The Rye Creek Project: Archaeology in the Upper Tonto Basin Volume 1: Introduction and Site Descriptions Mark D. Elson Douglas B. Craig Contributions by William H. Doelle Gary Huckleberry Stephen H. Lekson Deborah L. Swartz Center for Desert Archaeology Anthropological Papers No. 11 The Rye Creek Project: Archaeology in the Upper Tonto Basin Volume 1: Introduction and Site Descriptions Mark D. Elson Douglas B. Craig Contributions by William H. Doelle Gary Huckleberry Stephen H. Lekson Deborah L. Swartz Submitted to Arizona Department of Transportation Phoenix, Arizona Contract No. 88-36 Center for Desert Archaeology Anthropological Papers No. 11 1992 ii Cover Center for Desert Archaeology 3975 North Tucson Boulevard Tucson, Arizona 85716 (602) 881-2244 ABSTRACT The Rye Creek Project involved testing and data recovery at 19 archaeological sites within the Upper Tonto Basin of central Arizona. The project area is situated along a 5.4 mile (8.7 km) stretch of State Route 87, apprOximately 10 miles south of the town of Payson, Arirona, within the boundaries of the Tonto National Forest. The project was undertaken for the Arizona Department of Transportation prior to the realignment and expansion of State Route 87. Thirteen sites were tested and then intensively investigated (Chapters 6, 7, 8, and 9) while six were only tested (Chapter 10). Considerable functional and temporal diversity was present; the sites ranged from small, isolated, single-room masonry structures and larger multiroom pueblos dating to the early Classic period (AD. 1150-13(0), to earlier Preclassic period (AD. 750-1150) sites with subsurface pithouse architecture. Of the intensively investigated sites, seven dated primarily to the Classic period and six dated to the Preclassic period, although a number of these contained both Classic and Preclassic components (Chapter 25). The Preclassic period sites were for the most part more substantial; a methodological analysis of the archaeological signatures of sedentism suggests that the majority of the Preclassic period sites were sedentary in nature (Chapter 26). This contrasts with many of the Classic period sites which appear to have been seasonally occupied field houses, although a larger, more permanently occupied (but severely disturbed) pueblo roomblock was also present. Given the traditional emphasis on Classic period sites in Tonto Basin archaeology (Chapter 3), this project represents one of the most complete investigations of the less well known Preclassic period. The Deer Creek site (AZ 0:15:52) contained 17 pithouses and dated primarily to the Gila Butte phase (AD. 750-850) with a possibly earlier Snaketown phase (AD. 650-750) component (Chapter 7). This is now one of the earliest excavated ceramic period sites within the Tonto Basin. Limited testing on a volunteer basis was also undertaken at Rye Creek Ruin (AZ 0:15:1) which, while probably originating during the early Classic period (AD. 1150-13(0), dates primarily to the late Classic period Gila phase (AD. 1300(1450) (Chapter 27). Rye Creek Ruin is one of the largest permanently occupied sites in the Tonto Basin, containing around 150 masonry rooms and two platform mounds. The site was undoubtedly the focus of the Classic period settlement of the Upper Tonto Basin. The long temporal span of the project area sites, running from the Gila Butte phase (AD. 750-850) through the early Classic period Roosevelt phase (AD. 1150-13(0) (and with the inclusion of Rye Creek Ruin, into the late Classic period Gila phase), allowed for diachronic modeling of prehistoric settlement and subsistence systems within the Upper Tonto Basin (Chapters 4 and 28). Most significantly, the data strongly suggest that the Upper Basin was occupied by an indigenous population who interacted with neighboring populations while remaining culturally discrete. Changes through time in the intensity and direction of the interaction networks are clearly apparent, although the overall intensity is believed to have been relatively limited. The earliest inhabitants, ca. AD. 700-1000, were tied in most closely with Hohokam groups to the south. This is indicated by the dominance of imported Hohokam buffwares (Chapter 12), although in very low frequencies. However, the clear absence of a Hohokam mortuary complex suggests that the inhabitants were not culturally Hohokam. The early inhabitants, particularly during the Gila Butte phase, interred their dead in small, rectangular, daublined crematoriums or primary cremations (Chapter 7). Although a small number of secondary cremations was also present, the use of this method is unlike what is known from contemporaneous Hohokam populations. This practice has now been documented at only three other Preclassic period sites in the Southwest, one additional site in the Upper Tonto Basin and two sites in the White Mountains of Central Arizona. These data suggest the possibility of a sub-Mogollon Rim cultural tradition distinct from neighboring populations (Chapter 28). Sometime shortly after AD. 1000 or 1050, interaction with Hohokam populations ended or was severely curtailed, while interaction with Tusayan populations to the north increased (Chapter 12). The source of this interaction is suggested to be the Flagstaff area, given the dominance of Tusayan whiteware ceramics and the presence of imported argillite artifacts from the Del Rio mines in the Upper Verde Valley. The presence of iv Abstract a relatively high frequency of nonlocal Del Rio argillite during this time is significant, particularly since the project sites are situated within a large, local argillite source area (the Deer Creek source) which was intensively utilized (Chapter 22). The fact that Tusayan whitewares and Del Rio argillite have similar temporal and spatial distributions within project area sites suggests that they are moving within the same networks and probably from a single regional source zone (Chapter 28). Interaction with Tusayan populations appears to have ended by around AD. 1100 or 1150 at the start of the Classic period. Tusayan ceramics are first replaced by Little Colorado whitewares and later by Cibola whitewares, although both wares are present throughout the early Classic period (Chapter 12). Little Colorado whitewares initially dominate the Classic period decorated assemblage but are replaced after AD. 1250 (when production of Little Colorado whitewares ends) by Cibola whitewares. Finally, during the following late Classic period Gila phase (AD. 1300-1450), Cibola whiteware groups continue to dominate the interaction networks, joined by White Mountain and Roosevelt redware producers, and to a lesser extent groups producing Hopi Wares and Winslow Orangewares. Rye Creek Ruin appears to have been initially settled during the early Classic period, although there are some indications of an earlier Preclassic period occupation (Chapters 27 and 28). The most intensive occupation was during the late Classic period Gila phase. Given the paucity of sedentary sites within the project area and the Upper Tonto Basin in general during the late Classic period, it is suggested that the Gila phase was a period of population aggregation into Rye Creek Ruin from the surrounding area. The location of Rye Creek Ruin along a natural trade corridor on Rye Creek above the confluence with Tonto Creek, combined with the abundance of intrusive northern ceramic wares (8 wares containing 20 distinct ceramic types were recovered from the sampling of three trash mounds), suggests that Rye Creek Ruin may have served as an important gateway site into the Tonto Basin from points north. Interesting changes in subsistence practices were also documented. Corn agriculture and wild plant gathering combined with limited hunting appears to have been the subsistence base throughout time (Chapters 18, 20, and 21). In fact, the only other cultigen recovered outside of com was a single grain of squash pollen from the Deer Creek site. The dominance of com agriculture and plant gathering in the economy is secondarily supported by the analysis of the ground stone assemblage (Chapter 15). What is perhaps most significant is the prominent role of agave, which currently grows naturally in the Upper Tonto Basin, in the subsistence base. Although agave was used during the earliest periods, it became an increasingly important resource by the late Preclassic period Sacaton phase (AD. 950-1150), increasing from around 20 percent of the recovered plant parts to over 70 percent (Chapter 18). This trend is even more apparent during the early Classic period, where agave accounts for almost 85 percent of the recovered plant parts and was by far the most ubiquitous species. These data suggest possible specialization and perhaps cultivation of this foodstuff. Agave may have been part of an eXChange network between the Upper and Lower Tonto Basin, since it does not grow naturally in the Lower Basin. Interaction between these areas is supported by petrographiC analyses of the plainware and redware ceramic assemblages which suggest that a significant number of vessels are moving from the Lower Basin into the Upper Basin (Chapter 13). Therefore, unlike previous models of Tonto Basin settlement, there appears to be little need to invoke colonization or migration models to account for the initial settlement of the Upper Tonto Basin. Furthermore, occupation within the Upper Basin was more-or-Iess continuous, at least from the Snaketown or Gila Butte phase through the late Classic period; the notion of a Sacaton phase hiatus originally proposed by the Gladwins (1935) can be finally put to rest. How these data relate to the Lower Tonto Basin, which is closer to the Hohokam core area and in a more similar riverine-dominated environmental zone, remains unknown. However, our data suggest that while there may have been limited migration into the Basin from points north and south, the Upper Basin was occupied primarily by an indigenous population who participated in various interaction networks, the nature of which changed through time. ACKNOWLEDGMENTS The testing phase of the Rye Creek Mitigation Project began on October 24, 1988. It is now close to March 1992 as I sit here and write this. Much has happened. Too much to acknowledge all those who have aided not only this project but life in general, since the two are not overly separable. Therefore, while these acknowledgments are basic, giving credit only to people who played a key role in the project per se, I am extremely grateful to all those who have aided and abetted me through this time. To start, the project would not have been successful without the excellent work of the field crew and supervisors. Fieldwork is the most critical aspect of any archaeological project, since without good base-line data, subsequent interpretations have limited value. Deborah Swartz, who was the assistant project director, and Douglas Craig, who graciously agreed to step down to a crew chief position from his normal project director role, deserve special commendation for the smooth running of the project, the happiness of the crew, and the high quality of the data recovered. Both Deb and Doug played key roles during a four-month leave of absence I took after the completion of the fieldwork, and I cannot thank them enough. Working and playing with these two was indeed a pleasure, and I am glad to say that the team has continued onto subsequent projects and will likely continue for some time. The three assistant crew chiefs, Annick Lascaux, Victoria (Tory) Clark, and Danielle Desruisseaux, were also instrumental in the quality of the fieldwork. Annick served as crew chief for the SWAT team, directing the excavation of most of the small sites in her customarily able manner. Finally, Martina Lavelle ran the field lab out of Payson and did an excellent job while serving both as lab director and administrative assistant. And on to the crew, the backbone of any project. The testing-phase crew, who were in the field from October 24 through December 7, 1988, consisted of Lambert Jos~, Joan Lloyd, Jon Parslow, and Christopher Powell, with Archie Allen and Rob Caccio as laborers. The data recovery phase began on April 10 and lasted through August 11, 1989. The crew consisted of Kari Chalker, Jeff Clark, Nicole Croft, H. Scott Goodman, T. Marshall Henderson, Lambert Jos~, Andrea Kayser, Jon Parslow, Brenda Randolph, Terry Samples, Dawn Snell, Ruth Van Dyke, and Rein Vanderpot. The laborers were Archie Allen, Rob Caccio, Kathy Clark, Betsy Flores, Ben Johnson, Shay Mallett, Wally Pottle, Keith Vlastos, Noland Wiggins, and Jeff Ureles. Greg Thompson ably operated the backhoe during the testing and data recovery phases. The crew and laborers did an excellent job both in the dirt and in their notes. Wide-ranging discussions with all of the crew, but particularly with Tory Clark, Terry Samples, Jeff Clark, and Jeff Ureles, were both informative and fun, and have lent much to this write-up. The crew additionally deserves special kudos for managing to get along so well during the months of close-quarters living together in the Payson "fishbowl." Support from the Desert Archaeology (then the Institute for American Research) home office in Tucson was critical, particularly since this was our first large out-of-town project. Key personnel in the smooth running and successful completion of the project included Jennifer Ballard, the officer manager, Linda Mayro and her mid-project replacement Patricia Castalia, the operations directors, and Lisa Eppley, the laboratory director. The report production could not have been accomplished without the perseverance and skilled efforts of Kara Myrick (word processor and report compiler), Linda Gregonis (technical editor), Becky Quayle-Craig (proofreader, compiler, and draftsperson), Ronald Beckwith (draftsperson) and Daniel Snyder (photographer). Catherine Gilman is to be commended for her diligence in compiling references. Other Desert Archaeology staff lent their support and expertise through numerous discussions concerning the archaeology, to which I am most grateful. These people included Allen Dart, James Heidke, Miriam Stark, and Henry Wallace. And last but not least, William Doelle, the head of Desert Archaeology and the Principal Investigator of the project, deserves special praise, not only for his continuous support of the project and myself, but for his insights into Tonto Basin archaeology and archaeology in general. Numerous additional people paid visits to the sites and/or participated in discussions of the project area and Tonto Basin archaeology. Although I did not, of course, agree with everything that was said, the discussions vi Acknowledgments certainly strengthened the final interpretations. Foremost among these is Tonto National Forest archaeologist J. Scott Wood, who is probably one of the most knowledgable people on Tonto Basin archaeology, and certainly one of the most opinionated. Discussions with Scott, as well as his review comments on several drafts of the report, greatly helped clarify my thinking on numerous issues, as well as provided much food for thought. Other insightful comments were provided by David Wilcox, Michael Bremer, Anne Baldwin, Stephen Lekson, Douglas (and Kim and Katy) Mitchell, Glen Rice, Phil Weigand, David Doyel, and Christian Downum. David Abbott provided statistical consulting. I would also like to thank Dudley Varner, director of the Orange County Museum of Natural History, for giving me the position of Visiting Archaeologist during my three months there. I would also like to thank en masse the numerous chapter authors, all 21 of them, for their excellent work. Their insights and interpretations greatly aided the final conclusions and quality of this report. A list of contributors and their institutional affiliations can be found at the end of each volume. The Arizona Department of Transportation, which provided the funding for the project, and particularly ADOT archaeologist Bettina Rosenberg, deserve special mention for handling many of the project logistics and helping things run smoothly. A great deal of archaeology has been done under ADOT auspices the past 20 years or so, and they have contributed much to what we know about Arizona archaeology. Finally, I would like to thank my wife Debbie, as well as Max and Zeke, for sticking with me throughout this process. Debbie, I cannot thank enough. I'd also like to thank Jeannie Webber, Debbie Grunloh, Jack Galvin, Lily Speer, A Trinkle Jones, and Bill Wilson for their support. It's been a long road to haul, but the load is getting lighter and the road smoother. Mark Elson Tucson, March 1992 TABLE OF CONTENTS ABSTRACf .................................................................... iii ACKNOWLEDGMEN1'S ........................................................... v LIST OF FIGURES .............................................................. ~ LIST OF TABLES ................................................................ xxi PART 1: INTRODUCfION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 1 1. INTRODUCfION AND PROJECf SETTING, Mark D. Elson. . . . . . . . . . . . . . . . . . . . . . . . . . . .. PROJECf HISTORY ........................................................ Testing Phase ...................................................... .. Data Recovery Phase .................................................. ORGANIZATION OF THE REPORT ........................................... 3 5 5 5 8 2. ENVIRONMENT, Mark D. Elson and Gary Huckleberry ................................. GEOLOGy .............................................................. PHYSIOGRAPHY, VEGETATION, AND FAUNA ................................ HYDROLOGY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. SOILS ................................................................... CLIMATE ............................................................... GEOMORPHOLOGY AND PEDOLOGY OF THE RYE CREEK PROJECf AREA ..... Geomorphology ..................................................... Pedology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Summary .......................................................... DISTURBANCE FACfORS .................................................. CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 11 11 12 13 14 14 15 15 16 19 19 20 3. PREVIOUS RESEARCH AND CULTURE HISTORY, Mark D. Elson, Stephen H. Lekson, and Douglas B. Craig . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. PREVIOUS RESEARCH IN THE TONTO BASIN ................................ TONTO BASIN CULTURE HISTORy ......................................... Paleo indian and Archaic Periods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Preclassic Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Classic Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ARCHAEOLOGICAL RESEARCH WITHIN THE UPPER TONTO BASIN ............ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 23 23 24 27 27 29 32 37 4. RESEARCH DESIGN, Mark D. Elson and William H. Doelle ............................. DEFINmON OF HISTORIC CONTEXTS ...................................... HISTORIC CONTEXT #1: ADVANCES IN ARCHAEOLOGICAL METHODS -CONTEXTUAL ANALYSIS ........................................... HISTORIC CONTEXT #2: CHRONOLOGY BUILDING ........................... Relative Dating Methods .............................................. Absolute Dating Methods ................. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. HISTORIC CONTEXT #3: SUBSISTENCE AND SETTLEMENT SYSTEMS ............ Subsistence Strategies ................................................. Settlement Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 39 39 40 40 41 42 43 43 44 viii Table Of Contents HISTORIC CONTEXT #4: DEMOGRAPHY AND COMMUNITY ORGANIZATION .... Population Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. The Role of the Household in Community Organization . . . . . . . . . . . . . . . . . . . . . .. Community Structure and Integration ............. . . . . . . . . . . . . . . . . . . . . . . .. HISTORIC CONTEXT #5: EXCHANGE AND INTERAcrION ..................... HISTORIC CONTEXT #6: CULTURAL AFFILIATION ........................... ADDmONAL ISSUES: PROTOHISTORIC AND HISTORIC COMPONENTS .......... 45 45 46 47 47 49 49 SUMMARy .............................................................. 50 5. PROJECT METHODS, Mark D. Elson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • .. TESTING PHASE ......................................................... Brushing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Site Gridding, Mapping, and Surface Collection ............................. Backhoe Trenching and Feature Profiling .................................. Hand-dug Trenching and Wall Clearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Testing Phase Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. DATA RECOVERY PHASE ................................................. Feature and Strata Designations ......................................... Field Methods ...................................................... 51 51 51 51 52 53 53 53 54 55 PART 2: SITE DESCRIPTIONS .................................................... 61 6. THE HARDT CREEK DRAINAGE, Mark D. Elson ................................... 63 AZ. 0:15:96 ..................................................................... 65 EXCAVATION METHODS .................................................. Testing Phase ....................................................... Data Recovery Phase ................................................. FEATURE 1 DESCRIPTION ................................................. ARTIFACT SUMMARIES ................................................... POLLEN AND FLOTATION SUMMARIES ..................................... SITE CHRONOLOGY ...................................................... SITE INTERPRETATION .............................. ........ ............ 65 65 65 67 67 67 67 68 AZ. 0:15:71 .............................................. ....................... 69 EXCAVATION METHODS ............................ ........ .............. Testing Phase .................................. ..... . . . . . . . . . . . . . . .. Data Recovery Phase ........................... .......... ............ FEATURE DESCRIPTIONS ............................ .......... ........... Feature 1 ................................... ............ ........... Feature 2 ................................... .......... ............. Feature 4 .................................... ........... ........... Feature 5 .......................................................... Feature 6 ..................................... ....... .............. ARTIFACT SUMMARIES ............................. ............ .......... POLLEN AND FLOTATION SUMMARIES ................ ... ... ... ............ SITE CHRONOLOGY ...................................................... SITE INTERPRETATION ................................................... 69 69 71 71 71 74 74 74 75 75 76 76 76 AZ. 0:15:70 ..................................................................... 78 EXCAVATION METHODS .................................................. Testing Phase ....................................................... Data Recovery Phase ................................................. FEATURE DESCRIPTIONS ................................................. 78 78 80 80 Table Of Contents ix Feature 1 .......................................................... Feature 2 .......................................................... ARTIFACf SUMMARIES .................................................. POLLEN AND FLOTATION SUMMARIES ..................................... SITE CHRONOLOGY ...................................................... SITE INTERPRETATION .................................................. 80 80 81 81 81 81 TI-IE OVERLOOK SITE: AZ. 0:15:89 ................................................. 82 EXCAVATION METHODS .................................................. Testing Phase .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Data Recovery Phase ................................................. FEATURE DESCRIPTIONS ................................................. Feature 1 .......................................................... Feature 2 .......................................................... ARTIFACf SUMMARIES ................................................... POLLEN AND FLOTATION SUMMARIES ..................................... SITE CHRONOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. SITE INTERPRETATION ................................................. .. 82 82 84 84 84 87 88 88 89 89 7. THE DEER CREEK DRAINAGE ................................................ . 91 THE DEER CREEK SITE: AZ. 0:15:52 (ASM), Deborah L. Swartz .......................... 93 EXCAVATION METHODS ................................................. . 93 Testing Phase ...................................................... . 93 Data Recovery Phase ................................................ . 95 Summary ......................................................... . 96 PITHOUSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 96 Feature 2 .......................................................... 96 Feature 6 ......................................................... 101 Feature 9 ......................................................... 103 Feature 11 ........................................................ 106 Feature 12 ........................................................ 108 Feature 13 ........................................................ 108 Feature 14 ........................................................ 110 Feature 18 ........................................................ 115 Feature 21 ........................................................ 117 Feature 22 ........................................................ 120 Feature 25 ........................................................ 120 Feature 32 ........................................................ 122 Feature 34 ........................................................ 126 Feature 36 ........................................................ 128 Feature 59 ........................................................ 129 Feature 62 ........................................................ 131 Feature 65 ........................................................ 132 EXTRAMURAL FEATURES ............................................... 133 Extramural Surfaces ................................................. 133 Trash Mounds and Trash Areas ........................................ 136 Pits .............................................................. 136 Roasting Pits ...................................................... 138 Extramural Hearths ................................................. 140 Rock Concentrations ................................................ 141 MORTUARY FEATURES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 141 Crematoriums and Primary Cremations ................................... 141 x Table Of Contents Cremations and Inhumations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. APACHE REMAINS ...................................................... ARTIFACTS~llES .................................................. POLLEN AND FLOTATION S~IES .................................... Pollen Data ....................................................... Flotation Data ..................................................... SITE CHRONOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. SITE IN1ERPRETATION .................................................. 154 155 156 156 156 160 161 163 THE HILLTOP SITE - AZ 0:15:53 (ASM), Douglas B. Craig ..... ......................... 165 EXCAVATION METHODS ................................................ . Testing Phase ..................................................... . Data Recovery Phase ............................................... . Summary ........................................................ . PITHOUSES ............................................................ . Feature 1 Feature 6 ........................................................ . Feature 9 ........................................................ . Feature 14 ....................................................... . Feature 15 ....................................................... . MASONRY ROOMS ..................................................... . Feature 5 and Feature 16 ............................................ . EX1RAMURAL FEATURES .............................................. . Pits ............................................................. . Roasting Pit ...................................................... . Rock Alignment ................................................... . Possible Crematoriums .............................................. . ARTIFACT SUMMARIES ................................................. . POLLEN AND FLOTATION SUMMARIES ................................... . Pollen Data ...................................................... . Flotation Data .................................................... . SITE CHRONOLOGY .................................................... . SITE IN1ERPRETATION ................................................. . 165 165 167 167 167 167 170 172 174 174 177 177 180 181 181 181 182 183 185 185 185 185 186 8. THE CLOVER WASH DRAINAGE .............................................. 187 THE CLOVER WASH SITE - AZ 0:15:100 (ASM), Deborah L. Swartz . ..................... 189 EXCAVATION METHODS ................................................ . Testing Phase ..................................................... . Data Recovery Phase ............................................... . Summary ........................................................ . PITHOUSES ........................................................... . Feature 1 Feature 3 Feature 4 ........................................................ . Feature 6 ........................................................ . Feature 12 ....................................................... . EXTRAMURAL FEATURES .............................................. . Pits ............................................................. . Roasting Pits ..................................................... . Rock Clusters ..................................................... . Other Extramural Features ........................................... . 189 189 191 191 191 193 195 198 200 200 202 202 204 204 205 Table Of Contents MORTUARY FEATURES ................................................. ARllFACTS~nES .................................................. POLLEN AND FLOTAll0N SUMMARnES ................................... Pollen Data ....................................................... Flotation Data ..................................................... SIlE CHRONOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. SIlE INTERPRETAll0N ................................................. xi 205 206 206 206 208 208 209 THE REDSTONE SIlE - AZ 0:15:91 (ASM), Douglas B. Craig . . . . . . . . . . . . . . . . . . . . . . . . . . .. 