was accomplished using a  Caterpillar tracked hydraulic excavator with a 36‐inch ditch clean‐out (toothless) bucket to  remove overburden and expose subsurface features (postmolds, pits, etc.) (Figure 3.6).  The  equipment was selected for its size, tracked wheels, and maneuverability to minimize any  potential damage to exposed archaeological deposits, existing sidewalks, and lawns.   Individual  strip areas were labeled by letters (e.g., Strip Area A, B, C, etc.) and these are shown on Figures  3.7‐3.9.    The  placement  and  size  of  the  strip  areas  were  determined  by  several  factors,  including previous survey and test excavation results, the GPR results, and the need to avoid  various obstacles (buildings, sidewalks, buried utilities, and large trees).  Nineteen were located  in Operation A, eight in Operation B, and three in Operation C.  Table 3.4 provides data on the  size and volume of each of the 30 Strip Areas.    All mechanical stripping was monitored by an archaeologist and was conducted according to  the following procedure.  The hydraulic excavator first removed fill and overburden where  present.  The boundaries and surface elevations of each strip area then were recorded with the  total station in relation to the site grid system and datum plane.  Excavation with the hydraulic  excavator then proceeded in 10‐to‐15‐cm increments in order to expose large areas in plan  view (see Figure 3.6).  At the completion of each 10‐to‐15‐cm “level,” exposed surfaces were  shovel‐shaved and any midden deposits, features, or diagnostic artifacts were photographed  and mapped, and their locations and elevations recorded with the total station.  Selected  features also were excavated at this time (see Excavation Methods below).   After all mapping  and  hand  excavation  were  completed,  stripping  resumed  and  was  carried  out  in  a  similar

86

Research Design and Methods 

  

Figure 3.7.  Strip areas and test pit locations in Operation A. 

87

Bayou Park, 8OK898 

Figure 3.8.  Strip areas and test pits locations in Operations A and C. 

fashion until no additional midden deposits or features were encountered.  The termination  depths  varied,  but  usually  were  no  less  than  1.25  m  below  ground  surface,  with  several  trenches excavated to approximately 1.5 m below ground surface.    At least one wall in each strip area was cleaned and photographed, and profile drawings were  created to document the stratification of the entire site from its northwestern edge, near  building 591, to its southeastern terminus near Postl Lake.  In addition, a perpendicular profile  from several trenches located in Operation A also were photographed and drawn in order to  document  stratification  through  the  midden  area  from  southeast  to  northeast.    Stratum  designations for all profiles followed existing nomenclature established in previous work by PTA  (see Chapter 4).  Bottom elevations were obtained with the total station for each strip area  prior to backfilling.   

Excavation Methods  A total of 30 test excavation units of various sizes and eight 50‐cm‐in‐diameter shovel tests  were excavated at 8OK898 (see Figures 3.7‐3.9).  Continuing the practice established by PTA,  each unit and shovel test received a unique identifying number as well as a grid coordinate 

88

Research Design and Methods 

Figure 3.9.  Strip areas and test pit locations in Operation B. 

89

Bayou Park, 8OK898 

    Table 3.4.  Size and volume data for mechanical strip areas.  Strip Area  A  B  C  D  E  F  G  H  I  Ja  Jb  K  L  M  N  O  Pa  Pb  Q  R  S  T  U  V  Wa  Wb  X  Y  Z  AA  Totals 

Operation  A  A A A A A A A A A A A A B  B  A  B  B  A  B  B  B  B  A  C  C  C  A  A  A 

Perimeter (m)  18.86  58.53  39.83  56.48  71.86  39.51  19.82  36.95  39.25  92.37  77.84  96.59  68.81  40.02  52.02  41.53  20.01  127.96  72.86  48.71  18.93  26.56  34.09  57.28  20.91  18.10  44.64  88.46  31.61  27.43 

Area (m2)  21.82  61.32  93.62  150.30  190.66  75.71  24.17  72.90  75.08  303.94  249.36  291.02  182.97  56.73  131.67  71.56  16.82  137.42  223.59  77.05  20.71  32.26  51.83  162.38  20.76  19.00  92.93  159.12  46.49  46.78  3113.19 

Volume (m³)  32.74  91.98  140.42  225.45  285.98  113.57  36.25  109.35  112.62  455.91  374.04  436.53  274.45  85.10  197.50  107.34  25.24  206.13  335.39  115.57  31.07  48.39  77.75  243.57  31.14  28.50  139.39  238.69  69.73  70.16  4669.79 

  designation.  The southwest corner was used as the identifying pair of coordinates for the test  unit while the center points were used for the shovel tests.  The total volume of excavated  sediments from these units was 76.21 cubic meters (Table 3.5).      Shovel tests were judgmentally excavated within Building 588 (n=3), Building 591 (n=3), and  just  outside  and  to  the  northwest  of  Building  591  in  order  to  determine  the  presence  or  absence of midden within and adjacent to these demolished buildings.  The shovel tests were  excavated to 100 cmbs unless  dense  midden  was  encountered,  in  which  case  they  were  

90

Research Design and Methods 

Table 3.5.  Grid coordinates and elevation data for SEARCH shovel tests and test pits.  Grid Coordinatesa, b Elevation  Volume  Type  Number  Size  Operation  (m)b  (m3)  N  E  ST  257  Void  ST  258  50 cm dia.  977.00  1038.10  3.04  0.20  A  ST  259  50 cm dia. 983.81  1028.80  3.17  0.20  A  ST  260  50 cm dia. 995.88  1012.80  3.45  0.06  A  ST  261  50 cm dia. 1033.64  1013.29  3.68  0.20  A  ST  262  50 cm dia. 1039.90  1018.34  3.61  0.09  A  ST  263  50 cm dia. 1053.60  1025.24  3.71  0.20  A  ST  264  50 cm dia. A  1040.94  1010.10  3.76  0.20  ST  265  50 cm dia. A  1043.67  1012.00  3.97  0.20  TP  58  1‐x‐2  3.61  2.00  A  976.98  1035.87  TP  59  2‐x‐2  3.19  4.00  A  984.46  1028.40  TP  60  1‐x‐2  3.42  2.00  A  994.99  1010.61  TP  61  1‐x‐2  3.22  1.92  A  987.03  1021.71  TP  62  1‐x‐2  3.71  1.94  A  1052.58  1025.5  TP  63  1‐x‐2  3.61  2.00  A  1037.29  1018.16  c  TP  65  1‐x‐1  1010.39  1002.83  2.90 0.70  A  TP  66  1‐x‐1  1011.13  1003.48  2.99c  0.69  A  TP  67  1‐x‐1  1012.87  1004.18  3.02c  0.57  A  TP  70  1‐x‐1  1009.69  1003.55  2.50c  0.32  A  c  TP  71  1‐x‐1  1010.45  1004.22  2.29 0.29  A  TP  74  1‐x‐1  1008.32  1003.61  2.93c  0.67  A  TP  75  1‐x‐1  1009.01  1004.28  2.89c  0.96  A  TP  76  1‐x‐1  1009.78  1004.97  2.83c  1.02  A  c  TP  79  1‐x‐1  1007.64  1004.31  3.11 0.72  A  TP  80  1‐x‐1  1008.36  1005.00  3.08c  0.91  A  TP  84  .5‐x‐1  3.15c  0.49  A  1010.91  1026.31  TP  85  2‐x‐2  3.85  6.20  A  1080.30  1071.39  TP  86  2‐x‐2  3.44  6.12  A  1122.97  1106.21  TP  87  2‐x‐2  2.97  6.56  B  993.47  1252.10  TP  88  2‐x‐2  2.93  4.80  B  953.20  1245.13  TP  89  2‐x‐2  2.88  5.24  B  947.57  1252.75  TP  90  1‐x‐1  2.14c  0.50  A  1004.67  1044.47  c  TP  91  1‐x‐2  1.97 0.84  A  1009.09  1041.31  TP  92  1‐x‐2  2.67c  2.80  B  933.21  1238.43  TP  93  2‐x‐2  3.00  6.00  C  989.46  1160.03  TP  94  2‐x‐2  3.00  6.32  C  993.64  1166.72  c  TP  95  1‐x‐2  2.11 2.80  C  1061.50  1208.35  TP  96  1‐x‐2  2.00c  2.68  C  1049.88  1220.03  TP  97  1‐x‐2  2.09c  2.80  C  1063.23  1207.27  Total    76.21  a  Coordinates based on 1000N, 1000E steel rebar located at UTM E549051 N3371632 (Zone 16 WGS84). b  Shovel test coordinates and elevations are center points; test pit coordinates and elevations are the  southwest corner.  c  Ground surface after stripping.   