210 EXCAVAll0N ME11IODS ................................................. Testing Phase ...................................................... Data Recovery Phase ................................................ Summary ......................................................... PI1HOUSES ............................................................. Feature 5 ......................................................... Feature 11 ........................................................ EXTRAMURAL FEATURES ............................................... Activity Areas or Ramadas ............................................ Pits .............................................................. Roasting Pits ...................................................... Rock Alignments ................................................... AR11FACT SUMMARnES .................................................. POLLEN AND FLOTATION SUMMARnES .................................... Pollen Data ....................................................... Flotation Data ..................................................... SIlE CHRONOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. SIlE INTERPRETAll0N .................................................. 210 210 211 211 211 211 215 221 221 221 223 223 223 225 225 225 225 226 9. 11IE RYE CREEK DRAINAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 229 11IE ROOTED SITE - AZ 0:15:92 (ASM), Deborah L. Swartz . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 231 EXCAVAll0N ME11IODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Testing Phase ...................................................... Data Recovery Phase ................................................ PI1HOUSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Feature 14 ........................................................ RAMADAS ............................................................. Feature 15 ........................................................ COBBLE STRUCTURE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. MORTUARY FEATURES .................................................. Feature 13 ........................................................ Feature 16 ........................................................ CHECKDAMS ........................................................... ARllFACT SUMMARIES .................................................. POLLEN AND FLOTAll0N SUMMARIES .................................... Pollen Data ....................................................... Flotation Data ..................................................... SIlE CHRONOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. SIlE INTERPRETAll0N ................................................. 231 233 233 235 235 237 237 238 240 240 240 240 241 242 242 242 243 244 THE COBBLE SIlE - AZ 0:15:54 (ASM), Deborah L. Swartz ............................. 245 xii Table Of Contents EXCAVATION MErnODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Testing Phase ...................................................... Data Recovery Phase ................................................ Summary ......................................................... MASONRY STRUCflJRES ................................................. Feature 5 ......................................................... Feature 8 ......................................................... Feature 9 ......................................................... EXTRAMURAl.. FEATURES ............................................... Trash Mounds and Trash Areas ............................•. . ......... Other Extramural Features .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. MORTUARY FEATURES ................................................. ARTIFACf SUMMARIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. POLLEN AND FLOTATION SUMMARIES ................... . ................ Pollen Data ...................................... . ................ Flotation Data ..................................................... SIlE CHRONOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. SIlE INTERPRETATION .................................................. 245 247 248 250 250 250 251 253 257 257 257 261 261 262 262 262 263 263 mE BOONE MOORE SIlE - AZ 0:15:55 (ASM), Douglas B. Craig .. ...................... 265 EXCAVATION MErnODS ................................................. Testing Phase ...................................................... Data Recovery Phase ................................................ Summary ................... . ..................................... MASONRY STRUCTURES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Feature 1 ......................................................... Feature 18 ........................................................ COBBLE-LINED ADOBE PITROOMS ........................................ Feature 5 ......................................................... Feature 6 ......................................................... PI1HOUSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Feature 9 ......................................................... Feature 11 ........................................................ Feature 19 ........................................................ MORTUARY FEATURES ................................. . ................ Feature 3 ......................................................... Feature 7 ......................................................... Feature 8 ............................................... . ......... Feature 17 ........................................................ Feature 21 ........................................................ Feature 23 ........................................................ EXTRAMURAL FEATURES ............................................... Roasting Pits ...................................................... Extramural Hearths ................................................. Homo ........................................................... Pits .............................................................. ARTIFACf SUMMARIES .................................................. POLLEN AND FLOTATION SUMMARIES .................................... Pollen Data ....................................................... Flotation Data ............... . ..................................... SIlE CHRONOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. SIlE INTERPRETATION .................................................. 265 265 266 266 266 266 271 274 274 277 279 279 281 285 287 287 288 290 290 290 291 292 292 292 292 293 293 295 295 295 295 297 Table Of Contents xiii THE COMPACf SITE - AZ 0:15:90 (ASM), Douglas B. Craig ............................. 298 EXCAVATION MElHODS ................................................ Testing Phase ..................................................... Data Recovery Phase ............................................... Summary ........................................................ PI1HOUSES ............................................................ Feature 2 Feature 3 Feature 4 ........................................................ Feature 5 ........................................................ EXTRAMURAL FEATURES .............................................. Homo .......................................................... Roasting Pits ..................................................... ARTIFACf SUMMARIES ................................................. POLLEN AND FLOTATION SUMMARIES ................................... Pollen Data ...................................................... Flotation Data .................................................... SITE CHRONOLOGY .................................................... SITE INTERPRETATION ................................................. 298 298 298 300 300 300 300 . 303 . 306 . 308 . 308 . 308 . 310 . 311 . 311 . 312 . 312 . 312 . . . . . lHE ARBY'S SITE - AZ 0:15:99, Mark D. Elson .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 314 EXCAVATION MElHODS ................................................ . Testing Phase ..................................................... . Data Recovery Phase ............................................... . FEATURE DESCRIPTIONS ............................................... . Feature 1 Feature 2 Feature 3 Feature 4 Feature 5 Feature 6 ........................................................ . ARTIFACf SUMMARIES ................................................. . POLLEN AND FLOTATION DATA ......................................... . Pollen Data ...................................................... . Flotation Data .................................................... . SITE CHRONOLOGY .................................................... . SITE INTERPRETATION ................................................. . 10. TESTING PHASE SITES, Deborah L. Swartz and Mark D. Elson ....................... AZ 0:15:51 (ASM) ....................................................... Testing Methods ................................................... Feature Descriptions ................................................ Artifact Assemblage ................................................ Conclusions ...................................................... AZ 0:15:95 (ASM) ....................................................... Testing Methods ................................................... Feature Description ................................................ Artifact Assemblage ................................................ Conclusions ...................................................... AZ 0: 15:93 (ASM) ....................................................... Testing Methods ................................................... Feature Descriptions ................................................ . . . . . . . . . . . . . . 314 316 316 317 317 320 320 323 324 324 325 326 326 326 326 326 329 330 330 330 330 332 332 332 334 334 334 334 334 336 xiv Table Of Contents AZ. AZ. AZ. AZ. Artifact Assemblage ................................................. Conclusions ....................................................... 0:15:94 (ASM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Testing Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Feature Description ................................................. Artifact Assemblage ................................................. Conclusions ....................................................... 0:15:97 (ASM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Testing Methods .................................................... Feature Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Artifact Assemblage ................................................. Conclusions ....................................................... 0:15:98 (ASM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Testing Methods .................................................... Feature Description ................................................. Artifact Assemblage ................................................. Conclusions ....................................................... 0:15:101 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Testing Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Feature Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Artifact Assemblage ................................................. Conclusions ....................................................... 336 336 336 336 337 338 338 338 340 340 340 341 341 341 341 343 343 343 343 345 345 345 REFERENCES CITED ........................................................... 347 LIST OF CONTRIBUTORS ....................................................... 363 LIST OF FIGURES Figure 1.1. Location of the Tonto Basin in relation to defined prehistoric culture areas . . . . . . . . . . . . 4 Figure 1.2. Location of the Rye Creek Project area within the Tonto Basin . . . . . . . . . . . . . . . . . . . . . 6 Figure 1.3. Location of the 19 archaeological sites investigated as part of the Rye Creek Project ..... 7 Figure 2.1. Schematic diagrams of the geomorphology and stratigraphy for the Upper Tonto Basin (a), and a representative drainage/interfluve along State Route 87 (b) 17 Figure 2.2. Areas and sites disturbed through root-plowing ................................ 21 Figure 3.1. Locations of major archaeological sites and previous project areas within the Tonto Basin .................................................. 2S Figure 3.2. Phase systematics commonly used in the Tonto Basin (modified from Dean 1990; Kidder 1927; and Wood 1986) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 26 Figure 5.1. Strata designations used on the Rye Creek Project .............................. 54 Figure 6.1. Map of project sites included in the Hardt Creek drainage area .................... 64 Figure 6.2. Site AZ 0:15:96 (ASM) ................................................. 66 Figure 6.3. Photo of Feature 1 at AZ 0:15:% .......................................... 68 Figure 6.4. Site AZ 0:15:71 (ASM) ................................................. 70 Figure 6.5. Features 1,2, and 4 at AZ 0:15:71 ......................................... 72 Figure 6.6. Photo of west wall of Feature 1 at AZ 0:15:71 ................................ 73 Figure 6.7. Feature 4, slab-lined pit, at AZ 0:15:71 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 75 Figure 6.8. Site AZ 0:15:70 (ASM) ................................................. 79 Figure 6.9. Site AZ 0:15:89 (ASM) ................................................. 83 Figure 6.10. Feature 1 at AZ 0:15:89 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 85 Figure 6.11. Photo of Feature 1 at AZ 0:15:89 .......................................... 86 Figure 6.12. Photo of east wall of Feature 1 at AZ 0:15:89 ................................. 87 Figure 7.1. Map of project sites included in the Deer Creek drainage area ..................... 92 Figure 7.2. Overall site map of the Deer Creek site (AZ 0:15:52 [ASM]) ..................... 94 Figure 7.3. Habitation area at the Deer Creek site (AZ 0:15:52 [ASM]) ...................... 97 xvi List of Figures Figure 7.4. Feature 2 at the Deer Creek site 100 Figure 7.5. Feature 6 at the Deer Creek site 102 Figure 7.6. Feature 9 at the Deer Creek site 104 Figure 7.7. Feature 11 at the Deer Creek site 107 Figure 7.8. Feature 12 at the Deer Creek site 109 Figure 7.9. Feature 13 at the Deer Creek site 111 Figure 7.10. Feature 14 at the Deer Creek site 112 Figure 7.11. Photograph of Feature 14 at the Deer Creek site. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 113 Figure 7.12. Feature 18 at the Deer Creek site 116 Figure 7.13. Feature 21 at the Deer Creek site 119 Figure 7.14. Feature 22 at the Deer Creek site 121 Figure 7.15. Feature 25 at the Deer Creek site 123 Figure 7.16. Features 32 and 36 at the Deer Creek site ................................... 124 Figure 7.17. Photograph of Feature 32 at the Deer Creek site . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 125 Figure 7.18. Feature 34 at the Deer Creek site ......................................... 127 Figure 7.19. Features 59 and 65 at the Deer Creek site ................................... 130 Figure 7.20. Extramural surfaces Features 20 and 66, superimposed on pithouse Feature 59 at the Deer Creek site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 134 Figure 7.21. Photograph of crematorium Feature 46 ..................................... 142 Figure 7.22. Crematorium Feature 48 at the Deer Creek site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 143 Figure 7.23. Crematorium Features 1 and 70 at the Deer Creek site ......................... 144 Figure 7.24. Crematorium Features 88 and 117 at the Deer Creek site ....................... 145 Figure 7.25. Photograph of crematorium Feature 50 at the Deer Creek site. . . . . . . . . . . . . . . . . . .. 146 Figure 7.26. Photograph of crematorium Features 71 and 85 at the Deer Creek site ............. 147 Figure 7.27. Crematorium Feature 71 at the Deer Creek site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 148 Figure 7.28. Photograph of crematorium Feature 31 at the Deer Creek site ................... 149 Figure 7.29. Crematorium Feature 82 ................................................ 151 Figure 7.30. Photograph of crematorium Feature 52 at the Deer Creek site. . . . . . . . . . . . . . . . . . .. 152 List of Figures xvii Figure 7.31. Crematorium Feature 37 at the Deer Creek site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 153 Figure 7.32. Overall map of the Hilltop site (AZ 0:15:53 [ASM]) ........................... 166 Figure 7.33. Feature 1 at the Hilltop site 169 Figure 7.34. Feature 6 at the Hilltop site 171 Figure 7.35. Feature 9 at the Hilltop site 173 Figure 7.36. Pithouse Features 14 and 15, with possible wing wall, Feature 20 at the Hilltop site ..................................................... 175 Figure 7.37. Masonry structures Features 5 and 16 at the Hilltop site ........................ 178 Figure 7.38. Crematorium area at the Hilltop site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 183 Figure 8.1. Location of sites within the Clover Wash drainage area ......................... 188 Figure 8.2. Overall site map of the Clover Wash site (AZ 0:15:100 [ASM]) .................. 190 Figure 8.3. Area of data recovery at the Clover Wash site (AZ 0:15:100 [ASM]) . . . . . . . . . . . . . .. 192 Figure 8.4. Feature 1 at the Clover Wash site ......................................... 194 Figure 8.5. Photograph of Feature 1 at the Clover Wash site . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 195 Figure 8.6. Feature 3 at the Clover Wash site 197 Figure 8.7. Feature 4 at the Clover Wash site 199 Figure 8.8. Feature 12 at the Clover Wash site ........................................ 201 Figure 8.9. Photograph of roasting pit Feature 13 ....... . .............................. 205 Figure 8.10. Overall site map of the excavated area at the Redstone site (AZ 0:15:91 [ASM]) ..... 212 Figure 8.11. Feature 5 at the Redstone site ............................................ 214 Figure 8.12. Feature 11 at the Redstone site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 217 Figure 8.13. Photograph of Feature 11 at the Redstone site ............................... 218 Figure 8.14. Hearth Features 11-4 and 11-5 within pithouse Feature 11 at the Redstone site ....... 220 Figure 9.1. Location of sites within the Rye Creek drainage area. . . . . . . . . . . . . . . . . . . . . . . . . .. 230 Figure 9.2. Overall site map of the Rooted site (AZ 0:15:92 [ASM)) ....................... 232 Figure 9.3. Area of data recovery at AZ 0:15:92 ....................................... 234 Figure 9.4. Feature 14 at the Rooted site ............................................ 236 xviii List of Figures 239 Figure 9.5. Feature 15 at the Rooted site Figure 9.6. Overall site map of the Cobble site (AZ 0:15:54 [ASM]) ........................ 246 Figure 9.7. Locus B at AZ 0:15:54 ................................................. 249 Figure 9.8. Feature 5 at the Cobble site ............................................. 252 Figure 9.9. Feature 8 at the Cobble site 254 Figure 9.10. Feature 9 at the Cobble site ............................................. 255 Figure 9.11. Photograph of Feature 9 at the Cobble site .................................. 256 Figure 9.12. Petroglyphs within Features 1 and 2 at the Cobble site ......................... 258 Figure 9.13. Feature 1 at the Cobble site: area of disturbed masonry rubble ................... 259 Figure 9.14. Photograph of Feature 1 at the Cobble site .................................. 260 Figure 9.15. Overall site map of the Boone Moore site (AZ 0:15:55 [ASM]) .................. 267 Figure 9.16. Feature 1 at the Boone Moore site ........................................ 269 Figure 9.17. Upper floor of Feature 18 at the Boone Moore site. . . . . . . . . . . . . . . . . . . . . . . . . . .. 272 Figure 9.18. Lower floor of Feature 18 at the Boone Moore site ............................ 273 Figure 9.19. Feature 5 ............................................................ 275 Figure 9.20. Feature 6 at the Boone Moore site ........................................ 278 Figure 9.21. Cobble wall showing reuse of Feature 6 at the Boone Moore site. . . . . . . . . . . . . . . . .. 280 Figure 9.22. Feature 9 at the Boone Moore site ........................................ 282 Figure 9.23. Feature 11 at the Boone Moore site 284 Figure 9.24. Feature 19 at the Boone Moore site ....................................... 286 Figure 9.25. Feature 7 at the Boone Moore site ........................................ 289 Figure 9.26. Feature 21 at the Boone Moore site ....................................... 291 Figure 9.27. Site map of the Compact site (AZ 0:15:90 [ASM]) ............................ 299 Figure 9.28. Feature 3 at the Compact site ............................................ 302 Figure 9.29. Feature 4 at the Compact site 304 Figure 9.30. Feature 5 at the Compact site 307 Figure 9.31. Homo Feature 6 at the Compact site ....................................... 309 List of Figures xix Figure 9.32. Photograph of Feature 6 at the Compact site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 310 Figure 9.33. Overall site map of the Arby's site (AZ 0:15:99 [ASM]) ........................ 315 Figure 9.34. Feature 1 at the Arby's site .............................................. 318 Figure 9.35. Photograph of Feature 1 at the Arby's site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 319 Figure 9.36. Feature 3 at the Arby's site .............................................. 321 Figure 9.37. Photograph of Feature 3 at the Arby's site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 322 Figure 10.1. Site map of AZ 0:15:51 (ASM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 331 Figure 10.2. Overall site map of AZ 0:15:95 (ASM) ..................................... 333 Figure 10.3. Site map of AZ 0:15:93 (ASM) ........................................... 335 Figure 10.4. Overall site map of AZ 0:15:94 (ASM) ..................................... 337 Figure 10.5. Site map of AZ 0:15:97 (ASM) ........................................... 339 Figure 10.6. Overall site map of AZ 0:15:98 (ASM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 342 Figure 10.7. Site map of AZ 0:15:101 (ASM) .......................................... 344 xx List 0/ Figuns LIST OF TABLES Table 1.1. General characteristics of testing and data recovery phase sites ...................... 9 Table 2.1. Qimatological data from Tonto Basin area recording stations between 1952-1972 ...................................................... 15 Table 2.2. Archaeological sites and landform associations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 18 Table 6.1. AZ 0:15:71 artifact totals by feature and stratum ............................... 76 Table 6.2. AZ 0:15:89 artifact totals by feature and stratum ............................... 88 Table 7.1. Feature list for AZ 0:15:52 ............................................... 98 Table 7.2. Metric data from pithouses at AZ 0:15:52 ................................... . 99 Table 7.3. Metric data from sampled extramural pits at AZ 0:15:52 ........................ 137 Table 7.4. Metric data from roasting pits at AZ 0:15:52 ................................. 139 Table 7.5. Metric data from crematoriums at AZ 0:15:52 ................................ 150 Table 7.6. AZ 0:15:52 artifact totals by feature and stratum .............................. 157 Table 7.7. Archaeomagnetic dates recovered from the Deer Creek site (AZ 0:15:52) ........... 162 Table 7.8. Metric data from pithouses at AZ 0:15:53 ................................... 168 Table 7.9. Metric data from pits and roasting pits at AZ 0:15:53 .......................... 180 Table 7.10. Metric data from possible crematoriums at AZ 0:15:53 ......................... 180 Table 7.11. AZ 0:15:53 artifact totals by feature and stratum .............................. 184 Table 8.1. Metric data from pithouses at AZ 0:15:100 ...... . . . . . . . . . . . . . . . . . . . . . . . . . . .. 193 Table 8.2. Metric data from pits and roasting pits at AZ 0:15:100 ......................... 203 Table 8.3. AZ 0:15:100 artifact totals by feature stratum ................................ 207 Table 8.4. Metric data from pits and roasting pits at AZ 0:15:91 .......................... 222 Table 8.5. AZ 0:15:91 artifact totals by feature and stratum .............................. 224 Table 9.1. AZ 0:15:92 artifact totals by feature and stratum .............................. 243 Table 9.2. Metric data from excavated structures at AZ 0:15:54 ........................... 250 Table 9.3. AZ 0:15:54 artifact totals by feature and stratum .............................. 262 xxii List of Tables Table 9.4. Metric data from pithouses, pitrooms, and masonry structures at AZ. 0:15:55 ........................................................ 268 Table 9.5. Metric data from burials at AZ. 0:15:55 ..................................... 288 Table 9.6. Metric data from pits and roasting pits at AZ. 0:15:55 .......................... 292 Table 9.7. AZ. 0:15:55 artifact totals by feature and stratum .............................. 294 Table 9.S. Archaeomagnetic dates recovered from the Boone Moore site (AZ. 0:15:55) ......... 296 Table 9.9. Metric data from pithouses at AZ. 0:15:90 ................................... 300 Table 9.10. AZ. 0:15:90 artifact totals by feature and stratum .............................. 311 Table 9.11. AZ. 0:15:99 artifact totals by feature and stratum .............................. 325 ~ :.,~fi>;·:.~I:::.r;;i:I.::.y::::I! , .i';_:~I>.l 1,': !1 'm~: : : : : : : =m'l ~ .~.~ ».m m ~! ~ ~ ~f:Mf:f:~ I!; i~ ~1 :. ~ : ~:. :.' :. ~: ~ :.~ :~ :. : .:~ : !f: ~! ~1 :~·.~: . i~! t i!!!pMmf:f:~!!~f i! ~! !:. t.:·~:'-.~:~ ~ ~ ~f:~f:~ :~.:;~.~:; :. :.~: ~ i I!I M.~~..i~ : ~1I~wmmmmMmm~m1 ~!J I :~~ ~; I!II L:J =,~ i:. ~: .J~:.~ .~ ~ ~?:~ ~ ~1:.:.!.~ .................... .. :.::.![:.::. ... :.:.::--: i!!!l .. ..... i: ....::i: .:.:. .. . :.i.• r.: . :.·:. .·:.:'. ::,. .: ·. :. :. . .. :i. :'. .:,' ;" 1.:y:::: .. .. ;.:; ...:.:: .... ...... :.. : :. [.: :::i·.! .. imlmmmmwmMwJlll1 ...· :1.:1:: ... ::.... .. ". :!;. :!.::.:.:.:.: .:.:. . !.:.: ,.::!.;. jll~~~mmmmmmmmmmmmmmlllj 1llllnmmmmmmmmilll! PART 1: INTRODUCTION . .. . ::1.t. ..::.. .•..:. .:.:.:.::.: . . .:. . .:. ..:.: . .:. :... 2 InIroduction CHAPTER 1 INTRODUCTION AND PROJECT SETI'ING Mark D. Elson The Tonto Basin is situated in the approximate center of Arizona, in a transition zone bounded by high, rugged mountain ranges to the east and west, plateau uplands to the north, and the hot, arid desert to the south. It is surrounded by, and sometimes included in, the homelands of four defined prehistoric peoples: the Anasazi, Sinagua, Mogollon, and Hohokam (Figure 1.1). It is not surprising, then, to find significant variability within the Tonto Basin, both culturally and environmentally. The environment, well-watered and containing abundant resources, was conducive to the flourishing of a large and dynamic prehistoric population, which culminated between AD. 1150 and 1450 in what traditionally has been defined as the Salado culture. Even after 100 years of investigation and speculation, however, who these people were, where they came from, and how they lived, are questions still very much in the forefront of Southwestern archaeological research and debate. Unlike many areas of the American Southwest, our knowledge of the prehistory of the Tonto Basin can still be said to be in its infancy. This volume presents a report on the investigation of 13 sites in the Upper Tonto Basin excavated as part of the Rye Creek Mitigation Project. Work on this project was funded by the Arizona Department of Transportation (ADOT) and undertaken in coordination with the Tonto National Forest (TNF) and the State Historic Preservation Office (SHPO), based on a memorandum of agreement with the Advisory Council. The Rye Creek Project sites encompass considerable diversity, and range from small, isolated, single-room masonry structures, and larger multiroom pueblos, dating to the Salado period (AD. 1150-1450), to earlier, pre-Salado (AD. 750-1150) sites, with subsurface pithouse architecture. Functional, temporal, and perhaps cultural variability are clearly evident. There are several factors that formed the basis for structuring the research orientation of this project. First, the Tonto Basin lacks many elements that are essential to high quality archaeological research. For example, chronological controls are very poor, only a limited number of archaeological survey and excavation projects have been conducted to date, and much of the literature on the area expends substantial effort trying to establish the cultural affiliation of the local inhabitants. Although this latter concern is clearly a relevant issue, its research payoff is presently limited within the "data short" environment of the Tonto Basin. Furthermore, very little research has been undertaken on the nature of the pre-Salado occupation, or what may perhaps be termed the period of Salado development. The great majority of archaeological investigations have focused on the larger, more visible, and much more impressive pueblo remains of the Salado period. As a result, there are a multitude of opinions, many of them quite diverse and often contradictory, on the prehistory of the Tonto Basin. Given the nature and focus of the archaeological sample, it is apparent that the validity of at least some of the previous archaeological reconstructions remains highly questionable. Clearly, the excavation of 13 sites cannot resolve all of the questions raised by prior research. However, by focusing on a series of fundamental research questions that include chronology, subsistence and settlement systems, community organization, exchange and interaction, and cultural affiliation, along with specialized studies in ceramic petrography and artifact sourcing, we believe that significant progress has been made into unravelling the nature of the prehistoric settlement. Therefore, the research presented here is geared towards an initial, but extremely critical, investment in basic research. This involves building strong chronologies, emphasizing maximal control over archaeological context, examining in detail the systems present within the local prehistoric community, and patiently building a comprehensive data base before pushing toward conclusions of regional scope. The payoffs may not come for 4 Chapter 1 .... . " .. " ".,. .ot" ~~! _. ____ ._. __ .~.!~!L .. __ . _ ___ _ .. _____ .. __ ~_~~~!t~-- i-"0 ARIZONA )1. .,.~ " i"'''·~-· ··~·',,{'·' i I I ,/ ! .f ...:.•••, . ,1'" \ !!..)~.!r._i ,.. I i sr I ANASAZI I ! i \ r··--··_·· k' ... ./ ,~.!!!. , 1"''' - "I~··./·I I I, ~\,,} ......."- - - ...../ l I '- \ 35- If r ~oI'. . . . .' : I \I ! •• s.· " .) ~ ,.. o " " ~...~ - ~ \1-, \~'!'~~~~. ; I >S. ! 0 ; t o -~ -I(/) o:(/) 1200 «« UJ-' u 1100- So ho pm I--------~------- Miami /Hardt Santon Sacaton Sacaton >0: ~ Z 1000- Rooseve I t UJ o PI! UJ (/) ...J 900 « z Santa Cruz Santa Cruz 800- 0 Gila Butte Gila Butte Snaketown Snaketown 9 u 700 600500 1--------------Sweetwater 0: 1-------- UJ UJ Estrella z ? 0 300~--- i- -- - - - - - - ----- - -1------------__ « u 0: BM I! Red Mountain u J: 100 I- - BMm Yah k i 400- 200 PI Archaic Archaic « A.D...L-----Ll--------------..l--------------l----------------' Figure 3.2. Phase systematics commonly used in the Tonto Basin (modified from Dean 1990; Kidder 1927; and Wood 1986). Previous Research And Culture History 27 Paleoindian and Archaic Periods The Paleoindian and Archaic periods are poorly known in the Tonto Basin. Evidence for a Paleoindian occupation includes a basal fragment of a Clovis point reported in association with a small lithic and sherd scatter in the Upper Basin near Payson (Huckelll978), and an isolated Clovis point recovered subsurface by workers at the Fluorspar Mill in the Lower Basin near Punkin Center (Huckell 1982). Several Clovis points are also reported to be in private collections from the Upper Basin, although their original context is unknown (J. Scott Wood, personal communication, 1990). Clovis points generally are dated to between 9500 and 9000 B.C. (Haynes 1968). Archaic sites are more widespread, but the overall density continues to be low. Given the relatively limited amount of archaeological research in the Tonto Basin, however, coupled with the low visibility of Archaic period remains, the recording of even a few Archaic period sites may be meaningful and indicative of a relatively significant occupation. In the Upper Basin, Ciolek-Torrello (1987) excavated six Archaic sites within the uplands of the Mazatzal piedmont. Chiricahua points, dating from about 5000 to 1000 B.C., also have been found in the Tonto arm and the Payson area (Huckell 1973, 1978), and Archaic materials of an unspecified type have been found at sites on the upper Salt arm (Reid 1982) and on lower Cherry Creek (Wells 1971). Several Archaic sites have been identified for data recovery within the Roosevelt Lake area; however, they have yet to be fully described (Rice 1990). Although a number of isolated Archaic period projectile points were recovered from the Rye Creek sites, no Archaic components were defined. Preclassic Period The Preclassic occupation of the Tonto Basin traditionally begins with the first appearance of decorated ceramics. This mayor may not signify the first appearance of ceramics, however. The possibility of an earlier plainware, or plainware and redware, horizon would seem to be a very real one, given the existence of sites in the immediately surrounding area where such associations have been reported (e.g., the Red Mountain phase in the Phoenix Basin (Cable and Doyel 1985; Morris 1969) and the Hilltop phase in the Forestdale Valley (Haury and Sayles 1947». The Hohokam decorated ceramic series in the Tonto Basin extends from Snaketown Red-on-buff through Sacaton Red-on-buff; a Sacaton phase hiatus as proposed by the Gladwins is no longer tenable (Wood 1987, 1989; but cf. Ciolek-Torrello 1987:352,364). It should be noted, however, that decorated ceramics of any kind make up an extremely small proportion of surface and subsurface assemblages at nearly all Tonto Basin sites (Bruder and Ciolek-Torrello 1987; Rice 1985; Jewett 1986:121; Whittlesey and Reid 1982). The earliest datable ceramic type found in the Tonto Basin is Snaketown Red-on-buff (Elson 1989; Wood 1987), which dates to around AD. 650 or 700 in the Phoenix Basin (Dean 1990). Snaketown Red-on-buff ceramics were recovered from the Deer Creek site (AZ 0:15:52), excavated as part of this project (see Chapter 7). The few contexts producing Snaketown phase sherds are unfortunately ill-defined, and as of yet there are no excavated features that can be definitely assigned to this phase. Gila Butte Red-on-buff ceramics are well represented at sites such as Roosevelt 9:6 and Ushklish, as well as the Deer Creek site. A potentially large Gila Butte component has also recently been discovered at the Meddler Platform Mound site as part ofthe Roosevelt Lake Plan 6 excavations (Doelle et al. 1991). Although the Gila Butte phase was not recognized at the time of the Roosevelt 9:6 excavations, examination of the illustrated ceramics and their accompanying description clearly indicates the presence of both Gila Butte and Santa Cruz Red-on-buff (see especially Haury 1986:260). Roosevelt 9:6 is the quintessential Hohokam site of at least 15 (and presumably many more) pithouses, with a number of "smoking gun" Hohokam elements or traits, including secondary cremations and carved palettes (Haury 1932). From the point of view of both culture and research history, Roosevelt 9:6 remains the benchmark site for the Preclassic period in the Tonto Basin. 28 Chapter 3 Ushklish is located within the Upper Tonto Basin about 30 miles northwest of Roosevelt 9:6, and is comparable, or perhaps a little smaller, in size. Twelve pithouses were excavated out of an estimated 18 to 24 houses at the site (B. Huckell, personal communication, 1990). The predominant decorated type at the site is Gila Butte Red-on-buff, based on a retyping of the sherds in the Arizona State Museum collections (see Chapter 24, Volume 3). Ciolek-Torrello (1987:35) further notes that Ushklish is the largest of a small cluster of Preclassic sites along lower Hardt Creek. The Preclassic period settlement at Ushklish should probably include these nearby, contemporary sites; thus Ushklish may have been comparable to Roosevelt 9:6 in size. However, Ushklish is far from being a "textbook" Colonial period Hohokam site; indeed, its principal investigator ascribed the site to a mixed Hohokam-Mogollon cultural affiliation (Haas 1971), and it has definite similarities to excavated sites in the White Mountain area that have been called Mogollon (Halbirt and Dosh 1991). Whatever the merit of these cultural tags, they represent a short hand notation of formal variation beyond the Phoenix Basin Hohokam model. Colonial period sites -- or rather, sites with Colonial period ceramics -- are known from throughout the Lower Tonto Basin (Fuller et al. 1976; Hohmann 1985; Huckell 1977; Jeter 1978; Rice 1985; Rice and Bostwick 1986), and from the Globe-Miami area (Doyel 1978; Vickery 1945; Windmiller 1972). Colonial period buffwares are also found at pithouse sites in the Vosburg (Morris 1970) district near Young, Arizona, and at Mogollon sites further east (Halbirt and Dosh 1991). Late Colonial period buffwares are much in evidence at pithouse sites at Talkalai Lake near San Carlos, about 50 miles east of the Tonto Basin. One of these sites, AZ V:ll:ll (ARS) has been termed a Colonial period "Hohokam" pithouse village with five excavated pithouses, cremations, and a canal segment (Mitchell 1986). The ceramic assemblage includes Gila Butte Redon-buff, but is predominately Santa Cruz Red-on-buff. The point here is to show that Colonial period Hohokam ceramics are found in low frequencies in many areas in and around the Tonto Basin. Does this mean that these sites are all Hohokam sites? Haury (1932) and Gladwin (Gladwin and Gladwin 1935) believed them to be, and formulated the notion of a Hohokam expansion in the Colonial period. The "colonial" interpretation (with minor modifications) also has been accepted by many current archaeologists (Doyel 1978; Hohmann and Kelley 1988; Rice 1985; Wood and McAllister 1980); however, other researchers have questioned the model (Fuller et al. 1976; Neitzel 1985; Reid and Whittlesey 1982). Much of the argument revolves around Roosevelt 9:6, the first excavated pithouse site in the Tonto Basin. Decorated ceramics are not common at 9:6, but those present are predominantly Hohokam buffwares, and house forms, burial patterns, and other portable artifacts all appear to reflect a Phoenix Basin "prototype." Roosevelt 9:6 poses a fundamental problem in Tonto Basin archaeology. The questions it raises are methodological, and basic to issues of culture history and ethnicity. Cultural systematics (and chronological arguments based on systematics) are predicated on formal similarities between known sequences and areas to which a known sequence may be extended. Roosevelt 9:6 is, by both definition and tradition, a Hohokam site, and therefore appropriate for the application of Phoenix Basin Hohokam systematics; but is it "typical" of the Tonto Basin's Preclassic period archaeology? Since the excavation of Roosevelt 9:6, a number of Preclassic sites have been excavated in the general area -- Miami Wash (Doyel 1978), Ash Creek (Rice 1985), Slate Creek (Huckell 1977), and now Rye Creek -- and none exhibit the very clear Hohokam signature of Roosevelt 9:6. Indeed, the pattern at these sites and other upland Preclassic sites in the Fort Apache and San Carlos areas (discussed above) is one of small settlements with very few decorated ceramics of any kind, and a variety of pit structure forms that defy easy cultural categorization. In this context, Roosevelt 9:6 appears to be the exception. Had Haury excavated the Ash Creek or Rye Creek sites, with their irregular pithouses and low proportions of buffwares, instead of Roosevelt 9:6, how would the Tonto Basin Preclassic period be perceived today? That is, should we assume that Roosevelt 9:6 is representative of a typical Tonto Basin site, or should we assume that the pattern of small, culturally ambiguous pithouse sites documented from Rye Creek to San Carlos is the norm, and Roosevelt 9:6 the exception? The issue remains far from resolved but has important implications for our understanding of the prehistory. It is also curious to note that the hallmark of the Hohokam Preclassic period, the ballcourt, has yet to be recorded from the Tonto Basin, although more than 200 courts are now known from 165 sites in Arizona Previous Research And Culture History 29 alone (Crown 1990; Wilcox 1991; Wilcox and Sternberg 1983). Preclassic period courts have been found in all areas of the Southwest that were thOUght to have been either an actual part of the Hohokam settlement network, or strongly influenced by the Hohokam, including the Globe-Miami and Verde Valley areas south and west of the Tonto Basin. The adoption of the ballcourt system by outside groups has been suggested to represent significant economic and social interaction with the Hohokam core area, perhaps indicative of a shared ideology (Wilcox 1991; Wilcox and Sternberg 1983). The significance of the lack of Tonto Basin ballcourts is unclear; it may be due simply to sampling problems (although given the long-term reconnaissance nature of Tonto Basin research, it is probably safe to say that most of the likely areas have been inspected), or they may be underneath Roosevelt Lake or later Classic period sites. The fact that a large number of Classic period platform mound sites are known, however, which are believed to have served a similar integrative function as ball courts (Crown 1987, 1990), suggests that Roosevelt Lake may not be the real problem. A plausible argument has been made by Wood (1985:253), who suggests that the lack of ballcourts is related to the lack of exportable trade goods in the Tonto Basin prior to the Sacaton phase, when the ballcourt system was no longer widespread. This is based on the assumption that ballcourts played a significant role in trade, exchange, and redistribution networks, currently a debatable point. Regardless, the lack of ballcourts, if real, is another critical factor in our understanding of the nature of the Preclassic period occupation. Classic Period As in most parts of the Southwest, the Tonto Basin experienced a transition from pithouse to pueblo architecture, or, at least, the addition of pueblo-style structures to a tradition of pithouses. The appearance of masonry architecture is generally recognized as the benchmark of the Classic period in the Tonto Basin. This has been variously interpreted as a sociocultural reorganization by the local population, an intrusion of different (usually northern or eastern based) cultural groups, or some sort of combination. It is important to note that the transition from pithouse to surface architecture generally has been interpreted elsewhere in the American Southwest as an adaptive, not an ethnic, change. From the sample of sites currently available, it appears that many locations have multicomponent sites with late Preclassic period pithouses overlain by early Classic period surface structures. The stratigraphic superimposition has, in tum, been used to argue ethnic and cultural continuity between the two periods (Doyel 1978; Rice 1985), although the actual dynamics behind this transition are largely unknown. In the Miami Wash and Ash Creek areas, however, potential settlements apparently were restricted to the small tips of truncated ridges, increasing the chance of superimposition of otherwise unrelated occupations. In areas such as Rye Creek, with less restricted potential occupational surfaces, the degree of superimposition and multicomponency is much lower. As a result, the issue of cultural continuity between the Preclassic and Classic periods is far from settled and in need of more data and further evaluation. Related to these concerns, in both past and present research, is the specter of the Salado. "Salado" refers to an archaeological manifestation historically seen as emanating from the Tonto Basin; the Tonto Basin is the Salado heartland (Wood 1986, 1989; Hohmann and Kelley 1988). This is currently the subject of much debate, however, because other researchers, most notably Wilcox (Wilcox and Sternberg 1983), view the Salado not as a cohesive culture with a defined list of traits, but more as an economic interaction sphere or panSouthwestern ideological system. The Tonto Basin's importance in broader Southwestern culture history, however, comes from the early claims of Gladwin and Gladwin (1935) and Haury (1945) that Salado people left the Tonto Basin to influence or even truncate Hohokam development in the Phoenix Basin and much of southeastern Arizona. The Tonto Basin was seen as pivotal not because of the sites within it, but because of what happened once populations from there moved into Hohokam areas to the south. This view of the Tonto Basin stunts our understanding of what happened there, and why. The Tonto Basin of the Classic period is, in many ways, not at all unique. Similar shifts in settlement pattern, architectural form, and ceramic assemblage took place all along the transitional zone between the Colorado Plateaus and the low deserts of Chihuahua and Sonora. To the west, the middle Verde saw a similar change during the Honanki phase (AD. 1125-13(0), which traditionally is explained as an intrusion of southern Sinagua. To the 30 Chapter 3 east, the Bylas phase (AD. 1100-12(0) in the San Carlos area and similar, unnamed manifestations in the Safford Valley show major shifts in ceramics and architecture. In New Mexico, the "indigenous" Mimbres Mogollon sequence ends and is "replaced" by non-Mimbres Black Mountain (AD. 1150-13(0) and Animas (AD. 1150-13(0) phase populations, usually attributed to influences from Casas Grandes. Similar events also occurred within the Hohokam area, as manifested in the Soho phase (AD. 1150-13(0) of the Phoenix Basin and the Late Rincon (AD. 1100-1150) and Tanque Verde (AD. 1150-13(0) phases of the Tucson Basin. These synchronous shifts (or breaks) in local sequences suggest displacement or (more likely) reorganization of local populations on a huge, nearly pan-Southwestern scale. The Tonto Basin is only one in a chain of geographically (or artificially) disjunct districts where the "local" sequences all take sharp turns between AD. 1150 and 1200. The Classic period can be separated into at least two phases based on ceramic assemblages: an early Classic period, marked by black-on-white ceramics and dating from about AD. 1150 to 1300; and a late Classic period, marked by Salado and related polychrome ceramics, dating from about AD. 1300 to 1450. The traditional systematics for the early Classic period includes two phases, the Miami/Hardt phase (subsuming the Gladwin's Cherry Creek phase), dating from AD. 1150 to 1200, and the Roosevelt phase, dating from AD. 1200 to 1300 (Doyel 1978; Wood 1986). The late Classic period corresponds to the Gila phase, dating from AD. 1300 to 1450 (Wood 1986, 1989). Early Classic Period The Miami phase (in the Lower Basin) and the Hardt phase (in the Upper Basin) both refer to the same thing, a very brief (50-year) transitional stage between the Preclassic and Classic periods. The Miami phase was defined from a single site, the Columbus site (V:9:57), located just outside the east end of the Tonto Basin. This site had a number of noncontiguous rooms within a small, compound-like enclosure, and a ceramic assemblage including Snowflake Black-on-white, Reserve-Tularosa Black-on-white, St. Johns Polychrome, McDonald Corrugated, San Carlos Red-on-brown, and traces of other PIlI whitewares (Doyel 1978:194). The Scorpion Ridge site (AZ V:9:14) also is assigned to the Miami phase (Windmiller 1974). The Miami/Hardt taxon has important implications for the culture history of the Tonto Basin; as a transitional phase, it fills a perceived gap between the Preclassic and Classic periods, and between what has been seen as occupation of the Tonto Basin by two different ethnic groups. That is, its use in Tonto Basin systematics automatically validates argument for cultural continuity between the Preclassic and Classic periods. Transitional phases, which are almost always poorly defined, are a relatively common explanatory tool in Hohokam archaeology, and the Miami/Hardt phase is essentially identical to, and serves the same bridging purpose, as the Santan phase in the Phoenix Basin and the Cortaro phase (now generally discarded) in the Tucson Basin. Researchers who advocate such continuity in the Tonto Basin (e.g., Doyel 1978; Hohmann and Kelley 1988; Rice 1985; Wood 1986, 1989) may be absolutely correct, but due to the overall lack of concrete data, the Miami/Hardt phase at this time can only be viewed as tentative. It risks the false precision of an over-fine taxonomy. The following Roosevelt phase is also very poorly understood; very few previously excavated contexts have, to our knowledge, ever been assigned to this phase, unless Gladwins' and Gladwins' (1935) original definition was based on unreported excavations. A single Roosevelt phase site, the Boone Moore site (AZ 0:15:55), was excavated as part of this project, and several Roosevelt phase sites are present within the Roosevelt Lake Plan 6 area currently being excavated by Arizona State University and Desert Archaeology, although these data are still preliminary. Roosevelt phase sites are reported to be extremely widespread in and around the Tonto Basin (Wood 1989), but most appear to be small, short-lived occupations, with phase assignment based almost solely on surface data. Roosevelt phase material is also present at larger Tonto Basin sites with later Gila phase components, but the nature of the Roosevelt phase components, and whether they are truly temporally distinct, is almost entirely unknown. It is important to note that dating of the Roosevelt phase often depends on the absence of late Classic period polychromes, which in any case might be rare or absent on the surface of small Salado sites. This difficulty may not affect Wood's (1989) conclusions regarding the density of Roosevelt phase sites, but it should be considered in evaluations of Classic period settlement patterns. Previous Research And Culture History 31 What was the Roosevelt phase? At present, we do not know. Gladwin and Gladwin (1935) saw the Roosevelt phase as a major immigration of Anasazi peoples. Doyel (1978) argued that the Roosevelt phase was essentially a continuation of the Miami phase, itself a development of local Hohokam populations influenced by Western Pueblo/Mogollon neighbors. Wood (1986, 1989) and Rice (1985) stress continuity between the Roosevelt phase and earlier Hohokam populations, and de-emphasizes the importance of the Western Pueblo/Mogollon in its pattern. Whittlesey and Reid (1982) do just the reverse; they set the mix with a strong Mogollon base and a light touch of Hohokam. Various elements of the Roosevelt phase (and the following Gila phase) can be identified in surrounding areas, and the Tonto Basin pattern combines a number of elements that are perceived as either specific to surrounding areas or nonspecific to any single area. Further work clearly is needed to resolve the many issues raised here. Late Classic Period The late Classic period in the Tonto Basin includes a single phase, the Gila phase. The Gila phase is the culmination of Tonto Basin prehistory. The surface archaeology of the Gila phase has been synthesized in admirable detail by Wood (1986, 1989), and analyzed by Wood (1985), Tjaden (1978), Jewett (1978), and Hohmann (Hohmann and Kelley 1988). A series of very large sites, each encompassing a bewildering variety of domestic and public architecture -- platform mounds, compounds, pueblos -- were located along both the Tonto and Salt arms, and in favorable upland locations. Canal irrigation is assumed, based on the reported presence of prehistoric canals, and upland enhanced dry-farming can also be demonstrated. Salado polychromes, possibly originating in the Tonto Basin, but also suspected to be made in many areas of the greater Southwest (Crown 1984; Crown and Bishop 1987), are a major index of the ceramic assemblage. The extremely large regional distribution of the Salado polychrome types suggests that the Tonto Basin was integral to large-scale eXChange systems -- or at least, some kind of large regional "system." Because it is the Tonto Basin's most visible archaeology, Gila phase settlement patterns have been analyzed and reanalyzed, with many differing opinions about the social correlates of settlement architecture and settlement pattern. These range from a Hohokam proto-state (Rice 1990) on the high end, to a more moderately complex, smaller-scale Saladoan polity (Hohmann and Kelley 1988, Wood 1986), and on down to multiethnic egalitarian communities (Reid 1982) on the low end. Given the intensity with which many of these opinions are held, it is somewhat surprising to note that only three large Gila phase sites have been excavated in the Tonto Basin: Tonto Cliff Dwellings (Steen et al. 1962); Rye Creek Ruin (Gladwin 1957; Haury 1930); and the VIV or Meredith Ranch Ruin, (Mills and Mills 1975). Moreover, the latter two projects are only sketchily reported. About a dozen smaller sites, generally with no more than 5 or 6 rooms, have also been excavated (Rice 1985; Hohmann 1985). These sites provide a small and very difficult sample from what appears to have been one of the major fourteenth century population centers in the American Southwest. Rye Creek Ruin (Haury 1930) is a large Gila phase pueblo that incorporates a platform mound as one comer of a large, walled plaza; more than 150 single-and double-story rooms are present (see Chapter 27, Volume 3). Several extramural compounds are located to the west of the main structure, but the dating of these units has not yet been precisely determined (Haury's stratigraphic tests and Wood's surface assessments [personal communication, 1990] suggest that they may predate the main occupation.) Among other things, Haury's brief excavations disclosed one very unusual room with a sherd-paved floor associated with the platform mound and a road-like entry into the plaza, and a second room with the round base of a wicker-work fixed granary. Gila Pueblo sponsored extensive digging for burials within the plaza, producing a large number of whole pots but very little data. Although controlled stratigraphic tests by Haury in 1930 in several associated trash mounds failed to produce evidence of any substantial pre-Gila phase component, recent testing of three trash mounds as part of this project did document a pre-AD. 1300 occupation in two small mounds, although the extent of this occupation is uncertain. The Tonto Qiff Dwellings (Steen et al. 1962) provide superb architectural data for the Gila phase, keeping in mind the constant caveat of cliff dwellings that space constrained form. Construction details and, to some 32 Chapter 3 extent, interroom patterning may be usefully extrapolated to the large Gila phase riverine sites, but because of the size of the structures, the layout of the settlement may bear little relation to open sites. A question posed by the Tonto Cliff Dwellings is: why cliff dwellings at all? It appears that large open sites were supportable along the nearby Salt River and Tonto Creeks. There is no appreciable arable land in the drainage of the Tonto Cliff Dwellings; why were they built? The dating is disappointing (one tree-ring date at AD. 1lO9vv [Bannister and Robinson 1971D, but the ceramic assemblage appears to be almost purely Gila phase. What does the construction of a cliff dwelling here (and many others in the nearby Sierra Anchas) in the early-to-mid AD. 1300s indicate about settlement in the large riverine Gila phase sites of the Tonto Basin proper? Is cliff dwelling construction related to construction of large "upland" Gila phase sites such as Rye Creek Ruin and the Globe-Miami area sites? These data may be interpreted as either an expansion or an abandonment of riverine settlement established in the Roosevelt phase, although it should be noted that small Roosevelt phase sites are also reported from nonriverine, upland locations (Wood 1986.) The VIV Ruin (Mills and Mills 1975) is the only major Gila phase site within the Tonto Basin proper to be excavated prior to the current Plan 6 work at Roosevelt Lake. The Mills' concentrated on the platform mound, but also worked in surrounding compounds. Their report is confusing, but it contains much that is of interest. For example, a room in the VIV platform mound produced a wicker-work granary base identical to the one Haury found at Rye Creek Ruin; these also appear to be similar to granaries recovered by Arizona State University at Schoolhouse Point and other sites (G. Rice, personal communication, 1990). Moreover, the description and model of the platform mound (on display at the Eastern Arizona Community College museum), which was unrecognized as such by the Millses during their excavations, make it clear that this structure was indeed a platform mound with rooms on its upper surface. The results of the Bureau of Reclamation's Plan 6 excavations should completely eclipse the current data base from large Roosevelt and Gila phase sites in the Tonto Basin. Arizona State University is currently investigating platform mound communities at Cline Terrace, Rock Island, and Schoolhouse Point, and Desert Archaeology is investigating the Meddler and Pyramid Point platform mounds and the pueblo at Griffin Wash (Figure 3.1). In view of the ongoing nature of this work, no useful purpose will be served by attempting to extrapolate further from the three excavations briefly discussed above. ARCHAEOLOGICAL RESEARCH WITmN THE UPPER TONTO BASIN In this section, we examine in greater detail the previous research within the Upper Tonto Basin, the locale of the Rye Creek Project. The Upper Basin, although unquestionably part of the greater Tonto Basin regional system, can also be considered to be a distinct local subsystem. This is due largely to environmental differences, although sociocultural factors may also be involved. As outlined in the previous chapter, the Lower Basin contains large areas of arable alluvial floodplain, capable of supporting a sizable prehistoric population. A considerable number of very large habitation sites are known from this area; over 15 sites with platform mounds have been recorded. In contrast, the Upper Basin contains only two significantly large areas of arable land; one at the junction of Rye and Deer creeks, and the other along Tonto Creek near the town of Gisela. Contained within these two areas are Rye Creek Ruin and the Gisela Platform Mound, the two largest sites within the Upper Basin and the only recorded sites with platform mounds. The majority of the Upper Basin is within the higher pediment terraces and piedmont of the Mazatzal Mountains, where dispersed areas of arable land are found only within small, seasonal washes. Not surprisingly, the great majority of sites consist of small habitations, farmsteads, fieldhouses, and (presumably) special use or limited activity sites. Unlike the Lower Basin, investigations into the prehistory of the Upper Basin have only recently begun. Outside of the very limited excavations done by Gila Pueblo in 1929 and 1930 at Rye Creek Ruin (Gladwin 1957; Haury 1930), the majority of work in the Upper Basin has been undertaken within the last 20 years, primarily under the auspices of various contract and highway salvage programs. Although the recent and ongoing work of the Tonto National Forest archaeology program (Macnider and Effland 1989; Wood et. al 1989) has served to greatly augment the sample of known sites, excavation and systematic survey coverage within this region are still limited. Previous investigations have concentrated on either linear areas (Ciolek- Previous Research And CuJlW'e History 33 Torrello 1987; Hammack 1969; Hucke1l1978; Jeter 1978; Olson 1971; Reid 1982), isolated large sites (Gladwin 1957; Haury 1930; ASM Site Files), or miscellaneous sites located along roadways (Haas 1971; Huckelll973, 1977). As a result, any reconstruction of the sequence of occupation is necessarily incomplete, and based upon a somewhat biased site sample. A critical factor demonstrated by these studies is the overall lack of definitive patterning in site characteristics. Much variability is evident. Furthermore, the somewhat eclectic nature of the artifact assemblage and architecture, primarily consisting of small, nondescript, masonry structures or pithouses within low density plainware ceramic and lithic scatters, makes inferences based on analogy rather unreliable. In other words, modeling the archaeology of the Upper Tonto Basin by applying the archaeology and cultural systematics of neighboring areas is not currently supportable, and, if