91

Bayou Park, 8OK898 

terminated without impacting the midden further.  Excavated sediments were sifted through  .64‐cm (1/4‐in) mesh screens.    Most test units were excavated in Operation A (n=21).  Six 1‐x‐2‐m units (TPs 58‐63) were  excavated within the confines of demolished Buildings 588 and 591 prior to stripping (Figure  3.10).  These served to better delineate the midden deposits in these two areas.   A block of 20  1‐x‐1‐m units (TPs 64‐83) was laid out over a concentration of large pit features in Strip Area Ja  in order to identify these in more detail, but only 10 of these were actually excavated (Figure  3.11).  Test Pit (TP) 84 was a small (.5‐x‐1 m) unit excavated in the south wall of Strip Area K to  investigate a sand midden stratum.  TPs 85 and 86 were placed near PTA’s TPs 40 and 39,  respectively, both of which contained shell, fired‐clay fragments, and lithics.  TP85 was located  between Buildings 626 and 632 and TP86 was excavated northwest of Building 633.    In Operation B, a 1‐x‐2‐m unit (TP92) was excavated immediately to the south of PTA’s TP20  where a large shell‐filled pit and an abundance of lithic artifacts were identified in 2002.  PTA’s  TP20  was  relocated  through  mechanical  stripping;  however,  numerous  utility  lines  in  and  around the location of TP20 limited where additional hand excavation could be conducted and  only one unit was excavated.  Additional test pits in Operation B were excavated in locations  suggested by Eglin Archaeologist Joe Meyer.  One 2‐x‐2‐m unit (TP87) was placed southeast of  Building 635 and two more (TPs 88 and 89) were excavated southeast of Building 636.  TPs 93  and 94 (2‐x‐2‐m) were excavated in Operation C just to the southeast of Building 634.  This is  near shovel tests and a 1‐x‐1‐m test unit (TP42) excavated previously by PTA that recovered  shell and lithics.       The last three units (TPs 95, 96, and 97) all measuring 1‐x‐2‐m, were excavated to the north of  Building 635 in an effort to relocate and expose a black sandstone slab that was identified  during monitoring of a water line in 2003.  Using measurements from building corners provided  by Eglin Archaeologist Joe Meyer, as well as GPR data collected from GPR Grid 44, two possible  locations were identified.  The hydraulic excavator was used to remove overburden in the area  of the GPR anomaly and TP96 was excavated with negative results.  The anomaly turned out to  be large tree roots.  Mechanical stripping also was performed in the measured location and  fragments of black sandstone were encountered within a large modern pit.  TPs 95 and 97 were  excavated to better investigate the pit and to determine if any portion of the sandstone feature  remained intact.    All  hand  excavation  was  conducted  using  10‐cm  levels  within  observable    strata    and   archaeological  sediments  were  sifted  through  .64‐cm  (1/4‐in)  mesh screens.  Modern fill  was treated as a single stratigraphic unit and was excavated without screening.  Vertical control  was maintained with line levels using the unit’s highest corner as a datum plane.  Elevations for  each  corner  of  each  unit  were  established  with  the  total  station.    All  artifacts  and  bone  recovered from the .64‐cm mesh screen were collected and placed in 4‐mil plastic zip‐lock bags.   Complete  quahog  shells  (Mercenaria  spp.)  also  were  collected  for  use  in  the  seasonality  analysis.  Shell and charcoal samples for radiocarbon dating were piece‐plotted in situ, as were  diagnostic artifacts when possible.     92

Research Design and Methods 

Figure 3.10.  Excavating TPs 59 and 61 inside demolished Building 588, Operation A. 

Figure 3.11.  Block excavation in Strip Area Ja. 

   93

Bayou Park, 8OK898 

Figure 3.12.  Documenting a shell‐filled pit feature (Feature 36) in Strip Area B. 

  Documentation included the use of standardized excavation forms for notes.  Two wall profiles  were photographed and drawn in each test unit.  In addition, three 50‐x‐50‐cm column samples  were excavated to recover fauna and botanical samples from controlled contexts within the  midden.  The column samples were taken from TP60, TP63, and the west wall of Strip Area L.   These were collected as bulk samples from 10‐cm levels within observable strata.    All features exposed through mechanical stripping or hand excavation were documented with  photographs and plan view drawings, as well as x, y coordinates and elevation using the total  station.  Most features were bisected to document their profiles (Figure 3.12).  Depending on  the type of feature, bisection included hand excavation of one half of the feature and the sifting  of excavated sediments through .32‐cm (1/8‐in) mesh screen; bisection with the hydraulic  excavator; bisection with a shovel.  A few exceptionally large features were sampled using small  (25‐x‐25  cm)  sampling  units  sifted  through  .32‐cm  mesh.      Profiles  were  drawn  and  photographed.    Bulk samples were obtained from a representative sample of features for  detailed faunal and botanical analysis.  These bulk samples typically included the unexcavated  half of hand‐excavated features.      In addition, smaller soil samples were obtained from all pit features and many of the possible  postmolds/postholes for possible chemical analysis.  The sampling method included filling a 4‐x‐ 4‐in plastic zip‐lock bag (approximately 250 ml) with sediment from the center of the feature or  from  different  lenses  within  a  feature.    A  control  sample  also  was  collected  from  the 

94

Research Design and Methods 

surrounding matrix for possible sediment analysis.    In situations where multiple features were  present in close proximity to one another, a single control was collected equidistant from each  feature in the group.  The locations and elevations of soil and control samples were recorded  with the total station and written in the field notes and on various feature maps.   

LABORATORY METHODS    Initial cleaning and processing of artifacts and other materials was conducted in the field using  an on‐site building (Building 584) as temporary lab space.  The indoor space was equipped with  a  computer  station  with  printing  capabilities,  a  standing  rack  of  lipped  .16‐cm  (1/16‐in)  aluminum  drying  screens,  and  space  for  rough  sorting  and  storage.    A  washing  and  wet  screening station was established outdoors with water provided by a spigot alongside the  building and saw horses to support the aluminum drying screens (Figure 3.13).   As materials  were delivered to the field lab, they were sorted by provenience, then washed and allowed to  air  dry.      Once  dry,  materials  were  separated  by  general  artifact  classes  (lithics,  ceramics  [including baked‐clay objects], bone, shell, historic, and miscellaneous) and rebagged. All bags  were assigned sequential Field Specimen (FS) numbers according to collection unit (i.e., by test  unit, strip area, feature, etc.).   Artifacts were counted and weighed using an Ohaus Scout Pro  SP402,  and  all  data  and  provenience  information  were  entered  into  a  Microsoft  Access  database developed by Eglin AFB cultural resource personnel.       Preliminary processing of column samples and features also was conducted in the field. Column  sample material was wet‐screened through .16‐cm (1/16‐inch) mesh and allowed to air dry  before rebagging.  For botanical analysis, a five‐liter sample was taken from the bulk samples of  26 selected features.   The remainder of each feature was wet‐screened through .16‐cm mesh,  air‐dried, and rebagged for potential faunal analysis.       Reanalysis of previously recovered lithic artifacts and baked‐clay objects was performed at the  Eglin curation facility prior to field work (March 2011).  In‐house analyses of lithic, ceramic,  bone, and shell artifacts and faunal remains recovered from form 2011‐2012 data recovery  began after field work was completed in February 2012.  Diagnostic artifacts that required  detailed analysis or were planned for photographs were pulled from general provenience bags  and placed in individual zip‐lock bags with the appropriate provenience information written on  their exteriors.  Chronometric samples, sediment samples, botanical samples, and clam shell  samples  were  submitted  to  appropriate  specialists.    The  reader  is  referred  to  individual  chapters related to each of these materials for details of specific analysis methods.   

NAGPRA‐RELATED HUMAN REMAINS    Two human teeth were identified during analysis of the contents of Feature 44, a large refuse  pit.  Once the teeth were positively identified as human, they were separated from the rest of  the feature material and placed in a zip‐lock plastic  bag  and  then  within  a  small  acid‐free     95

Bayou Park, 8OK898 

Figure 3.13.  Water screening bulk feature sample through  .16‐cm (1/16‐inch) mesh screen.

archival quality box, to await further notification from Eglin AFB regarding their treatment.  The  teeth were handled by only a single individual, SEARCH faunal analyst Brian Worthington.     The  Eglin  Cultural  Resources  Manager,  Mark  Stanley,  was  notified  by  SEARCH  Principal  Investigator,  Robert  Austin,  about  the  discovery  and  a  written  report  was  submitted.     Consultation between Eglin AFB and NAGPRA representatives from the appropriate federally  recognized tribes was initiated by Mr. Stanley.  The teeth were wrapped in unbleached paper,  packed in a tightly sealed container, and shipped to Mr. Stanley.  After consultation with the  tribes, the teeth were reburied on site.      All unanalyzed features and column samples were scanned for any additional human remains.   None were identified.    

CURATION    All cultural material, samples, and associated paperwork were prepared for long‐term storage  in accordance with 36CFR79, Curation of Federally Owned and Administered Archaeological  Collections  and  Eglin  AFB’s  Curation  of  Archaeological  Collections  Requirements  (dated  September 2011).   All materials are permanently curated at Eglin Air Force Base under Task  Order number CR‐11‐0051.   