anything, potentially misleading. This lack of patterning, and a corresponding lack of consensus over the meaning of the extant data (i.e., different researchers are using the same data to arrive at very different conclusions), has important implications for the archaeology of the area. A wide range of prehistoric cultural traditions are claimed to be represented within the Upper Basin or in close proximity to it. In fact, various studies have purported to demonstrate the existence of the Southern Sinagua (Ciolek-Torrello 1987; Hammack 1969; Hucke1l1978; Pilles 1976), the Hohokam (Haas 1971; Huckell 1977; Jeter 1978), the Salado (Ciolek-Torrello 1987; Huckell 1977; Jeter 1978), the Mogollon (Haas 1971; Whittlesey and Reid 1982), and the Anasazi (Morris 1970). Archaic groups (Ciolek-Torrello 1987; Huckell 1973, 1978) and Clovis hunters (Huckell 1978) also are believed to be present. Furthermore, the Sinagua and Salado components are thought to be contemporaneous (cf. Pilles 1976), as are the slightly earlier Mogollon and Hohokam occupations. Thus, within an area of approximately 25 square miles there are Sinagua sites, Hohokam sites, Salado sites, and Mogollon sites, as well as one settlement suggested to have been occupied by both the Mogollon and the Hohokam (Haas 1971), and one settlement occupied by both the Anasazi and the Hohokam (Morris 1970). These data, if correct, would indicate that the Upper Basin was indeed a zone of intense cultural mixing and contact (Olson 1963). Examination of the various artifact assemblages recovered from these sites, however, particularly the ceramics, reveals that this interpretation of the Upper Basin as a cultural frontier or "no man's land" exploited by different, and possibly contemporaneous, cultural groups, may be suspect. This is based upon several lines of reasoning. First, it appears that different investigators are using similar lines of evidence, specifically paddleand-anvil plainware ceramics, a variety of pithouse types, and low-walled cobble masonry architecture, in the assignment of different cultural traditions. Second, the chronology is not well understood. Although several absolute dates have recently been recovered from the Ord Mine (Ciolek-Torrello 1987:339-344) and Ash Creek (Rice 1985:21-25) sites, these are for the most part so anomalous as to be of little value (see Chapter 25, Volume 3). The radiocarbon dates do not agree with the corresponding archaeomagnetic dates, which in tum do not agree with the ceramic dates. It is clearly apparent that significant discrepancies are present in both inter- and intrasite dating. Although some of these problems can be explained through sampling error or the "old wood" problem (Schiffer 1986), it is also likely that our understanding of the behavioral manifestations of Tonto Basin temporal patterns are far from complete. For example, does a cobble masonry structure always indicate a Roosevelt or Gila phase occupation? Are pithouses solely confined to the pre-Salado period? As a result of the lack of reliable absolute dates we are still in a situation where we cannot reconstruct the sequence of occupation except through relative methods using a few cross-dated ceramics, the significance of which are not well understood. That is, given the large chronological range of many of the ceramic types used for dating within the Upper Basin (particularly the plainwares but the decorated wares as well), statements of contemporaneity or temporal change are currently difficult to support. This problem is compounded by considerable controversy over both the identification of plainware and redware ceramics, and their corresponding cultural affiliation, due primarily to a lack of ceramic compositional and petrographic studies. As a result, there has been much speculation based on very little hard data. For example, various archaeologists working within the Tonto Basin and Verde Valley areas have noted that sherds of Verde Red and Verde Brown (or plain) are nearly identical to, and almost impossible to separate from, sherds of Tonto Red and Tonto Brown (Bruder and Ciolek-Torrello 1987; Hammack 1969; Huckelll978; Jeter 34 Chapter 3 1978; McGuire 1977). Although both the Verde and Tonto series are considered to be locally produced, they have traditionally been thought to represent different cultural groups. Colton (1946) and Colton and Hargrave (1937) include these types under Alameda Brownwares, which they believe were produced, used, and traded by the Sinagua. Breternitz (1960) and Schroeder (1975), however, in their work in the Verde Valley, interpret Verde Brown as Hohokam in origin and contrast it with later intrusive types brought into the valley by the Sinagua. Wood (1987) suggests that the Verde and Tonto ceramics are actually a single regional ware that can be subdivided into two slightly distinct local types through the inclusion of different ceramic tempers. According to Wood, this is indicative of two different ethnic groups who are within the same cultural tradition. Finally, within the Tonto Basin these same types have been variously, and perhaps most commonly, interpreted as either Salado, Hohokam, Sinagua, or Mogollon, depending upon their association with small amounts of diagnostic decorated pottery (Haas 1971; Hammack 1969; Huckell 1977; Jeter 1978; Whittlesey and Reid 1982). In addition, the plainwares are poorly dated and appear to have fairly long chronological ranges. Tonto Brown (or Red) for example, dates from AD. 1000-1400 (Breternitz 1966; Colton and Hargrave 1937) as does Verde Brown or Verde Red (Colton and Hargrave 1937). Wood (1987) gives both types an even longer span, from around AD. 500-1400, which seems more reasonable given the presence of plainware ceramics on Tonto Basin sites dated prior to AD. 1000. Unfortunately, within the Upper Basin the absence of well-dated diagnostic decorated ceramics (which generally comprise less than 5 percent of the ceramic assemblage) has often necessitated the use of plainwares for relative dating. Given the time ranges that these ceramics encompass, the establishment of contemporaneity or even site sequencing is impossible except on a very broad level. Even the various seriation methods that have been attempted comparing the relative frequencies of plain, red, and corrugated wares, have been successful only at a very gross and project-specific level (Woodward et al. 1985), while totally unsuccessful on others (Bruder and Ciolek-Torrello 1987). Ceramic and architectural data from the Upper Tonto Basin serve as an excellent illustration of how the lack of temporally diagnostic ceramics and the lack of terminological agreement affects interpretation of the area's prehistory. Although this has resulted in much apparent speculation and confusion in the published literature, some of which by today's practices seems unwarranted, it is important to note that these researchers were following the accepted methods and standards of their time. The great majority of these projects were undertaken more than 15 years ago. For example, both Hammack (1969) and Huckell (1978) working just north of the Rye Creek Project area, believe that their sites are indicative of the Southern Sinagua culture. Their interpretations are based upon three factors: the presence of Alameda Brownwares (typed as Verde Red and Verde Brown), the absence of significant numbers of diagnostic Salado, Mogollon, or Hohokam wares, and the presence of cobble or unshaped masonry structures. The Alameda Brownwares account for approximately 98 percent of the ceramics recovered from the six sites. In contrast, at the Slate Creek Ruin (AZ U:3:28) south of the present project area, Huckell (1977) proposed a multicomponent Hohokam and Salado occupation. At this site more than 90 percent of the recovered ceramics were a paddle-and-anvil brownware that Huckell typed as Verdeffonto Brown due to the difficulty in separating the two types. Huckell regards the brownwares as a local or indigenous type, and relies upon the small amounts of decorated or corrugated ceramics in the assignment of cultural affiliation; the 50 Salado Red sherds (comprising 2.8 percent of the recovered ceramics) are indicative of a Salado occupation, while the 15 Wingfield Plain sherds (0.84 percent) and the 5 Santa Cruz Red-on-buff sherds (0.28 percent) represent an earlier Hohokam occupation. Jeter (1978), working in the Lower Tonto Basin some 10 miles southwest of the Rye Creek area, excavated five sites; four of which were affiliated with the Salado, and one with the Hohokam. The Salado sites all contained a single dry-laid masonry structure, and Jeter assigned them to the Roosevelt phase (AD. 11501300) based on the presence of Salado Red Corrugated and the absence of any Gila phase diagnostics. Salado Red ceramics comprise 10.6,20.7, and 71.35 percent of the assemblages at AZ U:3:30, AZ U:3:32, and AZ U:3:31 respectively. At AZ U:3:33, which Jeter assigns to the Hohokam culture, sherds typed as Alameda Previous Research And CulfW'e History 35 Brownwares make up 91 percent of the recovered ceramics. These are considered to be locally manufactured. Diagnostic Hohokam wares comprise only a small percentage of the assemblage; Wingfield Plain makes up 2.5 percent of the assemblage and buffwares 4.9 percent. Just south of Hardt Creek, Olson (1971) excavated a lO-room pueblo (NA 8082). Again a paddle-and-anvilproduced brownware is the dominant pottery type, although in this situation it was typed as Tonto Red and Tonto Brown. These two types comprise almost 95 percent of the recovered ceramics. Salado Red was the next frequently occurring ceramic type (3.2 percent) followed by Tonto Polychrome (0.83 percent). Olson, however, makes no claims concerning the cultural affiliation of the occupants. Haas (1971), working about three miles east of the Rye Creek Project area at the site of Ushklish (AZ 0:15:31), found what he believes to be a pithouse village jointly occupied by both the Mogollon and the Hohokam. His evidence is largely ceramic, although he considers the various pithouse styles (i.e., house-in-pit versus true pithouse) to be indicative of the two groups; six Mogollon, three Hohokam, and three unidentified pithouses were present. According to Haas, the variability in the architectural styles is supported by a corresponding variability in the ceramic types. The decorated assemblage includes low percentages of Hohokam buffwares and Tusayan whitewares. The buffware ceramics, indicative of the Hohokam presence, are primarily Gila Butte Red-on-buff (see Chapter 24, Volume 3), and comprise around 6 percent of the sample of 26,568 typed sherds. The Mogollon, on the other hand, are ceramically represented by Forestdale Red and San Francisco Red, although Haas admits that if these sherds were mixed in with a sample of Gila Red, a Hohokam ware, they would be virtually indistinguishable. These types constitute 1.9 percent of the ceramic assemblage. The Tusayan whitewares are not used as a basis for assigning cultural affiliation, since they comprise less than 1 percent of the assemblage. Again, a paddle-and-anvil-produced plainware comprises 96 percent of the ceramics although Haas does not assign these sherds to any established tradition. The most complete research undertaken in the Upper Tonto Basin to date was conducted by the Museum of Northern Arizona along the section of State Route 87 just south of the present project area (Ciolek-Torrello 1987). This area is within the higher piedmont of the Mazatzals, and contains smaller and more ephemeral streams and significantly less area of arable land than the Rye Creek Project area. Twenty-four sites were thoroughly investigated during the mitigation phase. Six of these were classified through diagnostic artifacts as dating to the Archaic period. The majority of the sites, however, consisted of small one- or two-room masonry pueblos, although several larger pueblos and one possible compound were also present; no sites with pithouses were recorded. According to Ciolek-Torrello (1987) the small sites probably served as agricultural or wild food gathering locales whereas the larger sites were more permanent habitations. These in tum were integrated into larger community networks, the exact nature of which is unclear. Even with the absence of decorated or diagnostic ceramics (out of an assemblage of 22,264 sherds, 2.1 percent were identified as Salado Red Corrugated, and 0.27 percent were decorated), and reliable absolute dates, Ciolek-Torrello (1987:366-369) assigns the sites to the Salado or Sinagua cultures. As with previous researchers Ciolek-Torrello appears to be basing his interpretations primarily on the presence of unshaped cobble-masonry architecture and paddleand-anvil plainware ceramics. In this sense Ciolek-Torrello continues the long-standing and traditional method of dealing with chronology, cultural affiliation, and cultural origins, through the use of the identical data sets used by previous researchers. Therefore, except perhaps for the three Salado sites excavated by Jeter (1978), where Salado Red Corrugated comprises a relatively high percentage of the assemblage (it should be noted, however, that the significance and cultural affiliation of this type are also unclear), we have a situation in which the ceramic assemblages appear to be relatively identical; that is, they are dominated by paddle-and-anvil plainware ceramics and differ only in the types of painted wares represented. These polychrome, black-on-white, and red-on-buff ceramics stem from the Anasazi, Mogollon, and Hohokam areas, although the local manufacture of some of these types cannot be discounted without petrographic or other compositional analyses. The percentages of the painted wares vary in the Upper Basin assemblages between 0.5 to almost 10 percent; it is generally less than 5 percent, and on the average between 2 and 3 percent. These figures may indicate low-level regional trading networks, rather than cultural affiliation or migration (Neitzel 1985; Whittesey and Reid 1982). 36 Chapter 3 In conjunction with ceramics, site architecture often has been used as an indicator of both cultural affiliation and cultural sequencing. Within the Upper Basin, however, the architecture is relatively variable, and not easily categorized into predefined culturally specific types. Variability may occur in the number of structures, their size and shape, and their spatial arrangement, although the technique and style of construction are generally similar. Prior to the Rye Creek Project, only four sites with pithouses were known from the Upper Basin (Haas 1971; Hammack 1969; Huckelll977; Jeter 1978). Two of these were multicomponent, containing both pithouses and masonry structures; the other two contained only pithouses. Only Haas (1971), at Ushklish, found what might be termed a pithouse village; the other sites contained from one to three structures. Pithouses encompass a multitude of shapes and sizes, and range from Hohokam-Iike shallow "house-in-pits" to Mogollon and Anasazi style "true" pithouses to possibly Sinagua-Iike houses with large entrances or alcoves; most are relatively nondescript. Furthermore, data from the Rye Creek Project suggest that Ushklish is not unusual in exhibiting a mix of contemporaneous architectural styles within the same site. The majority of the masonry sites contain from one to three dry-laid cobble or unshaped masonry rooms constructed from material procured from the closest suitable source. The walls were probably only several masonry courses high, although some structures with full size walls may be present at the larger sites, and were most likely covered by jacal or ramada-like structures. Unfortunately, in terms of the problem of cultural affiliation, these architectural styles have been reported for the Salado (Doyel 1978), Sinagua (Pilles 1978), Mogollon (Tuggle 1970), and Anasazi (Dean and Lindsey 1978). In fact, it is now known that in areas lacking subsoil or containing available surface cobbles, the Hohokam also constructed dry-laid masonry structures (Downum et al. 1985; Doyel and Elson 1985; Wasley and Johnson 1965). Therefore, only limited conclusions regarding cultural affiliation can be drawn from the architecture. Furthermore, although there are some ceramic hints that the pithouse occupation of the Upper Basin was slightly earlier than the masonry pueblos (Haas 1971; Huckell 1977), the lack of absolute dates necessary to establish contemporaneity among the pithouses or between the pithouses and the masonry structures, makes any statement regarding contemporaneity or cultural change somewhat tenuous. In fact, it has now been established that in many areas of the Southwest, pithouses and masonry structures existed side-by-side, and that the transition to above-ground masonry was not a sudden occurrence (Lekson 1988). In the Tonto Basin, this pattern is tentatively suggested by Rice's (1985:86) findings at AZ U:3:51 (ASU) at Ash Creek, where a later pithouse at this predominantly Sacaton phase site may have been associated with a Salado occupation. Finally, due primarily to the small size of the sample from the Upper Basin area, it has yet to be demonstrated whether architectural variability corresponds in any way to functional variability. Previous research in the area has stressed the agricultural or wild plant processing functions of the small sites (Ciolek-Torrello 1987; Hammack 1969; Huckell 1978; Rice 1985). Other functions are possible, however, including animal as well as plant procurement, natural resource procurement (such as chert, argillite, ceramic temper, or clay), migratory stopping stations, or manufacturing areas. Ethnographic investigations into small sites have demonstrated that a wide range of activities are possible (Ward 1978) although they may not be visible archaeologically. In fact, it is quite likely that these sites, although architecturally similar, were characterized by widely varying ranges of activities. Regardless of the specific function, however, the great majority of sites within the Upper Basin appear to be locales for short term, and probably limited, activities. As a closing thought, it is interesting to note that historical factors may also be influencing interpretations of Tonto Basin prehistory. That is, it appears that a number of previous investigators have assigned cultural affiliation in accordance with the archaeology they were most familiar with -- if most of their work was in the southern desert areas, the sites strongly resembled those of the Hohokam; if most of their work was in the northern or eastern pueblo areas, the sites appeared to be Mogollon, Anasazi, or Sinagua. This further underscores the nondescript and relatively eclectic nature of the archaeology, as well as the difficulties inherent in interpreting it. Previous Research And Culture History 37 CONCLUSIONS This review has identified many questions and research topics for future study of Tonto Basin culture history. It also has raised more questions than it has resolved. As is readily apparent, much of the data, and the subsequent interpretations of the data, are confusing, if not outright contradictory. Two factors are responsible for this: the overall inadequacy of the database, and the apparent variability, or at least the lack of distinct analogy, in the archaeological record of the area (which may also be due to the lack of data). Data recovered from the Rye Creek Project, the largest project undertaken in the Upper Basin to date, will help to settle some of these issues, although many are simply not resolvable at the present time. In terms of the culture history of the Tonto Basin, two broad areas stand out as critical. The primary question, particularly for the Rye Creek data, is what is the nature and scale of the Preclassic period occupation? A secondary question is what is the nature and scale of the Roosevelt phase occupation? Research focusing on the latest, Gila phase, of the Tonto Basin seems premature without a better knowledge of the developments leading up to it. The two issues are similarly stated, but are operationally separable. Arguments about the Preclassic period in the Tonto Basin center on the questions of cultural affiliation and the mechanisms behind the initial settlement of the region. Whether the settlement of the Upper Basin is truly distinct from the Lower Basin, or whether they can be considered part of the same system and process, also needs to be investigated. Of course, a primary consideration is the role of Roosevelt Lake and later Classic period settlements in filtering the Preclassic period archaeological record. The role of the Roosevelt phase is critical on two counts, although the data from the Rye Creek Project are only peripherally applicable. First, if a substantial Preclassic period population can be demonstrated or, at least, validated, then it must be funneled through the Roosevelt phase both demographically and developmentally to produce the relatively substantial archaeology of the Gila phase. It appears that the Roosevelt phase marks the origins of the Gila phase pattern, with platform mounds, compounds, and (possibly) large pueblos, in a riverine settlement pattern. In most scenarios, the Roosevelt phase will either: mark the transformation of existing Sedentary period populations into the showpiece manifestations of the Gila phase, or mark the origins of a Gila phase that does not continue previous Sedentary period demographic and developmental trends. Either way, the Roosevelt phase is a critical period and severely understudied. Although this review of previous research and culture history has shown that we still have a long way to go, all is not bleak. On a very encouraging note is the fact that the Tonto Basin is poised on a threshold that will soon carry the area into a dramatically different future. There has been a slow accumulation of survey and excavation data through a series of Arizona Department of Transportation projects, the Tonto National Forest cultural resource program, and work associated with the Bureau of Reclamation's Roosevelt Lake Project. These have dramatically expanded the archaeological database, and the ongoing work in the Roosevelt Lake vicinity by Arizona State University (Rice 1990), Statistical Research, Inc. (Ciolek-Torrello et at. 1991), and Desert Archaeology (Doelle et at. 1991), has brought a major, long-term research program to the area. As a result, we will accumulate more data within the next 5 years than have been accumulated in the previous 100. Whether these data will resolve some of the issues raised above, or whether they will just add more fuel to the "controversial" fire, remains to be seen; either way, we will at least be standing on firmer interpretative ground. The following chapter, which presents the Rye Creek Project research design, discusses specific research questions related to these and other issues in more detail. 38 Clwptu 3 CHAPTER 4 RESEARCH DESIGN Mark D. Elson and Wtlliam H. Doelle As the previous chapters on the environment and culture history have discussed, the Tonto Basin is in an area of great diversity. This is reflected not only in the juxtaposition of different environmental and physiographic zones, but, as prior researchers have pointed out, in the possible prehistoric mixing of different cultural traditions and peoples. These factors have played a significant role in previous reconstructions of the prehistoric settlement. That is, along with environmental and, perhaps, cultural diversity there appears to be an equal, if not greater, diversity in the models that have been put forth by various researchers to explain the nature of the prehistoric record. Given the limited amount of formal research that has taken place to date, particularly in the Upper Tonto Basin, many of the previous reconstructions remain untested and highly speculative. This is not to place blame on prior researchers because the sample of sites they were working with was poor at best and insufficient for most inferences. Although the mitigation of 13 sites within the State Route 87 right-of-way will certainly not resolve all of the questions raised by previous research, work undertaken for the Rye Creek project has greatly augmented the data base, particularly for the little known Preclassic period. As a result, the analyses presented here are geared towards an emphasis on basic research. By basic research we mean those research questions that form the primary analytical building blocks necessary to transform speculation into hard data. By first investigating these fundamental research issues, we can begin building the data base necessary to explore some of the more complex questions presented in the previous chapter. This research design was largely written prior to the advent of the data recovery phase. As such it represents potential research problems that could be addressed through data expected to be present within the project area. As in all projects, not all potential data sets were recovered, and additional data were discovered that brought up research questions not addressed by the original research design. For the most part, however, we were able to investigate much of what we had originally planned. The research design is presented here with only minor modifications, stressing those data sets that were recovered over those that were not, and incorporating new data where applicable. DEFINmON OF HISTORIC CONTEXTS An historic context consists of a theme, or research issue, set in a specific time and place. Historic contexts recently have become the framework within which the federal historic preservation process is structured (National Park Service 1986), though many of the details for implementing the process still have to be worked out (Dart and Doelle 1988; Dart 1989). For the present study, six major themes, or in several cases thematic categories (a higher level grouping of related themes), have been identified. They are: contextual assessment, chronology building, subsistence and settlement patterns, community organization, exchange and interaction, and cultural affiliation. These historic contexts can be related to those developed for the Tonto National Forest (Macnider and Effland 1989:8-15) which were published after the completion of the Rye Creek research design and field work. The Tonto Forest themes, which are the approved basis for their dealings with the State Historic Preservation Office, include: demography, social-political-ideological systems, technology and industry, exchange-trade-commerce, subsistence, warfare and militarism, transportation and communication, recreation and tourism, education and art, and government. In this respect, our subsistence and settlement pattern theme is compatible with the Tonto Forest subsistence theme, our exchange and interaction theme 40 Chapter 4 is compatible with their exchange-trade-commerce theme, and our community organization and cultural affiliation themes are compatible with their demography and social-political-ideological systems themes. Although we have chosen to retain the historic context designation for our themes of contextual assessment and chronology, by Tonto Forest criteria these would not be recognized as themes, since they do not relate to the direct evaluation of cultural resources or historic properties. Instead these would be called "logistical issues," in the case of the contextual assessment or be part of determining the "time" component for chronology. The "place" component of most of the historic contexts to be considered is the Tonto Basin, more specifically the Upper Tonto Basin. By definition, the place component includes a consideration of the past environment (Dart and Doelle 1988), however, some of the contexts related to exchange and cultural affiliation, which involve large-scale regional patterns, require a slightly larger spatial scale. The "time" component will be made as specific as possible. All of the sites under consideration are believed to be prehistoric or protohistoric, and current evidence places aU of the sites within the ceramic period. Diagnostic ceramics recovered from the project indicate that all of the 13 sites date between AD. 700 and AD. 1400. One of the prehistoric sites also contains a small protohistoric Apache component, and minor historic components are present at several other sites. mSTORIC CONTEXT #1: ADVANCES IN ARCHAEOLOGICAL METHODS -- THE ASSESSMENT OF CONTEXT The temporal setting for this historic context is the present, and the geographic setting is considered to be the American Southwest, though the "place" component could be considered to be global for the methodological issue addressed here. A single aspect of methodology is considered: the need for a thorough assessment of the recovery context as an integral step in the archaeological research process. As is briefly outlined here, with a more expanded treatment in Chapter 11 of Volume 2, contextual analyses incorporates systematic, replicable, common-sense approaches to dealing with some of the complexities presented by the archaeological record. Karl Butzer (1982) defined the term "contextual archaeology," and much of what is proposed here is compatible with Butzer's formulation, although they are not identical. Of central importance, Butzer (1982) views archaeology not as the study of artifacts, but as the study of relations observable in the archaeological record. Butzer focuses on the need to develop method and theory to extract the significant information that is contained in those relationships, a point that is critical to our formulation as well. Our concept is not as broad as Butzer'S, which is why we employ the term contextual assessment rather than contextual archaeology. At its most basic level, contextual analysis is primarily concerned with developing explicit and replicable methods for evaluating the integrity of archaeological deposits. The development of a method for contextual assessment is a major component of the present research, and an important consideration in all historic contexts defined below. Contextual analysis is a methodological tool through which greater control can be gained over the formation of the archaeological record, therefore increasing confidence in the integrity and subsequent interpretation of archaeological deposits. Our concern with context is based upon the premise that it is only when the mechanisms behind the formation of the archaeological record are understood, that meaningful statements can be made with any degree of confidence. Although the study of archaeological context is by no means new (Binford 1981, 1983; Butzer 1982; Schiffer 1976, 1987), and much of what we propose is still preliminary and of an experimental nature, we believe that the methods developed here represent advances in this process. HISTORIC CONTEXT #2: CHRONOLOGY BUILDING As noted in the previous chapter, Tonto Basin archaeology is sorely in need of chronological refinement. The present assemblage of absolute dates is quite small in comparison to the span of time being dated, and very Research Design 41 few of the dates can be assigned to particularly strong archaeological contexts. Furthermore, the use of decorated ceramics for relative cross-dating has also not been particularly successful, due to a variety of reasons discussed below. As a result, the sequence of occupation in both the Upper and Lower Tonto basins is not well understood. On a positive note, however, is the fact that the Tonto Basin is in a relatively unique geographical position, situated as it is in the transition zone between the desert and the mountains. As such, it contains several important characteristics that serve to increase the dating potential: the presence of treering dated (or potentially tree-ring dateable) intrusive northern ceramics, the presence of Hohokam buffware ceramics, and the potential for recovered wood that may be dendrochronologically sensitive. Both relative and absolute methods for chronology building are considered in the following discussion. Relative Dating Methods An essential element in the successful construction of a relative chronology and the dating of individual features at a site is the identification of temporally unmixed deposits through the contextual analyses. The identification of such contexts is not always simple. For this reason the excavation strategy focused upon contexts such as burned houses, trash mounds, or rapidly filled trash deposits as potential sources for ceramic studies. Superimposed features were another highly valuable context, although very few were noted during the data recovery phase. It is not enough, however, to just focus on high quality excavation contexts, because these can produce temporally mixed data as well. Evaluation of other data such as sherd size, artifact density, and indications of sherd weathering, are all indicators of the relative degree of mixing that may be represented in an archaeological deposit, and they were important variables in the contextual assessment. Decorated Ceramics Unlike much of the rest of the Greater Southwest, the Tonto Basin did not have an extensive tradition of locally produced painted pottery (Wood 1986, 1989). Until the Classic period, beginning around AD. 1150, when a painted ware (the Salado Polychromes or Roosevelt Redwares) is assumed to have been locally produced, the only decorated wares found within the Tonto Basin are intrusive whitewares stemming from the Anasazi and Mogollon areas to the north and east, and buffwares from the Hohokam area to the south. Even during the Classic period, however, intrusive ceramics comprise the great majority of the decorated assemblage. Therefore, intrusive ceramics have been the most common method for establishing the temporal placement of archaeological features and sites in the Tonto Basin, but to date this has not been a particularly productive or reliable method on which to build a chronology. This is due to several factors. For one, intrusive ceramic types, particularly in the Preclassic period, comprise a very low frequency of the ceramic assemblage, and many small sites contain no decorated ceramics at all. This creates obvious difficulties, because the dating of a site based upon only a few intrusive sherds is fraught with interpretative problems. For another, many previous researchers have failed to take into account the archaeological context of the recovered ceramics in assigning dates, instead relying on the entire ceramic assemblage and generally dating the site at the point of ceramic overlap. Although this method is justifiable in some cases, it involves making the assumption that the site in question is single component, short-lived, and never reoccupied. This assumption mayor may not be valid, and needs to be demonstrated before the assigned date can be accepted with confidence. Unfortunately, this is rarely the case. And finally, most cross-dating of intrusive ceramic types relies almost exclusively on Breternitz's (1966) pioneering research on tree-ring dated ceramics, which, although a critically important study, is now largely out of date and contains relatively long life spans for the analyzed ceramics. As a result, due to these and other problems, such as the potential temporal lag in trade ware use-life, several researchers, most notably Schiffer (1982:309-312), have questioned the utility of crOSS-dating. The problems noted above, however, are by no means insurmountable, as the research conducted for this project demonstrates. Through careful control of archaeological context, and a reevaluation of tree-ring dated ceramics using more recent data (e.g. Downum 1988; Douglas 1987; Dean 1991; Halbirt and Dosh 1991), we believe that significant temporal information can be gained through the use of ceramic crOSS-dating. Wallace's research into the dating of the Gila Butte phase, presented in Chapter 24 (Volume 3) of this report, strongly 42 Chapter 4 demonstrates the utility of this method. Furthermore, although small sample size will always be a concern, the problems with this can be minimized if archaeological context is carefully considered. Given the relatively large standard deviations associated with both radiocarbon and archaeomagnetic dating, and the general lack of dateable tree-ring specimens from the Tonto Basin, cross-dating is considered to be by far the best means currently available for building a regional chronology. Plainware and Redware Ceramics Plainwares, redwares, and corrugated wares are the predominant ceramic wares found within the Tonto Basin, comprising over 97 percent of the ceramics recovered from the Rye Creek Project (see Stark and Heidke, Chapter 13, Volume 2). Unfortunately, these wares are not known to exhibit high rates of variation over time, as is often the case with decorated wares. Furthermore, previous attempts at seriating Tonto Basin assemblages based on the relative frequencies of these types have provided little convincing temporal data (Bruder and Ciolek-Torrello 1987; Woodward, et a1. 1985), other than to confirm the obvious fact that Classic period sites have greater percentages of redware and corrugated ceramics when compared to the plainwaredominated Preclassic period sites. In our research we found what we believe to be several dimensions of temporal variability that have not been adequately tapped by previous archaeological analyses. The present study represents a preliminary attempt to document these. Ceramic traits considered in the analysis include relative frequencies of ceramic types, vessel form, temper type, vessel-wall thickness, and surface treatment. Therefore, variation in plainware and redware ceramics also is believed to be a potentially important avenue for gaining temporal control, although not as conclusive, as tightly controlled, nor as well understood, as the cross-dated intrusive decorated ceramics. Absolute Dating Methods A major emphasis was placed on the recovery of high quality samples suitable for absolute dating. Radiocarbon and archaeomagnetic samples were the most common materials, though tree-ring specimens also were collected in the hopes that they could be dated. Once again, however, if the goal is chronology building, it is essential that the archaeological context of the specimen to be dated be controlled to the greatest extent possible. Dean (1978, 1991) and Schiffer (1976, 1982, 1986) have provided valuable discussions of issues related to absolute dating of past behavior and archaeological context. Despite the fact that these problems are widely recognized, poor control over the context of samples that are submitted for absolute dating continues to be a major problem that inhibits effective chronology development in archaeology. Furthermore, due to problems with the precision of radiometric dating, multiple dates from the same context are needed to confidently evaluate the accuracy of the assigned dates. That is, a single radiocarbon date cannot be evaluated because there is no comparative absolute basis for accepting or rejecting it (Jeff Dean, personal communication, 1990). Unfortunately, given the cost of radiometric dating, and the lack of understanding by archaeologists, single dates are the norm rather than the exception. The Tonto Basin has only a small number of absolute dates, and as reviewed in the previous section and in Chapter 25 of Volume 3, it is apparent that many of these are contradictory and of limited or no value. Whether a lack of context is symptomatic of the problems with the dates derived from these samples, or whether other problem-causing factors are present, is currently unknown. If the number of absolute dates for the Tonto Basin was to be increased on this project, however, we felt that it was essential that only samples from well-controlled contexts be submitted for dating, even if that meant that some contexts and sites could not be absolutely dated. All features that were considered to be potentially dateable through archaeomagnetic analysis were sampled, resulting in the recovery of 28 archaeomagnetic samples. Unfortunately, all of the collected radiocarbon samples were found through later analysis to be either in poor or ambiguous archaeological context, or ethnobotanically unsuitable for dating purposes. Ethnobotanist Charles Miksicek examined all wood samples that were to be submitted for radiocarbon analysis and found that the majority were composed of inner heartwood, which in a long-lived species such as juniper and pinyon pine (the two most commonly recovered Research Design 43 wood species) can produce inaccurate dates; the dates would reflect an unknown time when the tree was growing rather than when it was culturally used. As a result, no radiocarbon samples were submitted for analysis. In addition, seven potential tree-ring samples, all from juniper posts or beams found on pithouse floors, were submitted for dating. Although most of the samples had more than the 30 requisite rings needed for temporal placement, they unfortunately could not be matched with any known chronology (Jeff Dean, personal communication, 1990). Therefore, archaeomagnetic dates comprise the sole basis for the absolute dating of the Rye Creek sites. mSTORIC CONTEXT #3: SUBSISTENCE AND SETTLEMENT SYSTEMS Subsistence and settlement systems is a thematic category comprised of a variety of important themes. To date, very little work has been undertaken in the Upper Tonto Basin on settlement-pattern analysis, and for the most part the overall settlement systems and subsistence strategies are unknown. The present project has contributed new and significant information related to a number of these themes. This is particularly true for three important reasons. For one, the project sites are located within the Rye Creek drainage, which is a relatively well-defined catchment area, a large part of which has been already archaeologically surveyed. The "filtering" factor of Roosevelt Lake, situated more than 30 miles to the south, is not a consideration. The second, and perhaps most significant factor, is that the Rye Creek drainage area appears to be relatively free of complicated, overlapping, multiple settlement systems such as those found within the Lower Basin. Finally, the sites within the project area cover a relatively long temporal period, spanning the Gila Butte phase (AD. 750-850) through the early Classic period (AD. 1150-1300); if Rye Creek Ruin is included a late Classic period (AD. 1300-1450) component is present as well. As a result, both synchronic and diachronic views of the settlement and subsistence systems can be modeled. These data in tum can be compared to data recovered from the Lower Tonto Basin and elsewhere in the Southwest to provide a more complete picture of the prehistoric use of the Rye Creek area. Subsistence Strategies Previous research in both the Upper and Lower basins has established the presence of cultigens (primarily com) at sites along Tonto Creek (Haas 1971; G. Rice 1985) and at upland sites (Hucke1l1978; Ciolek-Torrello 1987). The present project has provided an excellent opportunity to examine a variety of environmental settings to assess the extent to which different localities were utilized for food production. Site settings range from relatively large and small settlements adjacent to the floodplain of the major drainages such as Rye and Deer creeks, to smaller sites on secondary drainages such as Clover Wash, as well as a variety of sites adjacent to very minor drainages that mayor may not have offered any advantages to food producers. A large suite of subsistence samples was recovered from this diversity of settings in order to establish the relative importance of food production at each, as well as to provide important clues about site function. These analyses are detailed in Volume 2 in Chapters 18 through 21. In addition, a single agricultural field system (AZ 0:15:92) is present within the project area. This system, although partially disturbed through root-plowing, is situated directly south of the large pueblo site of AZ 0:15:54 and may have served as the agricultural component. The system, although not large, has at least 10 checkdams, and may have had more that were disturbed through the root plowing or the construction of State Route 87. Many of the small isolated sites in the project area may have served as bases for wild resource procurement and processing activities, as were the majority of sites in the Ord Mine project area to the south (CiolekTorrello 1987). It is also clear that a wide variety of wild plant and animal resources would have been within easy access of the larger settlements as well. Given the environmental diversity and high productivity of the Upper Tonto Basin, it would have been possible for wild resources to have comprised a major dietary element of the local occupants. Our research has focused upon the identification of the specific resources used and with providing an assessment of their relative contributions within the subsistence economy. 44 Chapter4 Change in Subsistence Strategies Major differences in the ways in which upland and riverine environments were utilized have been noted by previous researchers (G. Rice 1985, 1990; Ciolek-Torrello 1987). The present project also provides an excellent opportunity to investigate the manner in which these strategies changed through time. This is particularly true given the long temporal range of project area sites. Changes in food production strategies have been correlated with changes in sociocultural complexity in other areas of the Southwest (Huntington 1986; G. Rice 1987, 1990) and Mesoamerica (Wilk and Netting 1984). The key contrast is the apparent riverine focus of the early ceramic period occupants of the area, whereas the later, Classic period occupants show a greatly expanded utilization of upland resources. These patterns have a variety of potential implications for the prehistory of the Tonto Basin. Such a pattern might support arguments of population growth forcing the utilization of more marginal environments, for example. Alternatively, the differential archaeological visibility of the remains of Preclassic versus Classic period settlements must also be considered, as should the possibility that different logistical strategies could have been used by the two groups to exploit similar kinds of resources. Settlement Systems There is a clear range of variation in the settlement types that are present within the State Route 87 right-ofway. Some of the smallest sites were almost certainly seasonal and served as locations for only a limited range of activities, whereas the largest sites in the project area appear to represent small hamlets that may have been occupied for most of the year. Rye Creek Ruin, a large, presumably permanently occupied village, and one of the largest sites in the Tonto Basin, is situated less than a kilometer east of the project area. Three lines of evidence were used in the reconstruction of settlement systems: evidence for seasonal versus year-round occupation; the degree of functional specialization indicated by artifact and subsistence data and other data classes; and evidence of site size and intensity of occupation. Although, the length of site occupation is perhaps one of the more difficult aspects of the prehistoric settlement to assess, a detailed analysis of seasonality was undertaken for this project. The success (and failures) of this analysis are detailed in Chapter 26 (Volume 3). In addition, to place the sites within the right-of-way in a larger context, information from recent areal surveys by the U.S. Forest Service was employed, as well as additional supplemental surveys conducted by Desert Archaeology. Wood et al. (1989:21) report that over 166 square miles have been intensively surveyed within the Tonto National Forest as of two years ago, including nearly 40 square miles within the Tonto Basin subarea. Most significantly, a large part of this survey has been focused within the two townships that include the project area, where over 50 percent of the included sections have been surveyed or sampled in some manner (Macnider and Effiand 1989:125-126). For the survey data set, it was necessary to depend more on evidence of site size and intensity of occupation in order to classify sites into types. Information from the present project as well as previous excavations (Ciolek-Torrello 1987; Haas 1971; Huckell 1978; Jeter 1978; G. Rice 1985; Hohmann 1985) are used as a basis for inferring subsurface remains likely to be present at unexcavated sites. Settlement Distribution Two key issues are of concern regarding settlement distribution: the spatial relationship between settlement types and resource distributions, and the spatial relationship among settlements. A third dimension is how the settlements are structured chronologically, and whether changes in settlement structure are occurring over time. It is believed that some of the smaller or limited-function settlement types have a very direct spatial relationship with key resources. Furthermore, it is increasingly clear that human populations take into consideration the distribution of other human groups when making settlement location decisions (for example, most of the large sites along Rye Creek are in visual contact with Rye Creek Ruin). Survey data from the U.S. Forest Service, along with additional survey within key areas, make up a very important data set in addressing this theme. Research Design 45 mSTORIC CONTEXT #4: COMMUNITY ORGANIZATION This broad thematic category could be divided into several components, but it was decided that they were most appropriately combined for this study. Three themes are identified for consideration on this project. Population Growth Population growth is a key theoretical issue in current anthropology (e.g., Boserup 1965; Hassan 1978; Spooner 1972), and it is of specific concern to an understanding of the prehistory of the Tonto Basin. Two general topics merit discussion here: the wColonial expansion" into the Tonto Basin and the apparent population increase in the Tonto Basin during the Classic period. There has been a ready acceptance by archaeologists that the widespread distribution of Gila Butte phase pottery, pithouse styles, and in some areas ballcourts and mortuary customs, is evidence of a population expansion by Hohokam groups out of the Salt and Gila river areas into more peripheral areas (e.g., Doyel and Elson 1985; W.S. Gladwin and H.S. Gladwin 1935; Haury 1932; I. Kelly 1978; Masse 1980; Wood and McAllister 1984). Recently, Craig (1989) has asked the very obvious question: Could this area have generated all of the "colonists" that archaeologists have assumed were expanding into surrounding areas? Using archaeological data for the Pioneer period in the Salt-Gila area, Craig (1989) used assumptions of constant growth rates of 0.1 percent and 0.2 percent per year in order to obtain a preliminary answer to this question. His conclusion was that the Salt-Gila area could not have populated even the Tucson Basin with such growth rates, despite the fact that such rates were found by Hassan (1978) to represent relatively high rates among primitive populations. The Tucson Basin has a relatively clear cultural connection to the Salt-Gila Hohokam core area, and there is widespread agreement among archaeologists that the Tucson Basin was part of the Hohokam regional system; if there was an extensive Colonial period expansion involving migration to outlying areas, the Tucson Basin would have been a part of it. Given these results, then, it seems much more appropriate to consider the Hohokam "colonial expansion" to represent the rapid dissemination of a material culture complex that was probably tightly linked to an ideology that involved at least a specialized mortuary complex and the ballcourt system. In most cases this expansion represents an adoption of this cultural complex by existing populations, although limited migration may be occurring as well. What are the implications for such a model of the "Colonial expansion" for the Tonto Basin? Trends in recent survey and excavation data from the Tonto Basin and in other parts of southern Arizona provide some useful insights. First, there is increasing evidence of an Archaic occupation in the Tonto Basin (e.g. Ciolek-Torrello 1987; Huckell1973, 1978), and Wood (personal communication, 1988) reports that there are several sites with Snaketown phase diagnostics known from the eastern Tonto Basin. Our excavation of the Deer Creek site (AZ 0:15:52) documented a strong Gila Butte phase occupation with at least some occupation dating to the preceding Snaketown phase. In addition, supplemental survey by Desert Archaeology along Upper Rye Creek located a second Upper Basin site with Snaketown ceramics, and Tonto Forest survey crews recently located a third. Given the very low frequency of the Pioneer period diagnostics, even at sites along the Salt and Gila rivers, we believe that it is appropriate to place substantial weight on even very low frequencies of early diagnostics in the Tonto Basin. This is further supported by the fact that the intensity of survey and surface collection in the Tonto Basin has been very low. As a comparative example, Doyel (1984) was still arguing relatively recently that the Tucson Basin represented an "empty niche" during late Archaic times. Research since then has shown sites to have been relatively abundant during the Late Archaic (Dart 1986; Doelle 1985; Elson and Doelle 1987a; Huckell and Huckell 1984; Fish et al. 1986; Roth 1987). Fish et al. (1988) have recently pointed out that the discovery of the majority of these Late Archaic deposits have taken place while trenching for Hohokam period remains. Thus, there is strong reason to expect that what is now only a minor or incipient trend in the Tonto Basin data set will intensify and become much clearer with increased fieldwork in the area. Based on the these data, it is a reasonable working hypothesis that the "Colonial expansion" into the Tonto Basin represents primarily 46 Chapter 4 the incorporation of an existing local population into the Hohokam regional system. The present project has shed further light on this issue since two sites (AZ 0:15:52 and AZ 0:15:1(0) contain Gila Butte phase material. Data from the site of Ushklish (Haas 1971), another Gila Butte phase site in the Upper Tonto Basin, were also reexamined as part of this project. Furthermore, it is important that this working hypothesis be taken into consideration when considering the cultural or ethnic affiliation of the early inhabitants of the Tonto Basin, a point discussed under Historic Context #6. This hypothesis would allow for a greater degree of variation in the Hohokam cultural pattern of the Tonto Basin, for example, because it does not assume that an actual population that was adapted to the riverine environment of the Salt and Gila moved into this new area. Rather the local inhabitants would already have in practice an indigenous adaptive strategy and they would have adopted the elements of the Hohokam cultural system within the framework of that existing adaptive pattern. Moreover, our data suggest that the nature of this interaction was not static and clearly changed over time. It now appears that Hohokam contact and influence were strongest during the pre-AD. 900 period; after this time interaction with the Hohokam decreased and was gradually replaced by contact with Anasazi and Mogollon groups to the north and east. Another area where the general consideration of the issue of population growth has relevance is during the Classic period Roosevelt and Gila phases (AD. 1150-1450). Wood (1986:15) recently estimated that the Tonto Basin may have had a population of roughly 10,000 persons during the fourteenth century. Accepting this as a reasonable estimate, this would mean that the local Pioneer or Colonial period population would have had to have been about 5,500 or 3,000, if constant growth rates of 0.1 and 0.2 percent, respectively, are assumed. The current settlement pattern data for the Tonto Basin, even though incomplete, does not seem to support population numbers as high as 5,500 for the late Pioneer period, and the 0.2 percent growth rate is not likely to have been sustained over 600 years, which makes the second number untenable as well. Therefore, it would appear that population movement into the Tonto Basin is likely to have occurred during the Classic period. An alternative possibility is that Wood's population estimate is too high. Only substantial refinement of the regional data base will provide a basis for an empirically based estimate that can be better supported than Wood's assessment. The present project has provided important new demographic data on a much more local scale. Our research has delved into such variables as seasonality, site occupation span, and population mobility, in order to make reasonable inferences regarding past population from settlement pattern data. This basic information about population structure is essential in order to develop credible methods for transforming settlement pattern data into population estimates. The Role of the Household In Community Organization The data recovery phase yielded an abundance of data on houses and associated features, and recent research in both pithouse and pueblo archaeology has found that these sets of features frequently can be identified as the domains of individual households or courtyard groups (Elson 1986; Howard 1985; Wilcox et al. 1981; Lowell 1988; Lipe and Hegmon 1989). As Netting et al. (1984) point out, the household is a culturally defined concept, but unlike the family, which is defined by rules of kinship, the household is defined behaviorally. This behavioral focus makes the household a particularly appropriate unit of archaeological analysis. Because the household is the level on which people interact most directly with their natural and social environment, change in the size and structure of households often provides some of the clearest and most accessible information on changing patterns of adaptation, social integration, and population. In the Tonto Basin in particular it is important to focus research on the household, because in the past house types have been used as evidence of cultural affiliation. In the studies of the household cited above, however, there are clear indications that variation in house type is often related to the function of a structure within the set of facilities utilized by a single household. Therefore, consideration of such functional variation is a prerequisite to the use of architecture as evidence of cultural affiliation or variation. The data recovery phase excavations strongly suggest that all of the pithouse sites within the project area are structured in some manner, and several are probably organized by households or courtyard groups; some sites Research Design 47 represent small, short lived, settlements, containing only one or at absolute most, two households, while others are larger, and contain multiple households. Unfortunately, the largest pueblo site, the Cobble site (AZ 0:15:54), which probably was a small village or hamlet, was completely destroyed by root-plowing and very little architectural information was recovered. The remaining pueblo sites consisted of single-room fieldhouses. With the exception perhaps of G. Rice (1985) and Ciolek-Torrello (1987), almost no research has been undertaken in either the upper or lower basins on intrasite structure at the household level. This is due primarily to the nature of the excavated sample; many sites simply are not appropriate for this type of analysis. The excavation of sites containing only a few households, however, such as those present within the project area, is extremely amenable to this research. Small sites allow for a clearer look at intrasite structure than multicomponent sites, which generally are occupied longer and contain a fair amount of superposition and disturbance. Therefore, the excavation and analysis of these sites has the potential to add significant new information on intrasite structure within the Tonto Basin and in southern and central Arizona in general. Community Structure and Integration Wood (1986, 1989) and G. Rice (1990) have developed rather detailed, and somewhat different, models of the structure of late Classic period platform mound communities in the Tonto Basin. At issue is the scale and nature of the perceived complexity and the primary mechanisms behind community control and integration, although there is general agreement that platform mounds played a critical role in these processes. Rye Creek Ruin, which is just a short distance from the present project area, is one of the largest Classic period communities within the Tonto Basin. It is highly likely that any Classic period settlements found within the project area were once components of the larger Rye Creek community. There is very little in the existing literature as to what Preclassic period community structure may be like in the Tonto Basin. It is clear, however, that the relatively small settlements that appear to be the norm in the Tonto Basin in Preclassic times could not have functioned independently. Rather, interaction with one or more neighboring settlements would have been the norm. The exact nature of that interaction and the mechanisms used to integrate the Preclassic community are simply unknown at present, although the present project has added some intriguing data concerning this interaction. mSTORIC CONTEXT #5: EXCHANGE AND INTERACTION The study of prehistoric exchange has been of great interest to archaeologists since the beginnings of the discipline, and recently there have been great strides made in method and theory in this area (Bishop et al. 1982; Earle and Ericson 1977; Ericson and Earle 1982; Fry 1980; Plog 1986; Renfrew 1975; P. Rice 1984; Weigand et at. 1977). Prehistoric exchange is a difficult topic to address and one that frequently leads to highly divergent inferences from a single data set. In many ways this is due to the difficulty of transforming archaeological data into behaviorally meaningful constructs. For one archaeologist the presence of a few nonlocal sherds is the basis for positing substantial trade contacts, while for another the low sherd frequency is interpreted as evidence of very low level trade and indirect contact. This has been particularly true of analyses within the Tonto Basin where the same data sets have been interpreted in numerous ways (cf. G. Rice 1985; Neitzel 1985; Wood 1986; Whittlesey and Reid 1982; and Ciolek-Torrello 1987). Archaeologists generally deal with very biased samples from past behavioral systems, and generally only a few relatively rare material culture classes are amenable to sourcing with any degree of precision. Therefore, it is very difficult to estimate flow rates of exchange items in the past and thereby to make realistic assessments of the relative importance of those items in a prehistoric system of production and exchange, although recent advances have been made in this area (e.g., Wallace and Heidke 1986; Wallace et al. 1991). These problems must be overcome if real progress is to be made in this area, and the theme of ceramic production and exchange is an important component of the present research. 