96

Research Design and Methods  Appendix 3.1.  Grid coordinates and elevation data for SEARCH hard points.  Elevations  Data Points  N  E  Comments  (m)  Bldg. 591, SE corner  1020.28  1010.01  3.72  PTA point  Bldg. 591, SW corner  1029.32  994.52  3.63  PTA point  Bldg. 632, SE corner  1033.84  1084.66  4.14  PTA point  Bldg. 632, SW corner  1048.19  1063.32  4.22  PTA point  Bldg. 633, NE corner  1081.28  1171.02  3.92  PTA point  Bldg. 633, NW corner  1111.86  1125.97  3.85  PTA point  Bldg. 634, SE corner  987.98  1147.01  3.39  PTA point  Bldg. 634, SW corner  1005.19  1121.44  3.36  PTA point  Bldg. 635, outside side corner, NE  1017.87  1227.63  3.42  Bldg. 635, SE corner  986.81  1227.25  3.43  PTA point  Bldg. 635, corner near 1000N/1200E  996.29  1213.22  3.45  Concrete Sewer Marker  996.00  964.00  3.30  PTA datum, NE corner of  Inverness & Flagler; elevation is  PTA's  Fire hydrant 1, top  981.24  1270.56  3.80  Fire hydrant 2, top  925.71  1233.88  3.65  Fire hydrant 3, top, NE of Bldg. 633  1078.32  1173.68  4.35  Fire hydrant 4, top  1126.76  1126.02  4.51  Fire hydrant 5, top  1172.57  1104.33  4.01  Fire hydrant 6, top  1085.60  1021.90  4.26  Green electric box, NW corner  960.66  1233.08  3.22  Light Pole (LP) 1  914.13  1224.13  2.91  LP2  998.24  998.08  3.95  LP3‐Area A  1012.01  1023.91  4.05  LP3‐Area B  965.33  1271.80  3.10  LP5  1038.76  1064.23  4.09  LP6  994.36  1049.71  3.56  LP6  1047.65  1047.53  4.03  LP7  962.88  1042.32  3.39  LP8  1006.36  1087.43  3.46  LP9  1032.22  1069.90  3.85  LP13  970.77  1093.44  3.31  LP14  1012.79  1111.12  3.61  LP15  1034.15  1127.99  3.55  LP16  957.10  1123.22  3.44  LP18  962.90  1156.62  3.17  LP19  1002.53  1172.18  3.36  LP20  938.47  1165.55  3.08  LP21  971.59  1188.22  3.39  LP22  997.41  1200.96  3.40  LP23  928.23  1190.00  3.01  LP24  1070.83  1176.47  3.43 

97

Bayou Park, 8OK898 

Data Points 





LP26  LP60(?), NE corner  LP87(hand written)  LP‐No #  Manhole cover, SE corner concrete  pad  Power pole, spike at base 

1022.55  939.28  1047.95  1133.22  912.28 

1226.16  1252.28  1198.49  1090.28  1206.35 

1068.33 

1213.60 

Elevations  (m)  3.25  3.08  3.25  3.37  2.69 

Comments 

  3.10  3.53  3.23  2.69  3.88  3.28  2.78  3.05  3.06  2.80  4.48  2.87  3.33 

Top of Stake (TOS)  SEARCH datum  1000.00  1000.00  PTA 1000N/1200E  1000.20  1199.74  TOS  PTA 1000N/1300E  1000.45  1299.95  TOS  Manhole Cover  1126.40  1122.31  N of Bldg. 633, PTA Datum 1  Manhole Cover  1069.73  1173.67  PTA Datum 2  Rebar, possible PTA datum  920.76  1248.94  Ground surface  Telephone pole anchor, NE side  1147.16  1046.85  Temp BM‐PVC cap  978.12  1232.38  Temp BM‐PVC cap  1021.71  1255.35  Across from Bldg. 636  Water lift, top  1074.92  1035.63  Water meter, bolt on side  1035.26  1237.30  Water meter, bolt on side, N of  1099.11  1172.45  Bldg. 633    Water meter, top  1145.43  1127.28  4.80  White concrete wall, NW corner  1114.38  1018.63  3.51  NW side of Flagler St.  a  SEARCH coordinates based on 1000N, 1000E steel rebar located at UTM E549051 N3371632 (Zone 16 WGS84). 

98

Type  Number 

Year 

ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST  ST 

1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993 

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  26  27  28 

Appendix 3.2.  Original and converted coordinates for PTA shovel tests and test pits.  PTA Coordinates  SEARCH Coordinatesa, b Notes  North  East  North  East  ‐‐  ‐‐  923.48  1344.64  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  904.71  1346.09  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  893.48  1324.81  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  907.75  1322.30  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  920.84  1323.23  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  948.07  1320.85  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  977.81  1295.74  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  951.11  1296.40  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  927.71  1299.57  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  906.17  1300.63  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  888.19  1300.76  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  886.61  1276.44  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  913.30  1275.64  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  932.07  1274.85  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  960.49  1274.06  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  984.94  1280.53  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  1015.11  1260.97  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  1007.31  1247.81  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  975.19  1249.13  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  955.23  1249.13  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  937.38  1249.79  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  915.84  1249.00  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  887.69  1248.73  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  863.90  1251.91  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  873.41  1228.77  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  915.18  1221.90  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  936.59  1221.37  Location estimated from Campbell and Meyer 1993:Figure 3  ‐‐  ‐‐  953.38  1221.24  Location estimated from Campbell and Meyer 1993:Figure 3 

Type  Number  ST  29  ST  30  ST  31  ST  32  ST  33  ST  34  ST  35  ST  36  ST  43  ST  44  ST  45  ST  46  ST  47  ST  48  ST  49  ST  50  ST  51  ST  52  ST  53  ST  54  ST  55  ST  56  ST  57  ST  58  ST  59  ST  60  ST  61  ST  62  ST  63  ST  64 

Year  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993  1993 

PTA Coordinates  SEARCH Coordinatesa, b ‐‐  ‐‐  935.01  1207.89  ‐‐  ‐‐  915.58  1207.23  ‐‐  ‐‐  891.91  1206.04  ‐‐  ‐‐  847.44  1184.34  ‐‐  ‐‐  842.76  1155.44  ‐‐  ‐‐  825.24  1089.80  ‐‐  ‐‐  854.50  1002.10  ‐‐  ‐‐  874.60  936.10  ‐‐  ‐‐  997.40  1203.53  ‐‐  ‐‐  1004.00  1172.60  ‐‐  ‐‐  982.06  1172.34  ‐‐  ‐‐  962.77  1174.32  ‐‐  ‐‐  941.75  1152.64  ‐‐  ‐‐  962.37  1152.77  ‐‐  ‐‐  937.65  1130.96  ‐‐  ‐‐  957.74  1132.29  ‐‐  ‐‐  975.45  1132.95  ‐‐  ‐‐  974.00  1096.20  ‐‐  ‐‐  997.92  1096.86  ‐‐  ‐‐  1025.31  1102.68  ‐‐  ‐‐  1056.37  1105.45  ‐‐  ‐‐  1053.08  1132.95  ‐‐  ‐‐  982.88  1080.74  ‐‐  ‐‐  1003.90  1081.27  ‐‐  ‐‐  995.97  1056.81  ‐‐  ‐‐  1038.53  1056.15  ‐‐  ‐‐  1008.52  1033.02  ‐‐  ‐‐  1071.04  1058.40  ‐‐  ‐‐  1127.35  1055.09  ‐‐  ‐‐  1128.94  1105.45 

Notes  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3 

Type  Number  ST  65  ST  66  ST  67  ST  68  ST  69  ST  77  ST  78  ST  79  ST  80  ST  81  ST  82  ST  83  ST  84  ST  85  ST  86  ST  87  ST  88  ST  89  ST  90  ST  91  ST  103  ST  104  ST  105  ST  96‐1  ST  96‐2  ST  96‐3  ST  96‐4  ST  96‐5  ST  96‐6  ST  96‐7 

Year  1993  1993  1993  1993  1993  1994  1994  1994  1994  1994  1994  1994  1994  1994  1994  1994  1994  1994  1994  1994  1994  1994  1994  1996  1996  1996  1996  1996  1996  1996 

PTA Coordinates  SEARCH Coordinatesa, b ‐‐  ‐‐  1019.10  992.05  ‐‐  ‐‐  892.17  1072.26  ‐‐  ‐‐  904.20  1051.62  ‐‐  ‐‐  928.28  1018.02  ‐‐  ‐‐  970.48  957.17  ‐‐  ‐‐  1044.48  1490.24  ‐‐  ‐‐  1025.51  1447.24  ‐‐  ‐‐  1000.90  1422.63  ‐‐  ‐‐  967.69  1382.61  ‐‐  ‐‐  934.20  1357.14  ‐‐  ‐‐  968.87  1327.46  ‐‐  ‐‐  996.04  1298.42  ‐‐  ‐‐  1034.02  1265.00  ‐‐  ‐‐  1071.76  1231.61  ‐‐  ‐‐  1107.05  1204.23  ‐‐  ‐‐  1093.28  1335.82  ‐‐  ‐‐  1103.53  1359.97  ‐‐  ‐‐  1162.50  1329.10  ‐‐  ‐‐  1182.30  1360.40  ‐‐  ‐‐  1194.60  1383.10  ‐‐  ‐‐  1102.50  1463.20  ‐‐  ‐‐  1066.49  1376.50  ‐‐  ‐‐  1104.79  1198.11  ‐‐  ‐‐  1030.40  1244.10  ‐‐  ‐‐  1039.10  1227.90  ‐‐  ‐‐  1059.70  1217.20  ‐‐  ‐‐  1061.70  1199.20  ‐‐  ‐‐  1107.80  1188.40  ‐‐  ‐‐  1131.00  1166.60  ‐‐  ‐‐  1131.70  1156.60 

Notes  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Campbell and Meyer 1993:Figure 3  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Location estimated from Mathews et al. 1994:Figure 35  Estimated location based on Mikell et al. 1996:Figure 14  Estimated location based on Mikell et al. 1996:Figure 14  Estimated location based on Mikell et al. 1996:Figure 14  Estimated location based on Mikell et al. 1996:Figure 14  Estimated location based on Mikell et al. 1996:Figure 14  Estimated location based on Mikell et al. 1996:Figure 14  Estimated location based on Mikell et al. 1996:Figure 14 

Type  Number  ST  96‐8  ST  96‐9  ST  109  ST  110  ST  111  ST  112  ST  113  ST  114  ST  115  ST  116  ST  117  ST  118  ST  119  ST  120  ST  121  ST  122  ST  123  ST  124  ST  125  ST  126  ST  127  ST  128  ST  129  ST  130  ST  131  ST  132  ST  133  ST  134  ST  135  ST  136 

Year  1996  1996  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  2002  2002  2002  2002  2002  2002  2002 

PTA Coordinates  SEARCH Coordinatesa, b ‐‐  ‐‐  888.90  1225.20  ‐‐  ‐‐  889.70  1192.10  ‐‐  ‐‐  1003.13  974.80  ‐‐  ‐‐  1007.95  977.73  ‐‐  ‐‐  1013.82  981.38  ‐‐  ‐‐  1020.17  984.90  ‐‐  ‐‐  1025.93  988.78  ‐‐  ‐‐  1021.17  995.94  ‐‐  ‐‐  1015.77  992.76  ‐‐  ‐‐  1009.42  988.89  ‐‐  ‐‐  1003.90  985.01  ‐‐  ‐‐  998.85  981.60  ‐‐  ‐‐  994.54  987.46  ‐‐  ‐‐  999.83  991.11  ‐‐  ‐‐  1005.23  994.87  ‐‐  ‐‐  1011.81  999.10  ‐‐  ‐‐  1017.22  1002.74  ‐‐  ‐‐  1013.45  1009.10  ‐‐  ‐‐  1002.00  999.80  ‐‐  ‐‐  998.55  998.41  ‐‐  ‐‐  989.38  995.01  ‐‐  ‐‐  1008.18  1000.65  ‐‐  ‐‐  995.61  1003.23  1000.00  1000.00  1004.95  999.39  1005.00  1013.00  1008.47  1018.11  1005.00  1022.00  1008.05  1023.30  1023.00  1040.00  1023.72  1038.23  1040.00  1052.00  1040.72  1050.23  1056.00  1066.00  1054.92  1060.95  1031.00  1084.00  1076.32  1074.71 

Notes  Estimated location based on Mikell et al. 1996:Figure 14  Estimated location based on Mikell et al. 1996:Figure 14  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  Estimated from PTA field notes, 11/19/1999  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002 

Type  Number  ST  137  ST  138  ST  139  ST  140  ST  141  ST  142  ST  143  ST  144  ST  145  ST  146  ST  147  ST  148  ST  149  ST  150  ST  151  ST  152  ST  153  ST  154  ST  155  ST  156  ST  157  ST  158  ST  159  ST  160  ST  161  ST  162  ST  163  ST  164  ST  165  ST  166 

Year  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002  2002 

PTA Coordinates  SEARCH Coordinatesa, b 1120.00  1112.00  1114.05  1101.02  1047.00  1143.00  1148.04  1129.23  1131.00  1156.00  1132.04  1142.23  1114.00  1171.00  1115.04  1157.23  1098.00  1186.00  1099.04  1172.23  1084.00  1203.00  1085.04  1189.23  1071.00  1219.00  1072.04  1205.23  1056.00  1234.00  1057.04  1220.23  1040.00  1248.00  1042.37  1232.13  1023.00  1265.00  1024.04  1251.23  1010.00  1280.00  1011.04  1266.23  974.00  1288.00  989.07  1267.61  956.00  1276.00  965.98  1261.16  939.00  1262.00  948.98  1247.16  925.00  1949.00  940.07  1228.61  922.00  1240.00  928.82  1230.05  932.00  1228.00  938.82  1218.05  933.00  1256.00  942.98  1241.16  949.00  1269.00  958.98  1254.16  966.00  1282.00  975.98  1267.16  995.00  1290.00  1010.07  1269.61  927.00  1228.00  933.82  1218.05  927.00  1232.00  933.82  1222.05  932.00  1224.00  938.82  1214.05  1006.00  1008.00  1007.99  1009.79  1009.00  1030.00  1009.72  1028.23  1014.00  1033.00  1014.72  1031.23  1019.00  1036.00  1019.19  1034.86  1027.00  1043.00  1027.72  1041.23  1032.00  1046.00  1032.72  1044.23 

Notes  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002 

Type  Number  ST  167  ST  168  ST  169  ST  170  ST  171  ST  172  ST  173  ST  174  ST  175  ST  176  ST  177  ST  178  ST  179  ST  180  ST  181  ST  182  ST  183  ST  184  ST  185  ST  186  ST  187  ST  188  ST  189  ST  190  ST  191  ST  192  ST  193  ST  194  ST  195  ST  196 

Year  2002  2002  2002  2002  2002  2002  2002  2002  2007  2007  2007  2007  2007  2007  2007  2007  2008  2008  2008  2008  2008  2008  2008  2008  2008  2008  2008  2008  2008  2008 

PTA Coordinates  SEARCH Coordinatesa, b 1009.00  1035.00  1013.31  1032.92  1009.00  1025.00  1013.20  1028.10  1027.00  1048.00  1024.33  1042.95  1027.00  1038.00  1028.04  1037.71  933.00  1256.00  942.98  1241.16  943.00  1265.00  952.98  1250.16  952.00  1272.00  961.98  1257.16  929.00  1253.00  936.98  1238.16  ‐‐  ‐‐  965.67  1078.15  ‐‐  ‐‐  954.73  1085.98  ‐‐  ‐‐  987.62  954.36  ‐‐  ‐‐  927.79  1046.34  ‐‐  ‐‐  895.05  1094.76  ‐‐  ‐‐  931.82  1164.14  ‐‐  ‐‐  993.47  1212.83  ‐‐  ‐‐  990.56  1217.06  ‐‐  ‐‐  1024.62  1267.43  ‐‐  ‐‐  1050.55  1244.41  ‐‐  ‐‐  1078.86  1221.39  ‐‐  ‐‐  1107.05  1204.23  ‐‐  ‐‐  1130.60  1176.90  ‐‐  ‐‐  1149.51  1183.29  ‐‐  ‐‐  1156.91  1206.58  ‐‐  ‐‐  1129.40  1229.59  ‐‐  ‐‐  1103.73  1252.88  ‐‐  ‐‐  1076.48  1275.37  ‐‐  ‐‐  1049.23  1298.12  ‐‐  ‐‐  1028.86  1290.45  ‐‐  ‐‐  1052.14  1270.08  ‐‐  ‐‐  1077.54  1248.12 

Notes  PTA field notes 8/28/2002; relocated SEARCH 2012  PTA field notes 8/28/2002; relocated SEARCH 2012  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  PTA field notes 8/28/2002  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio 

Type  Number  ST  197  ST  198  ST  199  ST  200  ST  201  ST  202  ST  203  ST  204  ST  205  ST  206  ST  207  ST  208  ST  209  ST  210  ST  211  ST  212  ST  213  ST  214  ST  215  ST  216  ST  217  ST  218  ST  219  ST  220  ST  221  ST  222  ST  223  ST  224  ST  225  ST  226 

Year  2008  2008  2008  2008  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009  2009 

PTA Coordinates  SEARCH Coordinatesa, b ‐‐  ‐‐  1103.73  1224.83  ‐‐  ‐‐  1130.46  1202.08  982.00  1232.00  982.80  1232.90  986.00  1226.00  986.80  1226.90  1173.50  1116.50  1174.30  1117.44  1144.50  1133.50  1145.30  1134.44  1113.50  1166.50  1114.30  1167.44  1096.50  1150.00  1097.30  1150.94  1093.00  1149.50  1093.80  1150.44  1094.50  1152.50  1095.30  1153.44  1090.50  1151.50  1091.30  1152.44  1011.50  1169.00  1012.30  1169.94  1011.50  1163.50  1012.30  1164.44  1012.00  1159.50  1012.80  1160.44  1001.50  1137.50  1002.30  1138.44  1004.00  1137.30  1004.80  1138.24  1006.90  1141.00  1007.70  1141.94  1011.20  1141.70  1012.00  1142.64  1044.90  1163.80  1045.70  1164.74  1042.50  1166.50  1043.30  1167.44  1039.00  1160.50  1039.80  1161.44  1036.10  1160.30  1038.90  1161.24  1031.90  1157.40  1032.70  1158.34  1031.60  1156.30  1032.40  1156.24  1076.70  1157.90  1077.50  1158.84  1074.20  1154.50  1075.00  1155.44  1095.60  1121.80  1096.40  1122.74  1094.80  1127.10  1095.60  1128.04  1092.20  1130.90  1093.00  1131.84  1080.70  1100.30  1081.50  1101.24 

Notes  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  Estimated from Morehead, Campbell, and Aubuchon 2008: folio  PTA field notes 2008  PTA field notes 2008  Campbell, Morehead, et al. 2009:151‐152  Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152

Type  Number  ST  227  ST  228  ST  229  ST  230  ST  231  ST  232  ST  233  ST  234  ST  235  ST  236  ST  237  ST  238  ST  239  ST  240  ST  241  ST  242  ST  243  ST  244  ST  245  ST  246  ST  247  ST  248  ST  249  ST  250  ST  251  ST  252  ST  253  ST  254  ST  255  ST  256 

Year  2009  2009  2009  2009  2009  2009  2009  2009  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010  2010 