48 Chapter 4 Steve Plog (Braun and Plog 1982; Plog 1980, 1986) has challenged some of the common assumptions of Southwestern archaeologists in a series of articles over the past decade. Plog notes that most archaeologists use the "criterion of abundance" to support an assumption that each village had a high degree of autonomy and that pottery production, in particular, was carried out at the village level. Plog notes that there are major demographic constraints that will tend to force small settlements to develop regular patterns of interaction with other similar settlements. Exchange relationships are generally an important element in establishing and maintaining the social relationships that are necessary to ensure access to spouses as well as the material necessities of life (cf. Wobst 1977). Ceramics, and particularly decorated ceramics, have generally been the most common avenue for looking at issues of prehistoric trade and exchange. This is particularly applicable in the Tonto Basin, where nearly all of the decorated wares in the Preclassic period, and the majority of the decorated wares in the Classic period, were imported. As discussed in Chapter 12 of Volume 2, analysis of the source areas for these wares has strongly indicated that interregional exchange networks were extremely significant and changing over time. On a more regional level, the absence of a major tradition of locally produced painted pottery in the Upper Tonto Basin means that it is necessary to look to the redware and plainware assemblages for information about ceramic production and exchange. A program of petrographic analysis was undertaken as a means of beginning to address this issue. The methods and results of this analysis are presented in detail in Chapter 13 in Volume 2. Although this was originally conceptualized as a pilot study, our research strongly suggests it is highly productive method for dealing with issues of intra regional eXChange within the Tonto Basin. Furthermore, this technique has even greater potential for a long-term payoff given the current research being conducted in the Roosevelt Lake vicinity. Another commodity that appears to have played a major role in the Upper Basin exchange networks is argillite. Argillite, a soft, reddish brown mudstone or siltstone (Gundersen and Tiffany 1986:46-48), is known from only four or five source areas in the Southwest, one of which is within the project area along Deer Creek. Sites within the project area contained an abundance of both worked and unworked argillite. Argillite was used as a source for the carving of beads, bowls, censers, and other artifacts, and, as our research indicates, as a red-colored ceramic pigment. Through the process of X-ray diffraction, all of the known argillite sources in the prehistoric Southwest have been found to be mineralogically distinct, and therefore the raw material source areas of the argillite artifacts can be determined. Like the petrographic analysis, the investigation into the movement of argillite goods and pigments was conceived as a pilot study, because a full analysis is beyond the scope of this project. The data indicate, however, that not only is a project of this nature highly feasible, but that argillite was moving through the southwest in appreciable quantities and played an important role in Tonto Basin exchange networks. The results of the argillite sourcing analysis are presented in Chapter 22 (Volume 2). A final topic is the transportation of goods through the Tonto Basin. Wood (1985, 1986) has proposed that the Tonto Basin served as a major thoroughfare for the transportation of material goods and commodities between the Pueblo peoples of the north and the desert-dwelling Hohokam and Mesoamerican peoples to the south. Given the distribution of intrusive trade goods throughout the Southwest, such as ceramics, copper bells, macaws, exotic feathers, obsidian, turquoise, and shell, it is obvious that trade and exchange were major components of a pan-southwestern interaction network that extended north and south as well as east and west (Crown 1990; Doyel 1979, 1987; McGuire and Downum 1982; Wilcox 1987). Prehistoric trails and exchange networks are known from almost all areas of the Southwest. These include, for example: the New River area, where north-to-south exchange in ground stone has been petrographically documented between the New River and the Hohokam core area, and east-west prehistoric trails are known to connect the New River with the Agua Fria River (Doyel and Elson 1985); the Kayenta Anasazi and the Sinagua areas (McGuire and Downum 1982); the Gila Bend area (Dart et al. 1989); and between the Hopi Mesas and the Grand Canyon (Bartlett 1940), to name but a few. Wood (1986:17) has suggested that the inhabitants of the Tonto Basin may have specialized in the transportation of these goods, because the Tonto Basin itself lacks the necessary raw material resources, with the exception of argillite and poSSibly steatite, to have served as a primary procurement area. He proposes that several major trade routes ran through the Tonto Basin, and that the large Salado period Research Design 49 platform mound sites played an important role in the integration and organization of this trading network (Wood 1986: Figures 14 and 15). This has significant implications for the reconstruction of the prehistory of the sites within the project area, because Rye Creek Ruin, a platform mound site of major importance, is without a doubt the focal point of the Classic period project area community system. Whether the growth of Rye Creek Ruin and its surrounding community system was based at least partially upon its role in the transportation of goods and commodities is an important aspect of the present research, as is the nature of the preceding Preclassic period exchange networks. mSTORIC CONTEXT #6: CULTURAL AFFILIATION Questions regarding the cultural affiliation of the prehistoric groups in the Tonto Basin are among those most commonly raised. Were the original Ceramic period occupants of the Tonto Basin affiliated with the Sinagua, Anasazi, Mogollon, or Hohokam? Did this represent an actual migration of peoples or was it primarily the adoption of outside traits by a local population? Did the Salado cultural tradition originate in the Tonto Basin or elsewhere? The limited amount of research that has been conducted in the Tonto Basin is one of the factors that makes it possible for a diversity of opinions regarding cultural affiliation to flourish. When the initial research design for the data recovery phase was written, we favored the hypothesis that the Tonto Basin showed the strongest cultural affiliation with the Hohokam. The data recovered during the course of this project, however, makes us far less confident that we know the answers. It now appears more and more likely, at least in the Upper Tonto Basin, that we are dealing with an indigenous people who initially may have been affiliated with the Hohokam, but who changed their affiliation through time, first toward the north and then towards the east. The issue is far from clear cut, and much more data need to be accumulated before the problem can be addressed with any degree of certainty. In fact, we are firmly committed to the development of a stronger empirical base and explanatory framework for addressing this issue before making any conclusions. As a result, the question of cultural affiliation, though perhaps fun to speculate on (as indicated by the numerous researchers who have through the years), is probably premature. Therefore, we are in strong agreement with Wood and McAllister (1982) when they say: it is necessary to first define with greater specificity the material inventories of each group, then to identify those aspects which reflect cultural rather than technological or environmental selection, and finally to identify exchanges of such cultural commodities, the nature of the exchanged items or behaviors, the directions and rates of exchange, and their extent within different segments of each population [Wood and McAllister 1982:93]. Ceramic and petrographic studies undertaken in the course of this project have contributed to these issues. Other material culture items that merit consideration along these lines are architecture, frequency of decorated ceramics, and the presence or absence of other artifact types. The analysis of plainware and redware vessel form is believed to be a particularly illuminating and largely untapped source for assessing cultural affiliation as these "everyday" use items may be more culturally meaningful than the low frequencies of decorated wares. Consideration of strategies of community integration, and the study of long-term developments in mortuary practices including the physical anthropology of the human remains, also will contribute substantially to resolving this issue. Unfortunately, the question of cultural affiliation is a topic for which archaeological methods are not well developed. As a result, this problem could not be definitively resolved on this project, although implications of the data are considered. Progress has been made, however, in building the necessary data base needed to address these issues in the future. ADDmONAL ISSUES: PROTOHISTORIC AND mSTORIC COMPONENTS Information to be derived from protohistoric or historic remains within the project area is extremely limited, although not totally insignificant. A single site, the Deer Creek site (AZ 0:15:52), contains a small component of protohistoric or historic remains in the form of a few scattered Apachean ceramics, an Apache 50 Chapter 4 pot break, and a possible mescal-roasting pit, while one site, AZ 0:15:71, contains historic Anglo remains. Although these remains in themselves are not enough to allow for the reconstruction of Apache and (early?) Anglo use of the project area, they do document their presence and allow for at least some preliminary interpretations. Of these, the Apachean component of AZ 0:15:52 is perhaps the most promising. A small number of Apache and Yavapai ceramics were recovered from both the surface and subsurface components of the site, and a large, partially buried roasting pit visible on the site surface may be Apache-related as well. Unfortunately, not enough charcoal was recovered from the roasting pit to date it, although all indications suggest it is a slightly colluviated Apachean mescal pit. Apache mescal pits have been well documented in the surrounding area, and historically the Rye Creek area was the home of one of the six semi-bands of the Southern Tonto Apache (Goodwin 1942). In addition, a small locus of Apache material, in the form of a single pot break, is present outside of the right-of-way along the western boundary of the site. A description of this material with an accompanying discussion of the Apache occupation of the Tonto Basin is presented in Volume 3 in Chapter 23. The historic components are even more ephemeral and ambiguous than the Apachean component. Site AZ 0:15:71 contains two rubble-core constructed checkdams that appear to be historic in nature. These checkdams may have been built to prevent the erosion of a dirt road that crosses through the eastern portion of the site outside of the right-of-way. A portion of this same road also cuts through a masonry structure at site AZ 0:15:70 and is at least partially responsible for the destruction of this feature. The origin and date of construction of this road is currently unknown, because no diagnostic artifacts were recovered, but it is possible that it represents the original Payson-to-Phoenix highway (which also cuts through the western portion of Rye Creek Ruin). SUMMARY This review of the research design has identified six basic research questions, or historic contexts, that were investigated during the course of this project. These are: Contextual analysis, chronology, subsistence and settlement systems, demography and community organization, eXChange and interaction, and cultural affiliation. As should be apparent from the previous discussion, all of these issues are highly interrelated, with perhaps the exception of contextual analysis, which is more of a methodological tool (although a critical one). That is, each research question builds upon data from the previous research question, and these are arranged from the specific to the general. In this sense, it is necessary to first control the archaeological context, before determining the chronology, which is necessary to answer questions on settlement, needed to address community organization, demography, interaction, and so on. It is only by first considering these more basic issues that questions related to cultural affiliation, the final research question, can be addressed with any degree of confidence. CHAPTER 5 PROJECT METHODS Mark D. Elson Fieldwork conducted by Desert Archaeology, Inc., on the Rye Creek Project was divided into two phases: a testing phase followed by a data recovery phase. Prior to the testing phase an archaeological survey of the project area was undertaken by Lyle Stone (1986) of Archaeological Research Services, Inc. A general overview of the methods used for both the testing and data recovery phases are presented below. Specific siteby-site and feature-by-feature methods are detailed in the testing report (Elson and Swartz 1989a) and in the individual site descriptions presented in Chapters 6 through 10 of this volume. TESTING PHASE The testing phase was structured to determine the significance and National Register eligibility of the 19 sites identified as being within the proposed right-of-way by the archaeological survey (Elson and Swartz 1989a). Although the specific methods varied by site, in general the methods were relatively standardized and are described below. Brushing Brushing was the first step in the investigation of the sites within the project area. This was because nearly all of the sites were covered with varying amounts of juniper, mesquite, cat-claw acacia, crucifixion thorn, Christmas cholla, holly, and prickly pear. At some sites the vegetation was so heavy that surface features could not be defined, or even seen, until the vegetation was removed. Site Gridding, Mapping, and Surface Collection After the brushing was completed a 20-m by 20-m grid was established with a transit and tape over the surface of almost every site. This was undertaken to facilitate the surface collection and the laying in of the backhoe trenches, as well as for the mapping of features and site boundaries. The exceptions to this were sites AZ 0:15:95 and AZ 0:15:97 that contained extremely low-density artifact scatters and were collected as single units. Site boundaries were mapped either with the transit and stadia rod, or through the grid system. At each site the proposed right-of-way was shot in. At sites where it was possible the nearest centerline station also was shot in to allow for the accurate placement of the site on the ADOT right-of-way map. Once the grid was established the site was collected. This first involved a systematic investigation of the site and the flagging of the site boundaries for later mapping purposes. The surface collection was undertaken for each 20-m by 20-m sample unit in the following manner: 52 Chapter 5 1. Field personnel walked straight lines across each unit spaced at 5.0-m intervals so that four transects per unit were collected. The direction the field crew walked varied, but was consistent within each unit. Artifacts were collected within a 2.0-m wide transect; that is, 1.0 m on either side of the 5.0-m spaced transects. Both sides of the sample unit were pinflagged every 5.0 m so that the crew member could see the destination point and stay on-line. 2. Each crew person collected all sherds greater than 2.5 em in diameter (roughly the size of a quarter), all lithic tools, all ground stone, all shell, and any other rare artifacts encountered such as figurines, worked argillite, or tabular knives. Crew members did not collect lithic debitage. Human bone was noted and recorded but not collected. Material recovered from the individual crew collections was combined and bagged separately by unit. 3. The crew supervisor conducted a complete surface collection within his 2.O-m-wide transect. This included the same artifacts collected by the crew members plus all lithic debitage. This transect served as the control transect and was bagged separately from the artifacts collected by the crew members. A form characterizing each unit was filled out by the crew supervisor. The form described the vegetation cover, presence or absence of disturbance factors, whether any pot holes or cremated human bone was present, the number and types of cultural features, and whether the control transect appeared to be representative of the unit as a whole. The relative proportion of the unit surface that was obscured by vegetation, sheetwash, arroyos, deposition, and modern debris was ranked in five stages from 0 to 100 percent, and the artifact density and sherd-to-lithic ratios were qualitatively determined. Finally, a sketch map of the unit was drawn showing the locations of pot holes, features, cremated bone, and erosional/depositional areas. The collection strategy provided a systematic sample of the site artifact assemblage that ranged from 40 percent for items of high information value, such as ceramics and lithic tools, to as low as lO percent for items of lower information value such as lithic debitage. The 40 percent sample is due to the collection of 160 square meters (four 2-m-wide by 20-m-long transects) within each 400 square meter unit, while the lO percent sample comes from the fact that only the crew supervisor (a single 2-m-wide by 20-m-long transect) collected lithic debitage. This collection method has been tested by Desert Archaeology on numerous projects (Craig and Wallace 1987; Dart 1987; Elson and Doelle 1986, 1987b) and found to be statistically valid for making intrasite comparisons between units (Altschul 1986:25-30). Backhoe Trenching and Feature Profiling Once the surface collection was completed, backhoe trenches were laid out by the collection crew using the grid system and a tape. Generally the trenches were placed in the transect collected by the crew supervisor, because this transect was totally collected, although due to vegetation or terrain problems this was not always possible. Trenches were spaced at 20-m, 15- m, lO-m, or 5-m, intervals, depending on the surface artifact density, vegetation, and the site layout. At sites where no subsurface features were discovered through trenching at 20-m intervals, additional trenches were then placed between trenches at lO-m intervals. This interval was later reduced to 5 m during the data recovery phase to insure that at all sites the subsurface was truly sterile. Once the trenches were excavated both sides of the trench walls were faced and the trenches were inspected for cultural features. If subsurface features were found, they were given a feature number, profiled, photographed, and a feature form was filled out. At every site with surface or subsurface features a vertical datum point was selected and was assigned the arbitrary elevation of 10.00 below datum. This was used to vertically relate all of the features within a site. Project Methods 53 Trench forms were then filled out for the majority of trenches regardless of whether subsurface features were present or not. These forms were used to characterize the natural sediments within each trench and to describe any subsurface cultural material not associated with cultural features. Hand-Dug Trenching and Wall Clearing Hand-dug trenches measuring 0.50 m wide were excavated across all masonry structures. Depending on the feature, the trenches were excavated in either 10-or 2O-cm arbitrary levels or natural layers. All dirt was screened through Y4-inch mesh. Forms were filled out by level and for the feature as a whole. Walls were cleared of vegetation and dirt by scraping above them to define the structure outline. Finally, each structure was mapped and photographed. Testing Phase Results Of the 19 tested sites, 13 were determined to contain significant resources within the proposed right-of-way and to meet criteria for inclusion in the National Register of Historic Places. The six sites that were determined to not be eligible for inclusion in the National Register of Historic Places (AZ 0:15:51; AZ 0:15:93; AZ 0:15:94; AZ 0:15:95; AZ 0:15:97; AZ 0:15:98) contained either low density artifact scatters confined to the surface, or surface features situated outside of the right-of-way (see Table 1.1). In addition, extensive disturbance through root-plowing was discovered to be present at four (AZ 0:15:54, AZ 0:15:92, AZ 0:15:95; AZ 0:15:100) and possibly five (AZ 0:15:53) sites. A map of the root-plowed areas and the disturbed sites is presented in Figure 2.2. DATA RECOVERY PHASE The data recovery methods were designed to recover the information needed to address the research problems presented in the preceding chapter. As with all archaeological projects, however, a certain degree of flexibility was necessary in order to respond to unexpected situations or new and significant data bases. The testing program was designed to maximize coverage in order to minimize the possibility of "archaeological surprises," but contingency measures, such as alterations in the sampling strategy, were employed when necessary. It was considered neither necessary nor efficient (due to data redundancy and cost) to completely excavate every feature present at every site; certain data classes are clearly more important than other data classes. Therefore, a site-specific sampling strategy was implemented based on the results of the testing program that included the consideration of such aspects as the site chronology, site function, site size, distribution and nature of features, range of feature types, and the condition of the site. As a result, although the sampling strategy called for the investigation of all sites, this was undertaken at varying degrees of intensity on a site-bysite basis, and did not include the excavation of every feature at every site. This involved first the determination of the sample universe through extensive backhoe trenching at 5-m intervals. Once this was accomplished a sample of that universe was selected for excavation. Basically, the sampling strategy involved the investigation of all major features, such as structures and trash mounds, through partial or complete excavation, while confining the investigation of other features, such as small pits, trash pits, and trash concentrations, for example, to a certain nonredundant percentage. Therefore, the data recovery plan was structured to allow for the reconstruction of intrasite structure through the excavation or sampling of all major site features, without requiring total site excavation. The amount excavated varied by site. As a result, although this chapter describes the general methods used at all sites, the individual site and feature descriptions presented in the following chapters should be consulted for specific situations and greater detail. 54 Chapter 5 Feature and Strata Designations Two types of cultural features were defined: primary features and secondary features. Primary features are major feature types, such as structures, extramural pits, and trash mounds, for example, that can be defined functionally by their nature alone. They often contain secondary features. Primary features were assigned sequential feature numbers within each site. Secondary features, on the other hand, are contained within primary features and are related to the function of the primary feature. These include, for example, all features within a structure, such as hearths, internal storage pits, and postholes. Secondary features were assigned sequential feature numbers based on the primary feature number, such as Feature 10-1 through Feature 10-8, where Feature 10 was the primary feature and it contained 8 secondary features. It should be noted that intrusive features are considered to be primary features because they are not related to the function of the primary feature they are intruding into. A set of standardized strata designations were used on this project. These are presented in Figure 5.1. For the most part, these strata are the same as those used on previous Desert Archaeology projects, allowing for comparative data bases to be constructed. With the exception of Stratum 19, which is always an arbitrary 5 em level above the floor, all of the strata are defined as natural layers. Finer natural distinctions within the broader strata were given letter designations (eg., Stratum lOA could be an ash layer within the general fill), while arbitrary levels were given numerical level designations (eg., Stratum 10 Levell is the first 10 em of the fill). Surface: Plow Zone: Alluvium: Bedrock: Sheet Trash: Stratum Stratum Stratum Stratum Stratum 0 1 2 3 9 House Strata Stratum 10: Stratum 11: Stratum 12: Stratum 19: Stratum 20: Stratum 30: Stratum 31: Undifferentiated pithouse fill (St. 40 in sealed houses) Roof/Wall fall (St. 41 in sealed houses) Under Stratum II, above Stratum 20 (used only when Stratum 11 is present) Last 5 em. above floor, but not in direct floor contact (St. 49 in sealed houses) Floor (direct contact only) (St. 21 in sealed houses) Fill of secondary features located within houses Fill of sealed secondary features within houses Extramural Features Stratum 50: Fill of all unsealed primary extramural features including those intrusive into houses Stratum 51-54: Extramural surfaces (successive numbers used if superimposed) Stratum 55: Sealed extramural features (must be completely sealed, such as by a house floor) Disturbed Contexts Stratum 80: Disturbed contexts Other Contexts Stratum 99: Unknown contexts Figure 5.1. Strata designations used on the Rye Creek Project. Project Methods 55 Briefly, feature fill (Stratum lO for structures, Stratum 30 for secondary feature fill, and Stratum 50 for primary feature fill except for structures) is defined as always being entirely within the edges or boundaries of a feature. Sediments overlying a primary feature could be cultural (Sheet-trash Stratum 9) or natural (sterile Stratum 2). In addition, it should be noted that due to the difficulty in accurately determining whether roof fall or wall fall was being excavated, particularly in pithouses, but in masonry pueblos as well, these two designations were lumped together under Stratum 11 as roof/Wall fall. Stratum 11 was identified through the presence of definitive construction material, such as cobbles, daub, or burned beams. Stratum 12, while rare, is occasionally present as a depositional episode between the abandonment of a structure and the collapse of the roof and walls. Stratum 20 is defined as floor-contact artifacts only. Stratum 19 was instituted to separate potential floor artifacts (as Stratum 20 is floor-contact only) from the lower fill or roof/Wall fall layer, because floors can vary in level and be difficult to define at times. Because this is an arbitrary level it was recovered from every excavated structure. By separating Stratum 19 from the fill and floor there is better control over context, and recovered artifacts can be used either as floor artifacts or lower fill artifacts, depending on what the analysis eventually determines. In addition, parts of floor-contact reconstructible vessels are often recovered from this stratum during analysis. Field Methods Based on the results of previous projects within the area, specifically the Ord Mine (Ciolek-Torrello 1987) and Ash Creek (Rice 1985; Hohmann 1985) projects, along with the data from the testing phase (Elson and Swartz 1989a), several major classes of archaeological features were expected or known to be present. These included masonry structures, pithouses, roasting pits, trash pits, trash mounds, agricultural features, cremations, and burials. Excavation methods for the major feature types are briefly presented below by site type (divided into sites with masonry architecture and sites with pithouse architecture). To avoid redundancy within the description of the specific excavation methods discussed below, it should be noted that every feature selected for excavation was investigated and recorded through standard professional methods. These included excavation in either natural layers or arbitrary lO-or 20-cm levels using vertical below datum control, with material screened through quarter-inch mesh (except in the case of burials or cremations where 1!8-inch mesh was used), standard feature recording on Desert Archaeology forms by the crew member, additional note-taking in a separate field journal by the Project Director, Assistant Project Director, Crew Chief and Assistant Crew Chiefs, as well as mapping, profiling, photography, and the collection of various samples for flotation, pollen, and chronological analyses. Excavation units were divided into two types: control units and non-control units. Control units, from which much of the analyzed artifact sample