PTA Coordinates  SEARCH Coordinatesa, b 1076.60  1097.60  1077.40  1098.54  1072.80  1095.00  1073.60  1095.94  1055.60  1081.70  1056.40  1082.64  1052.80  1081.80  1053.60  1082.74  1072.80  1096.10  1073.60  1079.64  1071.50  1080.70  1072.30  1082.04  1076.20  1084.30  1077.00  1085.24  1077.80  1081.30  1078.60  1082.24  988.30  1239.20  985.01  1244.29  984.40  1242.20  980.14  1248.55  980.20  1244.90  974.38  1251.92  976.00  1242.00  968.81  1251.35  972.10  1239.40  964.91  1248.68  967.90  1236.90  960.95  1245.71  964.20  1234.70  957.55  1243.07  963.30  1231.10  958.09  1238.11  965.30  1228.20  962.36  1231.98  969.10  1225.90  971.12  1228.82  966.50  1224.10  966.74  1226.26  963.50  1222.20  962.11  1223.46  975.60  1246.40  968.47  1255.25  972.10  1244.10  964.51  1253.49  966.70  1243.80  960.47  1253.79  964.30  1238.90  958.09  1250.26  954.80  1235.80  951.94  1235.31  959.30  1229.40  952.20  1230.10  942.10  1229.90  940.95  1227.47  932.70  1232.70  929.36  1232.85  938.30  1238.50  938.37  1240.04  944.80  1244.80  945.56  1245.52 

Notes  Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Campbell, Morehead, et al. 2009:151‐152 Morehead et al. 2011:Appendix C  Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C Morehead et al. 2011:Appendix C

Type  Number  TP  1  TP  2  TP  3  TP  4  TP  5  TP  6  TP  7  TP  8  TP  9  TP  10  TP  11  TP  12  TP  13  TP  14  TP  15  TP  16  TP  17  TP  18  TP  19  TP  20  TP  21  TP  22  TP  23  TP  24  TP  25  TP  26  TP  27  TP  28  TP  29  TP  30 

Year  1996  1996  1996  1997  1997  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  1998  2002  2002  2007  2007  2007  2007  2007  2007  2007  2007  2007  2007 

PTA Coordinates  SEARCH Coordinatesa, b ‐‐  ‐‐  1004.46  1242.95  ‐‐  ‐‐  947.82  1272.90  ‐‐  ‐‐  949.33  1259.51  ‐‐  ‐‐  883.55  1196.06  ‐‐  ‐‐  861.21  1180.95  ‐‐  ‐‐  1002.00  999.80  ‐‐  ‐‐  1001.40  999.10  ‐‐  ‐‐  999.10  1002.70  ‐‐  ‐‐  1000.50  1002.70  ‐‐  ‐‐  1001.40  1000.50  ‐‐  ‐‐  1000.60  1000.00  ‐‐  ‐‐  1002.00  998.40  ‐‐  ‐‐  1003.50  1000.00  ‐‐  ‐‐  1000.00  999.30  ‐‐  ‐‐  1000.00  1000.70  ‐‐  ‐‐  1004.20  1000.70  ‐‐  ‐‐  1004.90  1000.00  ‐‐  ‐‐  997.70  1002.70  1026.00  1041.00  1024.55  1040.55  927.00  1252.70  933.63  1238.30  ‐‐  ‐‐  975.46  1079.04  ‐‐  ‐‐  945.64  1061.01  ‐‐  ‐‐  941.59  1060.98  ‐‐  ‐‐  1081.94  1065.15  ‐‐  ‐‐  1095.92  1096.57  ‐‐  ‐‐  1081.47  1124.22  ‐‐  ‐‐  1054.36  1154.04  ‐‐  ‐‐  993.03  1209.19  ‐‐  ‐‐  1080.90  1128.70  ‐‐  ‐‐  964.70  1067.30 

Notes  Estimated location based on PTA field notes 1996  Estimated location based on PTA field notes 1996  Estimated location based on PTA field notes 1996  Estimated location based on Campbell and Mathews 1997  Estimated location based on Campbell and Mathews 1997  Relocated SEARCH 2012  Relocated SEARCH 2012  Relocated SEARCH 2012  Relocated SEARCH 2012  Relocated SEARCH 2012  Relocated SEARCH 2012  Calculated from SEARCH relocation data  Relocated SEARCH 2012  Calculated from SEARCH relocation data  Calculated from SEARCH relocation data  Relocated SEARCH 2012  Relocated SEARCH 2012  Calculated from SEARCH relocation data  PTA field notes 8/28/2002; relocated SEARCH 2012  PTA field notes 8/28/2002; relocated SEARCH 2012  Estimated from PTA field notes 2007  Estimated from PTA field notes 2007  Estimated from PTA field notes 2007  Estimated from PTA field notes 2007  Estimated from PTA field notes 2007  Estimated from PTA field notes 2007  Estimated from PTA field notes 2007  Estimated from PTA field notes 2007  Estimated location based on Morehead et al. 2008:Appendix A  Estimated location based on Morehead et al. 2008:Appendix A 

Type  Number  Year  PTA Coordinates  SEARCH Coordinatesa, b Notes  TP  31  2008  1024.00  1223.00  1029.64  1225.74  PTA field notes 2008  TP  32  2008  1024.00  1219.00  1030.79  1219.90  PTA field notes 2008  TP  33  2008  1027.00  1225.00  1034.36  1229.01  PTA field notes 2008  TP  34  2008  1028.00  1221.00  1037.32  1221.44  PTA field notes 2008  TP  35  2008  1022.00  1221.00  1026.14  1223.05  PTA field notes 2008  TP  36  2008  985.00  1229.00  990.13  1234.69  PTA field notes 2008  TP  37  2008  987.00  1225.00  987.80  1225.90  PTA field notes 2008  TP  38  2009  ‐‐  ‐‐  1052.14  1270.08  Estimated location based on Bourgeois et al. 2009:Figure 22  TP  39  2009  1126.50  1116.50  1127.30  1117.44  Campbell, Morehead, et al. 2009:151‐153 TP  40  2009  1081.00  1076.00  1081.80  1076.94  Campbell, Morehead, et al. 2009:151‐153 TP  41  2009  1098.50  1151.50  1099.30  1152.44  Campbell, Morehead, et al. 2009:151‐153 TP  42  2009  1010.00  1172.00  1010.80  1172.94  Campbell, Morehead, et al. 2009:151‐153 TP  43  2009  1005.30  1139.10  1006.10  1140.04  Campbell, Morehead, et al. 2009:151‐153 TP  44  2009  1034.30  1157.70  1035.10  1158.64  Campbell, Morehead, et al. 2009:151‐153 TP  45  2009  1079.20  1149.70  1080.00  1150.64  Campbell, Morehead, et al. 2009:151‐153 TP  46  2009  1093.10  1127.40  1093.90  1128.34  Campbell, Morehead, et al. 2009:151‐153 TP  47  2009  ‐‐  ‐‐  1075.00  1097.04  Campbell, Morehead, et al. 2009:151‐153 TP  48  2009  ‐‐  ‐‐  1053.90  1081.64  Campbell, Morehead, et al. 2009:151‐153 TP  49  2009  1112.50  1166.50  1113.30  1167.44  Campbell, Morehead, et al. 2009:151‐153 TP  50  2010  940.4  1220.60  939.25  1219.53  Morehead et al. 2011:Appendix C  TP  51  2010  939.7  1215.20  938.00  1215.79  Morehead et al. 2011:Appendix C TP  52  2010  960.9  1233.00  954.32  1241.73  Morehead et al. 2011:Appendix C TP  53  2010  966  1229.50  963.45  1236.13  Morehead et al. 2011:Appendix C TP  54  2010  967.5  1231.50  966.25  1239.30  Morehead et al. 2011:Appendix C TP  55  2010  992.2  1234.50  991.17  1242.00  Morehead et al. 2011:Appendix C TP  56  2010  967.5  1237.00  966.56  1237.48  Morehead et al. 2011:Appendix C TP  57  2010  968.5  1231.00  967.47  1237.96  Morehead et al. 2011:Appendix C a  Because it was not always clear which corner of a shovel test or test pit was used to designate a specific unit’s coordinates, all SEARCH coordinates  are  assumed to be center points.  b  SEARCH coordinates based on 1000N, 1000E steel rebar located at UTM E549051 N3371632 (Zone 16 WGS84). 

4

   

STRATIFICATION AND CHRONOLOGY 

         

  In this chapter we discuss the stratification and radiometric chronology at Bayou Park.   The  basic sequence of strata at the site was established during PTA’s block excavation in 1998  and a subsequent survey of a utility line corridor in 2002 (Campbell et al. 1998; Meyer et al.  2002).    Samples  for  radiocarbon  assays  were  collected  during  1998  and  2002  and  these  provided  a  basic  time  frame  for  the  site’s  primary  occupation.    SEARCH’s  data  recovery  project  in  2011‐2012  served  to  refine  and  build  on  these  initial  interpretations.    Long  profiles  in  several  north‐south  and  east‐west  trenches  allowed  for  the  development  of  a  detailed site‐wide stratigraphy and a better understanding of site formation processes and  the  relationships  between  midden  deposition  and  natural  soil  deposition  and  geomorphology.  Twelve radiocarbon assays, coupled with the four dates obtained by PTA,  indicate occupation between 4210‐3690 BP (3210‐1817 cal BC) with additional occupations  at 1990 BP (cal 48 BC‐AD 72) and 250 BP (cal AD 1522‐modern).  These dates correspond  primarily  with  the  Late  Archaic  period,  or  what  is  referred  to  by  Thomas  and  Campbell  (1991,  1993)  as  the  Gulf  Formational  period;  one  date  corresponds  with  the  Middle  Woodland period, and one is possibly post‐Contact.  Temporally diagnostic artifacts (baked‐ clay objects, microliths, Florida Archaic Stemmed projectile points) indicate that the earliest  occupation was by participants in the Elliott’s Point culture.  A few sherds of Weeden Island  and  Fort  Walton  pottery  indicate  later  occupations  during  the  Woodland  and  Mississippi  periods, while deeply buried lithic artifacts suggest a pre‐Late Archaic occupation.  None of  these approached the intensity of the Elliott’s Point occupation, however.   

DESCRIPTION OF STRATA    Prior  to  entering  the  field,  the  original  strata  descriptions  from  various  PTA  reports  were  summarized and a master sequence was prepared for use during excavation.  This master  sequence,  which  is  shown  in  Table  4.1,  essentially  followed  PTA’s  descriptions  with  the  following  modifications.    The  midden  stratum  was  divided  into  two  major  substrata  IVa  (sand midden) and IVb (shell midden) to correspond with PTA’s Ab1 and Ab2 designations.   During  excavation,  a  lighter,  less  organic,  but  still  identifiable  stratum  was  identified  and  labeled  IVc.    Stratum  V,  the  immediate  submidden  stratum  was  retained,  but  the  various  soil horizons visible below this stratum, particularly in Operations B and C, were collapsed  into  a  single  stratum  designation,  Stratum  VI.    The  reason  for  this  last  decision  is  that  horizon  formation  within  the  site’s  sandy  non‐midden  sediments  is  typically  the  result  of  pedogenesis rather than anthropomorphic processes.  Therefore, any artifacts found within  these  lower  strata  either  were  deposited  much  earlier  than  the  formation  of  visible  horizons, or had migrated to their current position via bioturbation (see Leigh 1998, 2001;  Thulman 2012; Wilder et al. 2001).  In either case, the separate excavation of these horizons  

Bayou Park, 8OK898    Table 4.1.  Strata correlations between representative shovel tests and test pits, midden area.  SEARCH  Descriptions 

PTA Strata Descriptions  TP9  Depth in  Horizon  cmbs  0‐12  Fill 

ST163  Depth in  Horizon  cmbs  0‐22  Fill 

TP19  Depth in  Horizon  cmbs  0‐50  Fill 

Strata 

Horizons 



O/Ap/Fill 

II 

Ab 

12‐21 

Ap 

22‐38 

Ab 

50‐58 

Ab 

III 

AC/(Fill?) 

21‐31 

AC 

‐‐ 

‐‐ 

58‐108 

AC 

IVa 

Ab 

31‐58 

Ab1 

‐‐ 

‐‐ 

108‐150+ 

Ab 

IVb 

Ab 

32‐53 

Ab2 

38‐58 

Ab2 

‐‐ 

‐‐ 

IVc 

Ab  C1 (leach  zone)  C2 

‐‐ 

‐‐ 

‐‐ 

‐‐ 

‐‐ 

‐‐ 

38‐63 

Ab3 

58‐63 

Ab3 

‐‐ 

‐‐ 

55‐70 



63‐100 



133‐150+ 



V  VI 

  as  stratigraphically  distinct  collection  units  would  not  have  resulted  in  useful  contextual  information.  Stratum VI was, however, excavated in arbitrary 10‐cm levels (see Chapter 3).   The strata designations in the first column of Table 4.1 were used throughout the field work  and  are  referenced  throughout  the  remainder  of  the  report.    Descriptions  of  these  stratigraphic units are presented below.  Profile summaries are presented in Figures 4.1 and  4.2 and representative test pit profiles are presented in Figures 4.3‐4.9.  Detailed profiles of  strip areas are provided in Appendix 4.   

  Stratum I  Stratum  I  is  a  composite  of  the  modern  organic  (O)  and  mineral  (Ap)  horizons  and  the  underlying  fill,  which  for  the  purposes  of  excavation  were  lumped  together  as  a  single  stratum.  The modern topsoil (Ia) consisted of very dark gray (10YR3/1) to gray (10YR5/1),  loose, medium‐grained sand mixed with organic material and small grass roots.   The fill (Ib)  is primarily a homogeneous clay or sandy clay that ranges in color from red (2.5YR4/6, 5/6,  5/8)  to  reddish‐brown  (2.5YR4/8)  to  reddish‐yellow  (5YR7/8).    It  is  extremely  hard  and  compact, especially when dry, has a blocky structure, and an abrupt lower boundary.  The  thickness of the clay fill is variable ranging from a few centimeters to as much as a meter;  however,  on  average,  it  is  between  20  and  40  cm  thick.    The  clay  fill  was  commonly  encountered in Operations A and C, and occasionally in Operation B.    Lenses  of  sand  fill  (Ic)  and  isolated  fill  deposits  also  occur  which  appear  to  be  related  to  specific construction events or the installation of utilities.  These fill deposits usually consist  of  sand,  sand  and  rubble,  sand  and  small  rounded  pebbles,  sand  and  clay  fragments,  or  shell.  Color is extremely variable and includes very pale brown (10YR7/4, 8/2), pale brown  (10YR6/3),  brown  (10YR4/3,  5/3),  dark  yellowish‐brown  (10YR4/4,  4/6),  yellowish‐brown  (10YR5/6),  light  yellowish‐brown  (10YR6/4),  very  dark  gray  (10YR4/2),  very  dark  grayish‐ brown (10YR3/2), and white (10YR8/1).  PTA reports modern intrusions containing sand fill

110 

 

Figure  4.1.  Generalized north‐south profile through Operation A based on profiles in Strip Areas E, D, and K.  Vertical scale exaggerated and distances between profiled sections are not to scale. 

    Stratification and Chronology 

  

111

Figure  4.2.  Generalized east‐west profile through Operations A, C, and B based on profiles in Strip Areas.  Vertical scale exaggerated and distances between profiled sections are not to scale. 

Bayou Park, 8OK898 

 

112 

   

 

Stratification and Chronology 

Figure 4.3.  East wall profile, Test Pit 59, Operation A.

Figure 4.4.  East wall profile, Test Pit 60, Operation A. 

113

Bayou Park, 8OK898   

Figure 4.5.  West wall profile, Test Pit 62, Operation A.

Figure 4.6.  West wall profile, Test Pit 63, Operation A.

114 

   

 

Stratification and Chronology 

Figure 4.7.  South wall profile, Test Pit 85, Operation A.

  extending to depths greater than one meter in isolated areas, particularly near buildings or  within their footprints (e.g., Campbell, Morehead, et al. 2009:77‐90; Morehead, Campbell,  and  Aubuchon  2008:64‐74),  but  most  of  the  isolated  fill  deposits  encountered  during  the  2011‐2012 project were not as extensive.  That exception was to the north of Building 635  where TPs 95, 96, and 97 encountered sand fill extending to about 95 cmbs.  Fill deposits in  Operation B tended to consist of sand, sand and gravel, or sand and shell (Rangia cuneata)  rather than red clay (see Figure 4.8).    As  Stratum  I  was  not  screened  when  hand  excavated,  it  is  not  possible  to  discuss  artifact  content  except  for  materials  that  were  observed  as  work  progressed.    These  typically  included pieces of building rubble, cement fragments, glass fragments and occasional whole  bottles, PVC pipe fragments, bullets and bullet casings, iron fragments, nails, and occasional  marine shells presumably mixed into the fill from the underlying midden. 

115

Bayou Park, 8OK898   

Figure 4.8.  South wall profile, Test Pit 88, Operation B.

 

Stratum II  Stratum  II was  not  consistently  observed,  perhaps  because  it  was removed  in  some  areas  during site grading and the deposit of fill.  Where present in Operations A and C, it consisted   of light gray (10YR6/1), grayish‐brown (10YR5/4), or black (10YR2/1) fine to medium sand or  loamy sand immediately underlying the reddish clay.  It was usually rather thin (2‐10 cm),  but  sometimes  was  as  much  as  20‐25  cm  in  thickness.    It  was  typically  loose  to  slightly  compact  with  a  relatively  clear  lower  boundary.    Stratum  II  contained  occasional  modern  artifacts  (glass,  metal,  rubble)  and  marine  shell  fragments,  as  well  as  evidence  of  burning  (see Figure 4.7).  In Operations B and C, it was more variable and occasionally displayed a  distinct  buried  surface  horizon  (see  Figure  4.9).    More  common,  however,  was  a  zone  of  sand lenses of various colors or a mottled sand stratum.    

Stratum III    This  stratum  occurred  only  in  Operation  A.    It  typically  consisted  of  yellowish‐brown  (10YR5/4,  5/6)  to  dark  yellowish‐brown  (10YR4/4)  fine  to  medium  sand,  occasionally  mottled, sometimes with very dark grayish‐brown (10YR3/2) or dark brown (10YR3/3) fine  to medium sand.  It varied in thickness from a few centimeters to as much as 45 cm, and in  some  places  was  entirely  absent.    Stratum  III  contained  prehistoric  artifacts  and  faunal  material, but in relatively low quantities compared to Stratum IV.  It also contained pieces of  building rubble in a few locations.   

116 

   

 

Stratification and Chronology 

Figure 4.9.  South wall profile, Test Pit 94, Operation C.