is derived, are standardized (generally 2-m by 2-m or I-m by 2-m) units, excavated in natural layers or arbitrary lO-or 20-cm levels, with all material screened through Y-I-inch mesh. Control units were excavated in all structures, starting at the top of the feature fill, and serve as a comparative data base for artifact analyses. Non-control units were much more variable, and included screened units, unscreened units, and units that contained both screened and unscreened levels. In structures, for example, they generally involved stripping off the upper fill (Stratum lO) and screening the lower fill (Stratum 19) and floor (Stratum 20) contexts. Masonry Sites Within the project area nine sites contained evidence of masonry architecture. Although these sites were almost all small, one-to two-room, field house sites (AZ 0:15:53; AZ 0:15:70; AZ 0:15:71; AZ 0:15:89; AZ 0:15:96 and AZ 0:15:99), three were multicomponent, containing small, one-or two-room surface structures, and subsurface features (AZ 0:15:53; AZ 0:15:55 and AZ 0:15:92) and one was a small village or hamlet, also with a subsurface component (AZ 0:15:54). Unfortunately, several of these sites, including the two largest (AZ 0:15:54 and AZ 0:15:92), were severely disturbed through root plowing. Although some information potential remained at these sites, it was reduced significantly from what it could have been. 56 Chapter 5 Given the importance of sites with masonry architecture, their relatively limited numbers, and shallow depth, all undisturbed masonry structures were completely excavated. Areas of disturbed masonry remains, such as at root plowed sites AZ 0:15:54 and AZ 0:15:92, or site AZ 0:15:70, which was disturbed through road construction, were determined during the testing phase to lack intact subsurface deposits. Therefore, additional excavation within the disturbed areas was not undertaken during the data recovery phase. Each of these sites, however, contained other features that were investigated, such as a small intact masonry feature at AZ 0:15:70, an agricultural field system, and a pithouse and ramada at the Rooted site (AZ 0:15:92), and a trash mound and three pitrooms at the Cobble site (AZ 0:15:54). Masonry Structure Excavation Because a hand-dug trench was excavated through each structure during the testing phase, the depth and characteristics of the fill were already known. Therefore, the interior of each structure was divided into halves, and each halfwas excavated as a single unit using the standard methods described above. Either one, or both, of the halves was excavated as a control unit, meaning that all of the fill was screened in natural or arbitrary levels. The masonry structure fill (Stratum 10), roof fall (Stratum 11), Stratum 19, and floor strata (Stratum 20), were distinguished as natural strata when possible and artifacts from these strata were collected and bagged separately. All floor artifacts were mapped and photographed, and all internal house features, of which there were very few, were excavated and mapped. Extramural Features To determine the use of extramural space, shallow shovel-stripping to the prehistoric surface was conducted around each structure. In general, the stripping extended between 3 m to 5 m around the structure, although this varied by site and was dependent on surrounding terrain and location of other features. The few features found within these extramural areas were entirely excavated and mapped. The only masonry site with a trash mound containing subsurface trash deposits was the Cobble site (AZ 0:15:54). This was sampled through the excavation of severall-m by 2-m units in 10-cm screened levels. With the exception of the Cobble site and pOSSibly the Arby's site (AZ 0:15:99), the other sites with masonry structures were small, single-room, limitedactivity sites, that did not contain recoverable extramural features. Burials Given the concerns of Native American groups and the Advisory Council, particular emphasis was placed on the discovery of cemeteries and isolated inhumations. All human remains were treated with respect and care in accordance with the regulations under ARPA (36 CFR 296); the principles outlined in the Advisory Council's policy statement (September 27, 1988) regarding the treatment of human remains and grave goods; and Desert Archaeology's internal policy on the proper handling of human remains. Failure to comply with this policy was considered grounds for dismissal from the project. This policy is presented below in full. All human bone was collected and all material recovered from burial contexts was screened through 1/8-inch mesh to insure the recovery of smaller grave goods and human bone. Supplemental Backhoe Trenching Limited backhoe trenching was undertaken around the periphery of some of the masonry sites to determine whether additional features, such as buried pithouses or trash deposits were present. It was believed that the Cobble site (AZ 0:15:54), Boone Moore site (AZ 0:15:55), Rooted site (AZ 0:15:92), and Arby's site (AZ 0:15:99), had the potential to contain subsurface features, based upon surface artifact density and depth of subsurface sediments. Although these sites were trenched during the testing phase at 15-m and 20-m intervals, and no subsurface remains were recorded, additional trenching was conducted during the data recovery phase until all trenches were at 5-m intervals to insure that subsurface deposits were not inadvertently missed. Sites that had been severely disturbed through root-plowing, however, were trenched at 10 m intervals, which was considered sufficient to determine that no intact deposits remained. The supplemental trenching was extremely Project Methods 57 DESERT ARCHAEOLOGY'S POLICY ON TREATMENT OF HUMAN REMAINS It is the policy of Desert Archaeology to treat all human remains and associated offerings with dignity, care, and respect. As professional anthropologists and archaeologists we are concerned with the scientific value and information potential human remains can provide in our search for knowledge and understanding of prehistoric cultures. We also recognize the sensitivity and concerns that contemporary Native American groups share toward the treatment, study, and ultimate disposition of human remains. In those instances when we are required to remove human remains the following procedures will be implemented and closely monitored: 1. Excavation and recording of human remains will be conducted carefully only by experienced professional archaeologists or physical anthropologists. 2. Human remains will not be displayed to the general public. 3. No human remains will be left exposed overnight or over a weekend. 4. Human remains will be handled as little as possible. 5. Human remains will be treated with reverence, respect, and care by all employees at all times. productive, buried pithouses or pitrooms were recovered from all of these sites with the exception of the Arby's site. Due to the lack of sediment deposition at the other masonry sites, additional trenching was not undertaken. Backhoe trenching was not used to investigate intact masonry components, although because of the disturbance caused by root-plowing at the Cobble site, the primary area of masonry rubble (Feature 1) was tested with the backhoe. No intact subsurface features were found within this area, confirming the extreme disturbance caused by the root-plowing. Pithouse Sites The testing phase resulted in the discovery of four sites with pithouses (AZ 0:15:52, AZ 0:15:90, AZ 0:15:91, and AZ 0:15:1(0), and four sites with pithouses and masonry architecture (AZ 0:15:53, AZ 0:15:54, AZ 0:15:55, and AZ 0:15:92). Of these sites, only the Deer Creek site (AZ 0:15:52) represents what might pOSSibly be called a small hamlet; the other sites are most likely small farmsteads or fieldhouses. Pithouse sites were investigated during the testing phase through the excavation of backhoe trenches at 20-, 15-, 10-, or 5- m intervals, depending on the site (see Elson and Swartz 1989a). Due to the significance of pithouse sites in Tonto Basin prehistory, and the fact that the majority of the sites were small, containing between 2 and 6 pithouses (with the exception of the Deer Creek site, which contained 17), all recorded pithouses were sampled and the great majority were fully excavated. Supplemental Backhoe Trenching Additional backhoe trenching at 5-m intervals was undertaken at all sites containing subsurface features. Although backhoe trenching is potentially destructive to pithouses, the damage is relatively minimal (a 6O-cm swath through the house) in comparison to the labor and time saved. Previous research by Desert Archaeology has shown that 5-m intervals are the minimum interval needed to insure full data recovery (Doelle 1985; Elson 1986; Elson and Doelle 1986; Huntington 1986). Although the use of 5-m intervals at every site is considered extreme by some archaeologists, we strongly believe that this level of trenching is necessary to document feature distribution in sufficient detail to allow for an in-depth analysis of site structure. 58 Chapter 5 This is due to the fact that cultural features are not randomly distributed across the landscape and therefore not entirely predictable through random or stratified sample trenching. Pithouse Excavation In all pithouses a 2-m by 2-m control unit was excavated in 10-or 2O-cm levels down to the pithouse floor (the depth of the level was dependent on the natural stratification of the pithouse fill). Previous research by Desert Archaeology has shown that a 2-m by 2-m unit is usually sufficient for the collection of a representative artifact assemblage from pithouse fill. Pithouse fill (Stratum 10), roof fall (Stratum 11), and floor strata (Stratum 20) were distinguished as natural strata when possible and artifacts from these strata were collected and bagged separately. Stratum 19, an arbitrary 5-cm level directly above the floor, was also collected and bagged separately. All floor artifacts were mapped and photographed. Up until this stage the excavation strategy was the same for every house. After this, however, the strategy varied depending on whether the pithouse was to be fully excavated or just sampled. In deciding whether to completely excavate a particular house or just to sample a feature, several variables were considered. These variables included disturbance factors, the presence of a floor assemblage, chronological placement through diagnostic ceramics, location and orientation relative to other features, burning, and the density of trash fill. Time and budgetary concerns were also a factor, although generally there was enough of both to completely excavate the majority of the structures. Houses that appeared to be most productive (e.g., contained a floor assemblage or had a high-density trash fill) or significant for other reasons, were completely excavated. In houses that were entirely excavated the pithouse fill was hand-stripped without screening to expose the plan and orientation of the structure after the 2-m by 2-m unit was completed. In all units that were not screened, a grab artifact sample was always taken. Unless the house was severely disturbed through root-plowing or some other factor, Stratum 19 and Stratum 20 were carefully excavated and screened in every house. As mentioned, the great majority of pithouses were completely excavated. In houses that were sampled, excavation continued after the completion of the 2-m by 2-m unit to determine the orientation of the house (by locating the entrance or the hearth and long axis), its approximate size (either through stripping above the house to determine the outline or by trenching to the walls), and the location of the hearth for archaeomagnetic dating. Extramural Features and Burials Extramural features were excavated in a similar manner to the masonry sites described above. With the exception of inhumations, not all extramural features were completely excavated, particularly at the Deer Creek site; most were sampled and mapped. An attempt was made to thoroughly hand-strip extramural plaza areas or courtyard groups, although the amount of the stripping was dependent upon the size of the site. As with the masonry sites, particular attention was placed on the discovery of cemeteries. Cemetery areas were handstripped and all inhumations were excavated carefully, screened through lIS-inch mesh, mapped, and photographed. Only a single site, the Deer Creek site, was found to contain a defined cemetery area, although human remains were recovered from several other sites as well. Agricultural Sites and Features The project area contained a single site (the Rooted site [AZ 0:15:92]) that can be characterized as an agricultural field system, while two other sites contained isolated checkdams or agricultural terraces (AZ 0:15:89 and AZ 0:15:99). The Rooted site, although not large, contained an integrated system of at least 10 checkdams along a single small drainage; more may have been present but were destroyed either through the root-plowing or the construction of State Route 87. The site also contained a very disturbed (root-plowed), small, masonry pueblo of anywhere from one to three rooms, and two pithouses (or a pithouse and a ramada) that escaped the root-plowing because they were within the fence line of the original State Route 87 right-ofway. The density and diversity of the surface artifact assemblage suggests that a gOOd-sized Preclassic period Project Methods 59 pithouse village may have been present here. The site is just south of the Cobble site and may also represent the Classic period agricultural component of this small pueblo. Due to the importance and uniqueness of this site in the regional settlement system, the site was first intensively explored and then recorded, topographically mapped, and photographed. The other two linear terraces within the project area at the Overlook (AZ 0:15:89) and Arby's (AZ 0:15:99) sites were recorded and sampled in a similar manner. 60 Chapta 5 PART 2: SITE DESCRIPTIONS 62 Site Descripfiom .... ....... . ... . . :::~~~;........ . . . ... .... . ... . ......,,',. . . . . . . . . . .. ..... @mmmmm!m1.~ffwmmmwmm:.·'·'·'·!ll!f·'·'·'·'''''':':':':':'·':':.:..................... ~ii @~@@11111!!1..~m .:· ·mnmmm~!m·W· :~~~~ i: .':.It.: [ :.· :~~~~ :::~~: :~~~: :::::~~ :~~~:' :~~~:. =*~!:I:.'~d* :::??-: !!mlm~~Mm~.~* r:.t~,:f.,~:f .,~ ! ! IM1M~1*! !t !!!lWIIMM!!!l ~: :.~, : !.:~, .:~, .:~, i:!.~: i.:;,:.·:i. i,',.! ' . , ...:.!.: • r:::: .• r_r .:.••. .. ..:.: .. :;:::::! :::::::~ ~~~: ;~~~: l.: .~:~.<~.,~ .: .:.:~ :~l,~:.•;:.·. .'~.,§••~••~ .~:' . .l:.d~ ii!!f ~~~ S~ • !o__ w • ••••• !:::,!,,:.:,.I,,:· •• .• .• .•:::f:::. CHAPTER 6 THE HARDT CREEK DRAINAGE Mark D. Elson The Hardt Creek drainage is the third largest drainage in the project area. Like Deer and Rye creeks, it is classified as an intermittent stream, and presently flows seasonally or during periods of increased precipitation. Three of the sites within the Hardt Creek drainage area, AZ 0:15:96, AZ 0:15:71, AZ 0:15:70, are situated on a small tributary of Hardt Creek. The fourth site, the Overlook site (AZ 0:15:89), is actually situated above a tributary of Deer Creek. The Overlook site was included within this section due to both its proximity and similarity to the other three sites. All four sites consist of single-room masonry structures and are believed to have been occupied on a seasonal and resource-specific basis. AZ 0:15:51, a two-room fieldhouse, is also within the Hardt Creek drainage. This site was investigated during the testing phase and is described in Chapter 10. A map showing the location of the sites within the Hardt Creek drainage area is presented in Figure 6.1. ~r:;~~T::~" li;>:.: .\.:~.:~;>,.\: ~;.,:_i~,.":i_r,'.~!:·i.~': ,._!:'~,r_.>:;~iir,i~~._f.:i \:.~•.,; '·:~ ~:.',_;i:.r~.<_:i,.·_~':; ~:.Il.:.~;'·::~.:'~:~',.i <~:~. :,.~:~ ,;.t:~,·". :;,.~i'.:lfi!~ ' Feature " " -- --- o Meters oPS '"" N60 10 I \ \ N 70 0 \ \ N80 Rack Alignment / Terrace N \ f N 90 ~ \ Masonry Structure o o o UJ o ,/" o 'l / o C\I rt) q- UJ UJ UJ Figure 6.9. Site AZ 0:15:89 (ASM). I 84 Chapter 6 Feature 2, the linear rock alignment or terrace, was noted and recorded, although no excavation was undertaken during the testing phase. Due to the presence of natural cobble outcropping and terrace formation on the ridge, the feature was inspected by Gary Huckleberry, the project geomorphologist, who expressed an opinion that the feature was cultural in nature. Data Recovery Phase Work undertaken during the data recovery phase concentrated on the complete excavation of the masonry structure and the testing of the linear rock alignment. Feature I, the masonry structure, was initially explored through widening the testing hand-dug trench and excavating this beneath the wall fall; a layer of darker cultural fill was encountered mixed within the fall. Additional large cobbles were found beneath this layer, which were later determined to be part of the natural cobble bar. The structure was then divided into northern and southern halves; each half was excavated in natural and arbitrary levels corresponding with the fill (Stratum 10), wall fall (Stratum 11), a level 5 em above the presumed floor (Stratum 19), and the presumed floor (Stratum 20). All dirt was screened through Y ~ • 0 Pit 0 0 - Z a: Posthole ,.. -4,'1 III/ U ILl Lithic Rodent disturbance Rock i em N NI09 ....NN ,.., ....N ILl ILl .... ....Non 1&.1 ILl ~ N PROFILE Figure 7.5. Feature 6 at the Deer Creek site. ....NID 1&.1 The Deer Creek Drainage 103 Abandonment and Postabandonment. The lack of a floor assemblage and the presence of burned roof fall lying directly on the floor suggests that the house burned and collapsed relatively soon after it was abandoned. The high artifact density within the fill, along with the contextual analysis, further suggests that the abandoned house pit was used as a repository for secondarily deposited trash. This, in conjunction with the presence of intrusive and overlying features, indicate that the site continued to be occupied after the abandonment of this house. Feature 9 This feature was originally defined in trench profile during the testing phase. Work undertaken during the data recovery phase began with the excavation of a control 2-m by 2-m unit on the east side of the trench. An area measuring 7.0 m by 5.7 m was then hand-stripped to expose the pithouse outline beneath the overlying overburden (Stratum 2) and sheet trash (Stratum 9) layers. The stripping defined the eastern portion of the pithouse but the northwestern comer remained very unclear. A screened I-m by I-m control unit was then excavated in this area to determine if there were intrusive features here. The remainder of the pithouse was excavated without screening to 5 em above the floor (Stratum 19), which was screened. The floor (Stratum 20) was excavated separately. Description. This pithouse is a rectangular structure with rounded comers and a large entryway. The entrance is slightly ramped and rectangular, opening toward the northwest. The hearth is situated on the east side of the trench directly in front of the entrance (Figure 7.6). The maximum dimensions of the house are 5.4 m by 3.7 m (approximately 20.0 square meters). The fill of the house consisted of a gray brown silt with caliche and charcoal inclusions. It contained an extremely high artifact density. The ceramic density was 247 sherds per cubic meter and the lithic density was 235 lithics per cubic meter. Numerous whole manos were found within the fill, suggesting that the pithouse depression had been used as an activity area at some point after the house was abandoned. The fill lying directly on the floor was a reddish sandy silt, much like the sterile substratum. This occurred primarily in the northern half of the house and was interpreted as unburned roof and wall fall, although it may also be washedin sterile sediments. The floor was extremely disturbed by rodent burrows and was therefore very uneven. There was no evidence of preparation of the floor. The floor assemblage included a hammerstone, a scraper, two tabular knives, three manos or mano fragments, a polishing stone, and lithic debitage. The numerous floor features include a hearth, two ash pits, seven pits, and five postholes. All of the floor features were totally excavated. Four of the postholes are within the house and are probably roof supports. They range in size from approximately 10 cm to 22 cm in diameter and from 8 to 23 cm deep. The fifth possible posthole may just be a depression in the center of the entryway; it is small and shallow, measuring approximately 18 em in diameter by 6 em deep. The house walls were cut into the sterile substratum 17 em below the level of the sheet trash. The walls were easy to define on the eastern side of the trench but the small portion of the house on the western side, including the entrance, was difficult to follow. This was primarily because the pithouse intrudes into another pithouse (Feature 25), and a burial pit (Feature 49) intrudes on the southwest comer of the house. Internal Features. Feature 9-1 is an oblong pit situated 60 cm northeast of the hearth. The fill was a yellowish brown, loose silt with light charcoal flecking and a moderate artifact density. It measured 70 cm by 50 em with a depth of 17 em. There was extensive rodent disturbance throughout. Feature 9-2 is a pit 65 em by 50 cm by 15 cm deep, located along the rear wall of the house. The fill contained a light gray silt with a low artifact density and several unmodified rocks. 104 Chapter 7 AZ 0 : 15 :52 (ASM) Feature 9 0 N I4 3 a • 0 M~ 0 0 N I42 0 0 0 0 o . NI 4 1 o J: U Z '"....a: o NI 4 0 Hammersfone -0 O-M N I 39 • • D 0 NI 3B r .... I- T~ ~ 0 ~O 0 N ~ em NI3 7 0 '"'" '"0 '"'" .. 0 '"'" '"0 '"'" CD ~ '" PROF ILE 'Ba.e of Fea. 9- IO Fla. 9- 4 Hearth \ \ \ \ \ Figure 7.6. Feature 9 at the Deer Creek site. ...0 '"'" Posthol e Pi t In tr usive pit Sherd 0 Lit hic M Mono T Tabu lar kn ife a N Pr obab le edQ! Rock CD 0 '"'" The Deer Creek Drainage 105 Feature 9-3 is a basin-shaped pit in the southeast comer of the house. It measured 65 em in diameter and is 11 em deep. The fill was a gray brown, compacted silt with a low artifact density. Feature 9-4 is a large hearth situated directly in front of the entryway. It is approximately 41 em in diameter and 20 em deep. The hearth is constructed within the fill of Feature 9-10, a large pit. It contained fine white ash with two large sherds and a flake. There was a moderate amount of rodent disturbance, which mixed the fill from this feature with the fill from features 9-5 and 9-9, two nearby ash pits. The walls of the hearth were oxidized but only a small area of plaster was encountered. No archaeomagnetic sample was recovered. Feature 9-5 was an irregularly shaped ash pit or possible hearth. It is situated 20 em east of Feature 9-4, the hearth, and is badly rodent disturbed. The pit measures 25 em by 20 em and is only 7 em deep. The walls of the pit were oxidized and the fill contained a dark gray silt mixed with white ash, a sherd, and a lithic. The rodent disturbance has mixed the ashy fill from this feature with the fill of Feature 9-4, the hearth. Feature 9-6 is a large, deep, circular pit situated between the hearth and the entryway. It was bisected by the backhoe trench. The pit probably measured approximately 65 em in diameter and 21 em deep. The fill was a dark gray brown silt with abundant charcoal inclusions. There was no evidence of burning in the pit walls. Another large floor pit (Feature 9-8) intrudes on the southern edge of this feature. Feature 9-7 is a floor pit located along the front wall of the pithouse, just southwest of the entrance. It measured approximately 60 cm in diameter and was 17 cm deep. The fill contained abundant charcoal inclusions within a gray silty loam. There was evidence of root disturbance. Feature 9-8 is a circular pit that intrudes on the south wall of Feature 9-6 and is cut by the backhoe trench. The pit probably measured approximately 60 cm in diameter and was 18 cm deep. The fill contained a gray brown, silty loam with abundant charcoal inclusions and a low artifact density. The pit was heavily disturbed by a tree trunk growing through it and by rodent activity. There was no evidence of in situ burning. Feature 9-9 is an oblong pit 40 cm by 30 cm by 9 cm deep. It is intruded into by the hearth Feature 9-4 and may have been an earlier, informal hearth, or an ash pit. It is also constructed on the fill of Feature 9-10. The fill contained a fine white ash with a moderate density of sherds. There was root and rodent disturbance on the southeastern side. Feature 9-10 is a large pit underlying Features 9-4 and 9-9. It may be an earlier floor feature or may be a pit that predates the construction of the pithouse. The pit measured 1.50 m by 0.70 m but was truncated by the backhoe trench. It was 10 cm deep. The fill was an orange brown silt with charcoal inclusions and very few artifacts. No burning was evident in the pit walls. There was root and rodent disturbance in the western portion. Intrusive Features. Feature 49, an infant inhumation, intruded into the southwestern side of the pithouse, partially cutting the wall. The entrance of Feature 9 intrudes on the east side of pithouse Feature 25. Feature 9 may also overlie Feature 9-10, a large shallow pit. Alternatively, Feature 9-10 could be a floor feature within Feature 9 but earlier than Feature 9-4, the hearth that overlies it. Abandonment and Postabandonment. Feature 9 does not appear to have burned. The floor assemblage is mostly ground stone that was probably intentionally left when the house was abandoned. Soon after the house was abandoned it naturally collapsed, as indicated by what appears to be unburned roof fall lying directly on the floor. That this is unburned roof fall and not washed-in sterile sediments, which it resembles, is suggested by the contextual analysis, which indicates that very little sheet trash washed into the house and that the house pit was used primarily as a dump for secondarily deposited trash. The house was built after Feature 25 was abandoned, as the entryway cut through the wall of Feature 25. Feature 9-10, the large pit underlying the hearth, may have been associated with Feature 25, although this is unclear. Furthermore, there are indications from the ground stone assemblage that the house depression was used as an extramural activity area at some 106 Chapter 7 point after abandonment but during the trash filling. This is suggested by the large number of whole, useable, ground stone found within the fill, which would not normally be discarded as secondary trash. Feature 11 This feature originally was identified as a possible secondary cremation area during the testing phase. This was due to the fact that the feature in profile was not flat-bottomed like a typical pithouse floor, and several pieces of cremated bone were visible in the trench profile. During the data recovery phase, an initial 2-m by 2-m control unit was excavated in the approximate center of the feature as determined through the trench profile. It was excavated to the top of the roof fall layer (Stratum 11) where two intrusive crematoriums were defined. These were excavated separately and the control unit was completed to floor level. A 1-m by 2-m control unit was attempted west of the first unit but was found to be severely disturbed by an intrusive feature and several tree stumps, and so was not completed. An area 6.5 m by 4.75 m was then hand-stripped to define the pithouse outline. No clear outline was ever identified so the walls were defined through the excavation of the house fill (Stratum 10). The pithouse fill surrounding the control units and intrusive features was excavated without screening to 5 em above the floor (Stratum 19), which was screened. The floor (Stratum 20) was excavated separately. Description. Feature 11 is a small, irregularly shaped pithouse. The floor area measures 10.1 square meters (Figure 7.7). It appears to be oriented toward the east or the northeast but due to numerous intrusive features and abundant root disturbance the exact outline of the feature is somewhat unclear. The house walls were cut into the sterile substratum approximately 30 cm below the sheet-trash layer. The fill of the house consisted of a dark gray sandy loam with a low-to-moderate artifact density. The ceramic density was 84 sherds per cubic meter and the lithic density was 66 lithics per cubic meter. A layer of compacted orange-to-red roof fall was found overlying a thin layer of ash that was lying directly on the floor. Neither the floor nor the walls were prepared or plastered; the floor was identified by the difference between the dark cultural fill and the sterile substratum. The floor assemblage consisted of two manos, a core, and scattered individual sherds and lithics. The contextual analysis (see Chapters 11 and 14, Volume 2) suggests that the lithic floor assemblage shows evidence for tool manufacturing or possibly tool maintenance. The only possible floor feature (no feature number) was an area of oxidation on the floor that may represent the hearth. It was badly rodent disturbed and very irregular in shape. It could also have been a small ash pit or just a highly burned area on the floor. An archaeomagnetic sample was collected from this area. From the presumed location of the entryway, a hearth could have been removed by either the intrusive crematorium, Feature 52, or the backhoe trench. The posthole pattern, consisting of six postholes situated outside of the wall, suggests that this is one of the few pithouses at the site with exterior posts. All of the exterior postholes are fairly large and well defined. They range in size from 11 em to 20 em in diameter and 5 em to 15 em deep. None contained any post remnants or a rock lining. Internal Features. No internal features were identified. Intrusive Features. Features 48 and 52, both crematoriums, Feature 51, a secondary cremation, and Feature 63, a trash pit, all intrude into this house. Feature 48 is fully within the fill of the house, Features 51 and 52 cut through the floor, and Feature 63 cuts through the western wall. The pithouse does not intrude on any features. Abandonment and Postabandonment. The thin layer of ash lying directly on the floor with the roof fall directly above it, along with the light floor assemblage, suggests that the house was deliberately abandoned, and then burned and collapsed soon after. The moderate density of artifacts within the fill, along with the contextual analysis, suggests that the house pit was primarily used as a secondary trash area, although deposits of transformed or mixed secondary trash and sheet trash are also present. This area was converted into a cemetery at some later point in time, but still during the Gila Butte phase. The Deer Creek Drainage 107 AZ 0: 15: 52 (ASM) Feature II NI 16 t Nil:! ~"'T"'-(''''''1- Fea. 52 >(0{ -{ • NI14 A.