Stratum IV  Stratum IV is the prehistoric midden, an anthropogenically derived deposit that consists of  organic‐stained  sediments,  marine  shell,  animal  bone,  and  artifacts.    The  thickness  of  this  stratum varied from between 20 and 45 cm.  PTA divided Stratum IV into two substrata, one  containing mostly sand and the other dominated by shell, and this practice was continued  during the SEARCH project.  Stratum IVa was characterized by very dark brown (10YR2/2),  brown  (10YR3/3,  10YR4/3),  dark  grayish‐brown  (10YR4/2),  and  very  dark  gray  (10YR3/1),  slightly compact fine to medium sand while Stratum IVb was dark brown (10YR2/2), black  (10YR2/1),  or  very  dark  gray  (10YR3/1)  in  color  and  composed  of  slightly  compact  fine  to  medium sand and abundant marine shell.  The lower boundaries of these substrata tended  to be abrupt and irregular.  The mechanical stripping revealed that Stratum IVb was not a  true  stratum  but  instead  consisted  of  discontiguous  deposits  of  shell  within  the  larger 

117

Bayou Park, 8OK898   

midden deposit.  This was seen in both plan view and profiles (see Figures 4.1, 4.2, 4.4, and  4.6).    Stratum IVc was identified in TPs 58, 59 (see Figure 4.3) and 61 and in the floor of Strip Area  C.  This substratum was characterized by brown (10YR5/3), yellowish‐brown (10YR5/4), and  gray  (10YR5/1),  slightly  compact  fine  to  medium  sand  with  observable  charcoal  flecking.   Again, this was not a distinct stratum per se, but appeared to be an extension of the midden  deposit restricted to its periphery where the discard of organic materials (and hence organic  staining)  was  not  as  intense.    The  lower  boundary  of  Stratum  IVc  was  irregular  and  conformable with the underlying Stratum VI.  Thickness varied from 10 to 25 cm.   

Stratum V  Stratum  V  was  present  only  in  Operation  A,  usually  in  association  with  Stratum  IV  (see  Figures  4.1,  4.2,  4.4,  and  4.6).    It  is  characterized  by  a  mottled  appearance  with  dark  grayish‐brown  (10YR4/2)  or  dark  yellowish‐brown  (10YR4/4),  slightly  compact  fine  to  medium sand mottled with yellowish‐brown (10YR5/4) sand.  Marine shell and artifacts are  present in this stratum, but are not as abundant as in Stratum IV.  Beneath Stratum IVc, it  consists  of  gray  (10YR5/1)  or  yellowish‐brown  (10YR5/4)  sand  with  charcoal  flecking  (see  Figure  4.3).    Stratum  V  is  generally  about  20  cm  thick  with  a  lower  boundary  that  is  relatively  smooth  and  conformable  with  Stratum  VI.      It  is  best  observed  in  areas  where  Stratum IV is well developed.  This and the conformable nature of the lower boundary and  yellowish‐brown mottling suggest that Stratum V represents a zone of organic and mineral  leaching  from  the  overlying  midden  into  the  underlying  yellowish‐brown  sediments  of  Stratum VI.     

Stratum VI  As  mentioned  above,  Stratum  VI  represents  a  composite  of  the  naturally  occurring  soils  within  the  Bayou  Park  site.    In  Operation  A,  it  underlies  Stratum  IV  (the  midden)  and  Stratum V while in Operation C it underlies Stratum I.  It is characterized by yellowish‐brown  (10YR5/4),  light  yellowish‐brown  (10YR6/4),  brownish‐yellow  (10YR6/6),  and  yellow  (10YR7/6)  medium‐grained,  slightly  compact  sand.    Small  pockets  of  white  (10YR8/1)  clay  were sometimes encountered at depths of 1.25 to 1.5 meters below ground surface.  This  horizon  sequence  is  similar  to  Foxworth  soil  as  described  by  the  National  Resource  Conservation  Service  (https://soilseries.sc.egov.usda.gov/OSD_Docs/F/FOXWORTH.html)  (NRCS  2012).    In  Operation  A,  this  stratum  contained  numerous  features  but  artifact  content was low except near its contact with Strata IV or V.  The base of this stratum was  not reached in Operation A, but thicknesses of the exposed sediments ranged between 60  and 120 cm.    In Operation B, Stratum VI immediately underlies Stratum I and/or Stratum II and exhibits  the  horizonation  associated  with  Kureb  (https://soilseries.sc.egov.usda.gov/OSD   Docs/K/KUREB.html)  or  Resota  (https://soilseries.sc.egov.usda.gov/OSD_Docs/R/RESOTA.  html) soils (NRCS 2012).  This stratum consisted of a relatively thin (15‐30 cm) E horizon of 

118 

   

 

Stratification and Chronology 

white  (10YR8/1)  to  light  gray  (10YR7/1)  fine  to  medium  sand  underlain  by  an  poorly  developed B horizon or B/C horizon (10YR4/6, dark yellowish‐brown sand) about 15‐40 cm  thick, followed by a thick (50‐90 cm) C horizon (10YR6/6, brownish‐yellow sand).  Artifacts  contained within this stratum in Operation B consisted primarily of lithic waste flakes and  occasional tools.  Only a few features were identified but no midden (Stratum IV).    

Laboratory Analysis of Sediment Samples  In addition to the field descriptions, soil samples were obtained from Strata III through VI in  TPs  58  and  60  in  order  to  better  characterize  the  sediments  that  make  up  the  site.    All  samples  were  obtained  from  the  east  wall  profile.    The  samples  from  TP  60  were  subsequently  examined  in  the  laboratory  under  a  70x  Stereo  Zoom  microscope.    After  describing  their  content  and  color  while  wet,  the  samples  were  weighed  and  placed  in  a  conventional oven for 16 hours at a temperature of 230˚ F.  After removal from the oven,  the  samples  were  reweighed  and  the  difference  in  weight  between  wet  and  dry  samples  was  recorded  as  a  measure  of  the  moisture  content  of  the  sediment.    Each  sample  was  tested  with  dilute  (10%)  HCL  to  test  for  the  presence  of  carbonates.    This  information  is  presented in Table 4.2.    For  the  most  part,  the  laboratory  descriptions  and  analysis  are  consistent  with  the  field  descriptions in terms of color and content.  The main difference is in the characterization of  the  sand  fraction.    Field  observations  consistently  classified  the  sand  as  fine  to  medium‐ grained, but examination under a microscope indicated a substantial amount of coarse and  very  coarse  quartz  grains  in  all  samples.    In  all  cases,  the  grains  were  poorly  sorted  and  subangular to subrounded in shape.  Moisture content was very low, which probably is due  to  the  excessively  drained  nature  of  the  soils.    Also  of  interest  was  the  absence  of  HCL  reaction  in  Stratum  IVb,  which  contains  abundant  marine  shell.    The  HCL  solution  was  applied  to  the  quartz  fraction,  avoiding  contact  with  shell  fragments  that  contain  calcium  carbonate.  The absence of any reaction to the HCL is unusual given the length of time since  deposition  of  the  shell.    It  suggests  that  that  carbonate  leaching  into  the  surrounding  sediments  from  the  shell  has  not  been  extensive  or  that  carbonates  were  rapidly  leached  out of the excessively drained sands.    

Stratigraphic Interpretation  Operation A and parts of Operation C are covered with clay fill to varying depths, with the  shallowest occurrence in the open field in the northwest corner of the site and the deepest  to  the  southeast  of  Building  591.    The  clay  fill  appears  to  represent  a  fill  event  that  was  designed to flatten out the topography.  The fact that it appears in test units placed within  the footprints of Buildings 588 and 591 indicates that it pre‐dates the construction of these  buildings,  which  were  built  for  the  prison  camp  during  the  early  1970s  (Campbell  et  al.  2009:68).  Numerous historic features (large square postmolds, trash pits, refuse deposits,  utility  lines)  encountered  underneath  the  fill  are  related  to  military  use  of  the  property  during the 1940s and 1950s.  Historic features also have been identified by PTA in various  locations within 8OK898 (e.g., Bourgeois et al. 2009; Campbell, Morehead et al. 2009).  

119

    Table 4.2.  Characteristics of soil samples from TP 60.  Wet  Dry  Moisture  HCL  FS#  Color (wet)  Description  Wt. (g)  Wt. (g)  Content (%)  Reaction  Fine to Medium grained quartz sand with occasional  Coarse quartz grains and charcoal, poorly sorted,  59  III  13  10YR4/3, brown  81.1  78  3.8%  None  quartz subangular to subrounded  Fine to Medium grained quartz sand, with occasional  Coarse quartz grains and shell fragments, rare  10YR3/2, very dark  charcoal fragments,  poorly sorted, quartz subangular  60  IVb  25  grayish‐brown  70.3  68.3  2.8%  None  to subrounded   Fine to Coarse grained quartz sand with occasional  2.5Y3/2, very dark  Very Coarse quartz,  poorly sorted, subangular to  61  IVa  36  grayish‐brown  85  82.7  2.7%  None  subrounded  Fine to Medium grained quartz sand, with occasional  Coarse to Very Coarse grains, poorly sorted,   62  V  53  10YR4/3, brown  83.8  82.2  1.9%  None  subangular to subrounded  10YR6/4, yellowish‐ Fine to Coarse quartz sand, occasional charcoal,  63  VI  86  brown  90.4  89.2  1.3%  None  poorly sorted, quartz subangular to subrounded   a  Centimeters below unit datum (southwest corner). Depth  Stratum  (cmbud)a 

   

 

Stratification and Chronology 

The  isolated  fill  deposits  that  were  encountered  throughout  Operations  A,  B,  and  C  are  related to specific construction events or the installation of utilities.   Many of these post‐ date the deposition of the clay fill as they penetrate into or through the clay.  Meyer et al.  2002:43)  indicate  that as  many  as  7  fill  episodes  have  been  identified ranging  from  a  few  centimeters too as much as a meter in thickness.  A major fill event was identified by PTA in  the northern portion of Operation A within Buildings 632, 633, and 634.  Excavations near  the buildings encountered deep deposits of sand fill ranging from 65 to 95 cm in thickness  (Campbell et al. 2009:77‐83).  Several historic refuse pits, round postholes, and an asphalt  slab were encountered.  Deep fill deposits extend into Operation C where SEARCH exposed  sand  fill  extending  to  about  95  cmbs.    Fill  in  Operation  B  is  more  variable  and  appears  to  represent more recent activities.      The  fill  overlies  what  PTA  has  interpreted  as  a  buried  plow  zone  (Ap)  and  which  we  designated as Stratum II (see Table 4.1).  Because it was not frequently identified during our  investigations, Stratum II is difficult to interpret.  The interpretation of this stratum by PTA  as  buried  topsoil  or  ground  surface  is  plausible  and  is  not  contradicted  by  our  field  observations.      Stratum III has been interpreted by PTA as a transitional A/C horizon consisting of sand that  accumulated  over  and  buried  the  midden  after  prehistoric  abandonment  of  the  site  (e.g.,  Campbell et al. 1998:48; Meyer et al. 2002:45).  Campbell et al. (1998:48) suggest that our  Strata  II  and  III  reflect a  “redevelopment  of  the  Foxworth sand  above the  midden”;  i.e.,  a  new period of pedogenesis and the development of a new soil within the parent C material.      There are several reasons to question this interpretation.  Although Campbell et al. (1998)  do  not  mention  what  the  source  of  Stratum  III  sediments  might  have  been,  the  NRCS  indicates  that  Foxworth  Series  soils  formed  in  sandy  marine  or  eolian  sediments.   Deposition  of marine sediments could have occurred as a result of higher seas levels or a  major  storm  surge,  but  there  is  no  evidence  for  sea  levels  higher  than  1  m  after  3000  BP  (see Chapter 2) and the highest elevation of the midden (Stratum IV) is 3.61 m.   A storm  surge of that magnitude also is unlikely since the mainland is protected by barrier islands.   Storm  surges  from  hurricanes  have  overtopped  these  islands  in  the  past,  and  the  period  from  3400  to  1100  BP  was  apparently  quite  active  for  hurricanes  (Lui  and  Fearn  2000).    Prehistoric  hurricane  overwash  deposits  at  Western  Lake  near  Grayton  Beach  range  in  thickness from a few millimeters to over 10 cm, but most are only a few centimeters thick;  some contain small shell fragments (Liu and Fearn 2000:241‐242).  Stratum III sands are as  much  as  45  cm  thick  in  some  locations  and,  except  for  fragments  of  large  marine  shells  displaced from the underlying midden, no shells of small marine invertebrates are present.   Multiple storm deposits are unlikely since there is no evidence of buried A horizons within  the Stratum III sands which would have formed between storm events.    Alternatively,  Stratum  III  may  be  eolian  in  origin.    However,  the  last  period  of  substantial  eolian  deposition  in  the  southeastern  Coastal  Plain  was  likely  during  the  Hypsithermal,  a 

121

Bayou Park, 8OK898   

period  of  dry  climate,  reduced  flooding  of  interior  rivers,  and  lowered  lake  levels  which  generally is considered to have occurred between about 9000 to 5000 years ago (Delcourt  and  Delcourt  1993;  Holloway  2002:211‐226;  Otvos  2004b;  Stout  and  Spackman  2002:227‐ 235; Watts 1971), or before Bayou Park was being occupied by Elliott’s Point people.  Some  researchers  (e.g.,  Leigh  1998,  2001)  would  argue  that  no  significant  eolian  deposition  has  occurred in the Coastal Plain since the end of the Pleistocene.      After about 4000 BP, rainfall increased, the rate of sea level rise slowed, and modern rainfall  and vegetation regimes became established (Fearn and Cohen 1984:430; Watts and Hansen  1988).    With  the  region  heavily  wooded,  the  landscape  presumably  stabilized  and  wind  erosion decreased.  These conditions would seem to make it unlikely that significant eolian  deposition  would  have  occurred  after  the  Bayou  Park  midden  was  abandoned,  which  appears to have happened no earlier than the Mississippi period (AD 1000‐1500), although  the  exact  date  of  abandonment  is  not  known.    On  the  other  hand,  Johnson  and  Fredlund  (1993:71)  argue  that  broad  expanses  of  open  pine  forest  to  the  northwest  of  Bayou  Park  would have contained enough open areas beneath the pine canopy to allow wind to pass  through creating the conditions necessary for erosion of the unconsolidated surface sands  and  their  subsequent  redeposition  at  downwind  locations.    In  support  of  this  hypothesis,  they  point  to  a  prehistoric  feature  at  8OK387  about  22.5  km  northwest  of  Bayou  Park,  which was buried beneath 65 cm of eolian sand and was radiocarbon dated at 860 +120, ‐ 150 BP.    An alternative explanation is that Stratum III represents a fill event, perhaps coeval with the  deep fill event identified in Buildings 632, 633, and 634 and in TPs 95, 96, and 97.   A marine  shell from Stratum III was radiocarbon dated at 4140 ± 30 BP (cal 2381‐2141 cal BC).  This is  one  of  the  older  dates  obtained  from  the  site  and  it  is  the  only  date  that  is  out  of  stratigraphic  sequence,  i.e.,  it  was  obtained  from  a  higher  elevation  (18  cm  below  unit  datum  in  TP60)  than  its  date  would  suggest.    This  suggests  that  it  was  displaced  from  its  original context and became mixed with Stratum III when it was being deposited.  In several  other locations modern rubble appeared to be mixed with what was identified in the field  as Stratum III sediments.  It is also quite possible that Stratum III represents a combination  of eolian sands and sand fill, with the field identification of these depositional events made  difficult by modern land modifications.     Stratum  IV  is  the  prehistoric  midden  deposit  that  is  present  only  in  the  northwestern  portion  of  Operation  A.    PTA  distinguishes  between  the  sand  midden  (Ab1)  and  the  shell  midden  (Ab2)  while  the  original  ground  surface  on  which  the  midden  accumulated  is  designated  Ab3  (see  Table  4.1).    These  correspond  to  our  Strata  IVa,  IVb,  and  V,  respectively.   As discussed above, the shell deposits do not represent a distinct stratum or a  buried  surface  (Ab),  but  instead  are  discontiguous  lenses  and  deposits  within  the  larger  midden stratum.  In addition, we identified a light‐colored, charcoal impregnated extension  along the eastern periphery of the midden.      122 

   

 

Stratification and Chronology 

Campbell  et  al.  (1998:48)  indicated  that  Stratum  V  represents  the  original  ground  surface  prior to midden formation and assigned this an Ab (buried A horizon) designation.  This is  probably  true,  but  the  color  of  Stratum  V  is  the  result  of  the  leaching  of  organics  and  minerals from the overlying midden into the underlying yellowish‐brown sediments of the  preexisting  Stratum  VI,  the  C  horizon  of  the  Foxworth  Series.    The  fact  that  Stratum  V  is  visible and most strongly expressed only where Stratum IV is present tends to support this  interpretation.   

SITE CHRONOLOGY   

Radiocarbon Dates  Twelve samples (8 charred wood, 4 marine shell) were submitted to Beta Analytic, Inc. for  AMS  dating.    All  of  the  marine  shell  samples  were  quahog  clam  (Mercenaria  spp.).    The  results are presented in Table 4.3.  Also included in this table are the four dates on marine  shell  (species  unknown)  obtained  by  PTA.    All  of  the  conventional  radiocarbon  ages  have  been  calibrated  using  CALIB  6.01  (Hammer  et  al.  2001).    Figure  4.10  is  a  plot  of  the  calibrated ages and Figure 4.11 shows the calibrated radiocarbon dates.     The charcoal sample from Feature 38, a large wood post, returned the youngest age at 250  ±  30  BP.    The  intercept  crosses  the  calibration  curve  in  several  locations  ranging  from  modern (‐1 to 11 cal BP) to between 428 and 376 cal BP (cal AD 1522‐1574).   There is a 59%   probability that the true age is between 324 and 270 cal BP, or cal AD 1620‐1680, and a 75%  probability that the true age is between 428 and 270 cal BP, or cal AD 1522‐1680.  However,  no  definitive  contact‐period  artifacts  have  been  recovered  from  the  site  despite  the  numerous  investigations  that  have  been  carried  out  over  the  past  20  years.    While  the  probability  that  the  sample  is  modern  is  only  3%,  it  remains  a  possibility  and  so  interpretations of a contact‐period occupation of the site should be viewed with caution.     Feature 51, another large post, returned a radiocarbon age of 1990 ± 30, or 1997‐1878 cal  BP  (48  BC‐AD  72).    This  would  situate  the  feature  within  the  Late  Early  Woodland  period  (see  Chapter  2).    The  local  prehistoric  culture  would  have  been  Deptford,  although  no  definitive  Deptford  artifacts  have  been  recovered  from  the  site.    Feature  51  also  closely  resembles Feature 38, which dates much later in time, and both are at approximately the  same  elevation  (3.20  m  and  3.27  m  amsl,  respectively).      Because  both  features  provide  dates for which there are no supporting artifacts, their usefulness in establishing the timing  of post‐Archaic occupations is limited.    The  bulk  of  the  radiocarbon  dates  are  related  to  the  Late  Archaic  period  and  these  are  extremely consistent.  The maximum two‐sigma range of this occupation is 4481‐3766 cal  BP  (2529‐1817  cal  BC),  a  period  of  about  seven  centuries.  Using  only  the  median  dates  calculated by CALIB 6.01 yields a minimum range of 4301‐3971 cal BP (2352‐2002 cal BC), or  a little over three centuries.   

123

Lab No. 

Provenience 

Material 

 

Table 4.3.  Radiocarbon dates from Bayou Park.  Measured Age  13C/12C  Conventional  Calibrated  (rcy BP)  (‰)  Age  Age (2σ)a    270 ± 30 BP  ‐26.1   250 ± 30 BP  428‐376 (.15) 366‐363 (
Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.