( -< -(' '"T" )0- :l: V ...z: a a R>. .\ /.. I- I- Fea. 63 lIIt' '« / '( '( y '< -<'..J.,.y NI12 ).y t- Il) C\I • Postho le LT . ~ ...~ I ntrusive feature a a Co Sherd L i lhic D Rock :~ •• PI oster ~- N ... 00 if i ...a: • ).. .{ NII3 ,., 0 00 T,..,. -{ -'!' • '1-.,.. . 'l- Crematorium ~-< 0 Co A f. -<. -<. >. :: ::. Ol i di zed area • ",Xxx Ash 1/1/ Di sturbed area ~o 0 I I em :: o a> ... N C\I W PROFILE Figure 7.7. Feature 11 at the Deer Creek site. a> ... C\I C\I a> ... C\I 108 Chapter 7 Feature 12 Description. This is a small, relatively circular, pithouse with a ramped entryway oriented toward the southeast. The house does not contain a hearth. Instead, a likely trivet within an associated ash pit are situated directly in front of the entrance (Figure 7.8). The floor area measures approximately 9.3 square meters, making this one of the smallest pithouses at the site. The floor was not plastered or prepared in any manner, although there were areas of burning, and it was distinguished by the contact between the dark cultural fill and the lighter sterile substratum. The floor sloped from the edges of the house into the trivet pit in the center. The floor assemblage consisted of two manos, one of which was in the entryway, a mano fragment, a piece of lithic debitage, and a few isolated sherds and lithics. There was only one floor feature, the ash pit with trivet (Feature 12-1), and three very shallow, possible postholes. These were all situated on the south side of the feature and may have been roof supports. Their shallow nature suggests that the house may have been constructed with a bent-pole architectural style and that additional postholes were too ephemeral to locate. An intrusive secondary cremation pit (Feature 53) cuts through the floor in the north-central portion of the house. The fill was not homogeneous across the pithouse. The area with the darkest burned fill, was in the center of the house near the possible trivet. The walls and entryway of the house were difficult to define because the pithouse fill in these areas was a light brown color, similar to the sterile substratum. Also, a crematorium, Feature 46, intruded on the east wall of the house and the north side of the entrance. The fill did contain a small amount of charcoal flecking, however, which was not found within the walls, allowing for a subtle distinction. In addition, there was a small area of wall plaster, 10 em long and 5 em wide, along the east side of the house. The walls are approximately 12 cm high below the layer of sheet trash. The fill contained a very low artifact density. Ceramic density averaged 37 sherds per cubic meter and lithics averaged 43 lithics per cubic meter. Internal Features. Feature 12-1 is a circular pit, 62 cm in diameter and 18 cm deep. The fill contained light brown charcoal-flecked sediments mixed with a dark gray ash containing pieces of charcoal. There were several large rocks in the fill; a large cobble was embedded in the sterile alluvium at the pit bottom, with two other smaller cobbles resting upon it. There were several other smaller rocks in the pit also. The cobbles were positioned in a roughly triangular manner, suggestive of a trivet or pot-rest, although this is not entirely clear. The location of the pit, directly in front of the entrance and slightly offset from the middle of the house also suggests that this represents a trivet or some sort of similar cooking feature. There was no oxidation visible in the pit walls to suggest in situ burning, however. Intrusive Features. Feature 46, a crematorium, intrudes on the east wall of the house and the north wall of the entrance. It extends into the house 30 em. Feature 53, a secondary cremation, cuts through the floor of the house. It probably originated within the fill but was not recognized until floor level. Abandonment and Postabandonment. The relatively small amount of burning evident in the fill and on the floor of the house indicates that the house did not completely bum. The small floor assemblage further suggests that the house was abandoned prior to the collapse of the roof and walls. Although the abandoned house was not used exclusively as a trash area for secondary refuse, the contextual analysis indicates that some trash dumping was occurring, particularly lithics. The general area was used as a cemetery after the abandonment of the house, as indicated by the two intrusive mortuary features. Feature 13 This pithouse was identified during the testing phase in the profile of Trench 2. Excavations during the data recovery phase began by obtaining a control sample from two 1-m by 2-m units on the west side of the backhoe trench. It was necessary to excavate two units in order to obtain a 4-square-meter sample and avoid The Deer Creek Drainage 109 AZ 0: 15 : 52 (ASM) Feature 12 NI23 e 0 9 Fea. 53 ...... '(~ " "I c::;, -l,)- NI22 o 00 .., e NI2 1 ~'~' J: u z ~ It: l- ~ • e • () "'4 ' e 0 NI 20 Gi 0 50 I I em i N NI1 9 en ~ 0 '" I'J ~ U Z It: I- • III I'J J: Fea . 46 Crematori um iii I'J ~ I'J '" I'J ~ • E;) .,..,..,. Burned daub Posthole Pit Intrusive feature e Shard ® 0 Mono Rock .., '" v '" I'J I'J ~ ~ PROFILE Figure 7.8. Feature 12 at the Deer Creek site. 110 Chapter 7 major root disturbance while still falling entirely within the pithouse. Description. Feature 13 is a small rectangular pithouse with rounded comers. The entrance opens to the southeast and the hearth is situated directly behind it (Figure 7.9). The house contains a floor area measuring approximately 8.5 square meters, the smallest pithouse at the site. The walls were excavated to a depth approximately 10 em below the sheet trash, making this also one of the shallowest houses at the site. The fill was a reddish brown, silty loam mixed with darker black sediments. It contained charcoal and burnt daub inclusions as well as a low artifact density. The ceramic density was 51 sherds per cubic meter. The lithic density was 69 lithics per cubic meter. The floor was distinguished by the difference between the cultural fill and the underlying sterile alluvium substratum. It was not plastered or prepared in any manner although several burned areas were present. The only floor artifacts were a single mano near the southern wall and scattered individual sherds and lithics. A burned beam approximately 65 em long also was lying on the floor. Floor features consisted of a hearth, Feature 13-1, and two small postholes. One of the postholes was in the northeast comer of the house and measured 12 em by 9 em by 7 em deep. The other posthole was situated just south of the hearth and was very small, measuring approximately 5 em in diameter and 5 em deep. There are several rocks of no apparent function scattered across the floor. Some of the burned beams were on top of the rocks indicating they were in place when the house burned. A cluster of rocks situated against the wall along the south side of the house may have been used to bolster or reinforce the wall. Internal Features. Feature 13-1 is an unplastered hearth situated directly in front of the entryway. It appears to have been remodeled several times and moved southeast towards the entrance with each remodeling. The fill of the last used hearth, which was exposed at floor level, consisted of a fine light gray ash that extended under the house floor to the northwest. Upon excavation, the area sealed by the floor was found to be an earlier use of the same hearth. Separate hearths could not be identified, however, since they tended to conjoin, and the actual number of hearths remains unknown. The bottom and sides of the pit were oxidized. The entire feature, including all of the hearth area, measured 90 em by 75 em and was 10 em deep. An archaeomagnetic sample was recovered. Intrusive Features. There were no intrusive features and the pithouse does not intrude into any features. Abandonment and Postabandonment. The lack of a floor assemblage and the presence of the burned beams directly on the floor indicates that the pithouse was abandoned and probably burned soon after. The low density of the artifacts in the fill above the roof fall is within the range of sheet trash (or transformed secondary trash) deposits as determined through the contextual analysis. Some lithic reduction may have been occurring within the fill, however. The low artifact density and lack of intrusive features suggests that this feature was occupied relatively late in the site occupation, which is tentatively supported by the archaeomagnetic data. Feature 14 Description. This is an almost-square pithouse with a ramped entrance oriented toward the southeast (Figures 7.10 and 7.11). It measures 4.2 m by 4.0 m (16.8 square meters). The house is one of the deeper pithouses at the site with walls excavated into the sterile substratum 38 cm beneath the sheet-trash layer. The fill contained a gray sandy silt with numerous charcoal inclusions, areas of oxidized orange roof fall, and pieces of burned daub. The layer of roof fall was extremely dark and very thick, averaging approximately 17 em in thickness throughout the house. The upper layers of the fill contained a higher artifact density than the lower layers, although overall the artifact density was moderately high with an average of 138 sherds per cubic meter and 139 lithics per cubic meter. The Deer Creek Drainage 111 AZ 0: 15: 52 (ASM) Feature 13 PROFILE A- Figure 7.9. Feature 13 at the Deer Creek site. 112 Chapter 7 AZ 0 : 15: 52 (ASM) Feature 14 NI2' NI24 Mm @CJ G" '.~. ..~ NI23 !t O O-M fragment M Metate 3 o1) M • 0 M GS Pit Reconstructible • ., .. 1 Sherd Mono Ground stonl Tr Trl •• t rock RV 0 0 :z: u z .... ... \oJ a: .... .'• \]M <::J I\7M \ OM --- Probable edQI Posthole NI20 ... .... .:' ~Q-MMetat. 2 NI21 .... ,..... , 0 50 em i /" I , ~. N Rock .; :: Plott.r _ Pla."r .. all patch .. ...... ...on \oJ \oJ '" '" a II> ...... '" \oJ Figure 7.10. Feature 14 at the Deer Creek site. ...'" '" \oJ The Deer Creek Drainage 113 Figure 7.11. Photograph of Feature 14 at the Deer Creek site. The floor was heavily oxidized but had only small areas of plaster concentrated around the hearths. There were several floor features. Two hearths, Features 14-1 and 14-3, were situated approximately in front of the entryway. The third hearth (Feature 14-4) is located along the north side of the house. There were also two pits, one of them (Feature 14-5) possibly containing a trivet. A series of three postholes were found inside the southern wall and two other postholes were in the interior of the house. One of the two interior postholes was probably for one of the main roof supports. It measured 13 em in diameter by 16 em deep and contained burned post remnants. The other four postholes are smaller, ranging in size from 5 em to 9 em in diameter and from 7 em to 12 em deep. The three postholes along the southern wall were probably for part of the wall support while the other interior posthole may have been for an additional roof support. The house contained an extensive floor assemblage, including 3 whole trough metates, 16 manos, a single reconstructible plainware jar, and scattered sherds and lithics (Figure 7.10). The lithic assemblage shows evidence for primary core reduction. One of the trough metates (Number 2) was found in an upright position near the southwest wall and had five manos in close association. Metate Number 3 was found inverted in the northeast comer with six manos on the floor near it. The third metate (Number 1) was also upright, near the west comer of the house, and had at least two manos in direct association. Due to the location of the backhoe 114 Chapter 7 trench, additional manos associated with metate Number 3 could have been removed. The reconstructible jar was found in the north-central portion of the house. The walls of the house were mainly identified by the contrast between the dark cultural fill and the reddish sterile substratum. Both the northeast and southwestern walls have small areas of burned plaster which suggest that the pit walls were plastered and incorporated into the house. No exterior postholes were found, although they were intensively looked for. As a result, it is difficult to determine if this is a true pithouse that incorporates the pit walls, or, like almost all of the pithouses at the site, a house-in-pit. The entrance was not clearly defined because the fill within it was not as dark as the fill of the rest of the house and therefore did not obviously contrast with the sterile substratum. It is likely that the entrance did not bum. The location of the hearths, the lack of a well-defined wall in this area, and subtle differences in the sediments, strongly suggest that the house was oriented to the southeast. Internal Features. Feature 14-1 is a hearth set back approximately 1.0 m from the front wall of the house, slightly south of centering on the entryway. It contained a shallow, basin-shaped, caliche-plastered portion, 20 cm in diameter and 5 em deep, along its eastern side. The fill extended from this area to the northwest an additional 30 em in an irregularly shaped, un plastered but oxidized, depression, 7 em lower than the bottom of the plastered portion. It appears that the hearth was remodeled and moved, although the direction and sequence of use is unclear. The fill throughout the feature contained dark ash, three sherds, and a lithic. Feature 14-2 is a posthole in the northern portion of the house with a thin band of plaster floor surrounding it. It measured 13 cm in diameter and was 16 em deep. A portion of the post was still standing above the level of the floor and was collected. Feature 14-3 is a fully plastered, basin-shaped hearth completely sealed by the floor. It is situated approximately 10 cm north of hearth Feature 14-1, and is more directly centered on the entrance. It measured 23 em in diameter with a depth of 8 cm and contained a gray ashy fill. A 10 cm wide caliche-plastered collar was found around the feature. The use of this feature appears to predate the use of Feature 14-1. An archaeomagnetic sample was recovered. Feature 14-4 is the third hearth in this house. It is located in the northwest comer of the house, almost against the wall, which is an odd location for an interior hearth. The feature itself is also morphologically different from most hearths at the site, being large and deep with straight sides leading to a basin-shaped bottom. The hearth measured approximately 33 cm in diameter and was 21 cm deep. The caliche-plaster coating was extremely burned and found mostly along the south and east sides of the collar; it extended down the east side of the hearth interior approximately 10 em. The remainder of the interior was highly oxidized and without plaster. The fill consisted of a dark black ashy matrix. The feature was not sealed so was probably in use at the time the house was abandoned. Feature 14-5 is a large, irregularly shaped pit, with several large rocks resting on the pit bottom. The pit measured 1.10 m by 0.40 m by 18 em deep and was situated just west of the center of the house. The rocks may represent a trivet, although this is uncertain. The fill of the pit consisted of a dark gray, ashy silt with a low artifact density. The pit walls were not oxidized. The eastern side of the pit was heavily rodent disturbed. Feature 14-6 is a small pit or possible hearth. The walls of the pit were oxidized but not plastered. The fill consisted of a loose, ashy gray matrix with pebbles and a low artifact density. The pit measured 2S em in diameter and 7 cm deep. It was situated approximately 30 em southwest of Feature 14-5, the large pit with the possible trivet. Intrusive Features. There are no intrusive features, nor does this house intrude into any other feature. The Deer Creek Drainage 115 Abandonment and Postabandonment. The extensive floor assemblage along with the presence of burned roof and wall fall lying directly on the floor, suggest that this house was still in use when it burned. The fact that one of the metates was inverted, however, may indicate that the occupants were away from the site when the burning occurred. In addition, the house may have burned due to ritualistic or conflict-related events, rather than a catastrophic event. A ritual- or conflict-related burning is suggested by the extremely heavy and complete nature of the bum, which is difficult to achieve through natural means alone. After the house collapsed, the pithouse depression was used as a trash dump, as suggested by the contextual analysis and the high artifact density in the fill. Feature 18 This pithouse was recorded in the additional trenching undertaken during the data recovery phase. A 2-m by 2-m control unit was excavated on the east side of the trench, but it was difficult to discern from the fill whether the unit was entirely within the pithouse. Therefore, a second 2-m by 2-m control unit was excavated directly across the trench from the first, to obtain an undisturbed sample of the house fill. An additional1-m by 2-m unit was excavated to the west of the second 2-m by 2-m unit to increase the sample size. Additionally, a lower floor was visible in the trench profile, and was exposed once the upper floor and house were completely recorded. The house and all floor features were totally excavated. Description. This is the largest house at the site. It is an oval-shaped pithouse with a step entrance opening toward the east (Figure 7.12). The house measured 6.3 m by 5.3 m (33.4 square meters). The entrance was somewhat difficult to define due to heavy rodent disturbance, but appears to have a step at the east end of the ramp. The fill consisted of dark gray ashy sediments with numerous charcoal and burned daub inclusions. The evidence of burning increased in the fill closer to the floor. The fill contained a low-to-moderate artifact density. The ceramic density was approximately 61 sherds per cubic meter and the lithic density was approximately 41lithics per cubic meter. The floor was uneven with very little evidence of preparation. There is a small area of caliche plaster in the center of the house and around the hearth collar. The north side of the house is cut into a very gravelly substratum, evident in the floor and walls. The floor and walls in the remainder of the house are defined by the contrast of the dark feature fill with the light orange-to-brown sterile alluvium. A small portion of the floor sloped up in the area around the hearth. This is the same area where a second, lower floor, was visible in profile. Upon excavation, it was apparent that both floors were related to the same hearth (Feature 18-1), which had been remodeled from a lower hearth, and the upper floor represented a replastering of a low portion of the lower floor. It seems that the earlier floor sloped down around the hearth and the later floor was built to raise it up. The walls of the pithouse were excavated into the sterile substratum approximately 25 em below the sheet trash on the north side of the structure, but only 14 cm on the south side. This is probably due to the fact that the sheet-trash layer is thicker on the south side of the feature and has obscured more of the top of the wall, because it is unlikely that the prehistoric ground surface was so drastically unlevel. The upper floor assemblage consisted of only four pieces of ground stone, two cores, lithic debitage, and isolated sherds. The lower floor contained a large sherd cluster consisting of 39 body sherds from the same vessel but no rim sherds. There were only two floor features; a hearth associated with both floor levels and a small pit originating in the lower floor. Numerous postholes were present. These consisted of three large central postholes (Features 18-3, 18-4, and 18-5), fifteen postholes directly within the walls, one or two external postholes, two small postholes in the interior of the house, and nine shallow depressions scattered across the floor. These may be shallow post 116 Chapter 7 AZ 0 : 15: 52 (ASM) Feature 18 HI 3S NI34 (J o NI 33 • o ,,- N I 32 ". (Foe .... .:.... . . :.. :: HI " '. ' • H I 30 . I c\ ..... " .: ~ .'. ° 0 ' a c • '.' " ••••• STRATUM 2. . '. • ;0 .. \ H.a,", ····f.18-' Pla,t., colla, : •• • ./ Pla.ter a /-'_.-._.- : o~ \r:-h STRATUM 2 • \\. \./.° 0 1Q.os • 7 a \$ .,.*'" _.--;".. '. a " .,,' I o ~ ...'"a:z !' u a .. H12' • _. . . . -0-.------o 128 .... - - - - P, obable edge Po,thole • ... SO c Cl L-.-J cm .::. .. o ... N iii N ... .... N N .. ... N ...... N i @PIt a Sh.,d 0 Lithic GS Ground .tone .. ., ... N N Cor. Roc_ Oepr ...lon .. ... N .. ... N -It Figure 7.12. Feature 18 at the Deer Creek site. The Deer Creek Drainage 117 supports but could also be potrests or served some other function. The postholes along the walls range in size from 12 em to 23 em in diameter and 7 em to 35 em deep. The three large central postholes are 2.5 m apart and extremely large, measuring approximately 40 em in diameter and from 45 em to 65 em deep. None of the central postholes contained post remnants although several of the smaller postholes around the walls did. The two external postholes are both on the north side of the house, 50 em apart. The western one, Feature 45, may be a small pit rather than a posthole since it contained an argillite censer that had been intentionally buried within it. Internal Features. Feature 18-1 is an oblong, basin-shaped hearth situated in front of the entrance. It measured 20 em by 15 em and was 15 cm deep. The hearth is partially plastered; a collar extends out onto the house floor approximately 5 em to 8 em and the plaster extends to em down the sides of the basin. The bottom of the hearth is oxidized and an archaeomagnetic sample was collected. The fill consisted of a dark gray, sandy silt with burnt orange daub and charcoal inclusions similar to the fill of the pithouse. There is an earlier, nonplastered, hearth directly beneath this that was articulated with the lower floor. Features 18-2 to 18-5 are postholes from which artifacts or samples were collected, necessitating the assignment of feature numbers. Feature 18-6 is a small pit that originates in the lower floor approximately 20 em southwest of the hearth. The pit measured 30 em in diameter and was 18 cm deep. The fill contained a compact brown silt with charcoal flecking and a moderate number of artifacts. The pit walls and bottom were defined by the contact with the sterile red clay substratum. Feature 45 is either a small extramural pit or an external posthole. It is on the north side of the house and 50 em from another presumed external posthole. The reason this feature was assigned a primary feature number is because of the uncertainty of its relationship to the pithouse. Within the fill of this feature was a carved phallic censer made out of argillite (see Chapter 16, Volume 2). This may have been a fertility offering in either a small pit or a posthole, although this is purely speculation. The pit measured approximately 30 cm in diameter and was 19 cm deep. The fill consisted of a loose, gray brown sediment with charcoal inclusions. Other artifacts found within the fill included a few sherds and lithics, and a polished stone. Intrusive Features. No features intrude on this house nor does this house intrude on any other feature. Abandonment and Postabandonment. The evidence of burning and the lack of a floor assemblage suggests that the house was abandoned prior to the burning. The lack of burned post remains in the large central postholes, in contrast to the numerous burned remains of the smaller posts, further suggests that the burning was intentional and that artifacts and large posts were removed before the structure burned. This is additionally supported by the very intensive and complete nature of the bum, which is difficult to achieve under purely natural conditions. The low artifact density of the fill, along with the contextual analysis, suggests that the feature was used for only limited secondary trash deposition after abandonment. Feature 21 Feature 21 was recorded during the data recovery phase. A 2-m by 2-m control unit was first excavated using the standard methods described above. Because of the high artifact density, a I-m by 2-m control unit was then excavated east of the first unit. This was excavated in to-cm levels to see if a tighter vertical control would allow for the seriation of the ceramic assemblage. Description. This pithouse is a large rectangular house with rounded comers. The entrance opens to the west, possibly toward Feature 18 with which it may be contemporaneous. The entrance is rectangular and slightly 118 Chapter 7 ramped. The house measured approximately 5.7 m by 4.4 m (25.1 square meters) (Figure 7.13). It was excavated into the sterile red alluvium to a depth of approximately 20 em beneath the overlying sheet-trash layer. The fill contained a moderate-to-high artifact density within a fine dark matrix. There were approximately 134 sherds and 165 lithics per cubic meter. The 10 em directly above the floor consisted of a burned roof-fall layer containing burned daub and charcoal inclusions. The floor was defined by large areas of floor plaster. Plaster was found within the south side of the house and in the central area around the hearth and postholes. The northern portion of the floor contained compacted sterile sediments with oxidized areas but no formal preparation. The area around the hearth exhibited evidence for more than one floor, probably related to one or two minor remodeling episodes corresponding to a similar remodeling of the hearth. The house contained a relatively rich floor assemblage, one of the few found at the site. It included five reconstructible plainware vessels (one bowl and four jars), six manos, three tabular knives, a metate fragment, a chopper, a hammerstone, and a projectile point. Three of the vessels were near the rear of the house while the other two were in the southwestern comer. The stone artifacts were not clustered in anyone area. Three related floor features were present, all situated directly in front of the entryway. One was the hearth, Feature 21-3, and the other two were ash pits, Features 21-4 and 21-5. The hearth showed evidence of remodeling that would correspond with the three floor layers. The ash pits were associated with two different floor levels. Eleven postholes were found within the house. Six of these were situated around the walls and probably functioned as wall supports, while three internal postholes may have been for the central roof-support posts. The two other postholes are also in the interior and probably served as additional roof supports. All of the postholes are similar in size and range from 9 em to 15 cm in diameter and from 10 em to 18 em deep. Internal Features. Feature 21-3 is a hearth that shows obvious signs of remodeling. The feature was cut by the backhoe trench and observed in the trench profile but was not visible when the floor and associated floor assemblage were exposed during the excavation of the pithouse. As a result, it was apparent that the hearth in profile had been sealed by the upper floor of the pithouse. No hearth associated with the upper floor was found, suggesting that the hearth in use with the upper floor had been removed completely by the backhoe trench. This is unknown, however, and it is possible, although considered unlikely given the floor assemblage and formality of the architectural style, that the upper floor did not contain a hearth. Portions of two hearths remain, one within the other, and each articulating with different floor levels. They both contained evidence of being plastered and probably were basin-shaped, although so little remains it is impossible to be certain. The fill consisted of a dark matrix with a low artifact density. An archaeomagnetic sample was recovered. Feature 21-4 is an ash pit that was visible in the upper floor just east of the hearth area. It was circular in shape, approximately 20 em in diameter, and contained a fine gray ash with no artifacts. The pit appeared to be slightly bell-shaped but was difficult to define because it intruded on an earlier ash pit, Feature 21-5. Feature 21-5 is an ash pit associated with the second or lower floor of the house. It was situated within 20 em of the hearth. The ash pit was oblong and measured 75 em by 55 em with a depth of 14 em. It is intruded upon by Feature 21-5. The pit fill contained a fine gray ash with a few artifacts. Intrusive Features. Feature 47, a pile of roasting pit debris that could represent a single dumping episode, was within the fill of this feature. The pithouse does not overlie any features. Abandonment and Postabandonment. The floor assemblage directly overlain by a burned roof-fall layer suggests that the house catastrophically or ritually burned. A ritual bum is suggested by the complete nature of the burning, although this is conjectural and there is no way to be certain. The contextual analysis of both the lithic and ceramic assemblages suggests that the house was not intensively used as a secondary trash dump after The Deer Creek Drainage 119 AZ 0:15 : 52 (ASM) Feature 21 NI33 • N I 32 o o NI3 1 I I \ GS-G \)-GS N I 30 ~ R'(~.... \ _ ) 0 \l.:Hamm."tone Hearth " ProjecHie Fe • . 2 1- 3 ",~point .~V R.t ~ A- 0 N I29 • ! N I28 0' o T·"T·'" !Rv\ ~- ' o NI27 o N I 26 f0Ol N '" ~ '"uz ...'"a: - - - Probable odQo - . - .- Edoe romodeled .,eo • Postho le DO 0 Feo . 2 1-1 0 Pit c9 P it in lower floor o Reconstructible vessel Sho,d RV4 Lithi c 124 Chapter 7 AZ 0: 15:52 (ASM) Features 32 ond 36 I I ~ NI30 ,----... Fla. 37 "' ...... ," "'I ,I NI29 o " NI28 Spindle whorl ~ OMana NI27 o Hearth 0 " o NI26 \ ~O " \ \ N12~ , a - P l a s t l r callar I I \ \ 0 " NI24 •.... --- :" ® I -----I I NI23 0 50 '-------I cm .. III ... N .. ... ... N ... ... 0 N i=: ... N Posthole Posllbll posthole Un,,,cD'Ioted floor feature 0 Sherd " Lithic c::::> '.q. Probable edal Rock ... ... N N PROFILE Figure 7.16. Features 32 and 36 at the Deer Creek site. 1 N ..,... ... N .. ... ... N The Deer Creek Drainage 125 The walls were straight and vertical with well-defined, squared-off comers. They are at a height of 19 em below the sheet trash. Portions of the walls were heavily burned, giving them an almost plaster-like appearance. Internal Features. Feature 32-1 is a well-plastered, basin-shaped hearth, situated directly in front of the entrance. It is circular, measuring 25 em in diameter and 8 em deep. The plaster lines the entire basin and extends over a low collar approximately 8 em around. The fill was similar to the pithouse fill consisting of a gray brown, sandy silt; it contained only a few charcoal flecks and no artifacts. An archaeomagnetic sample was collected. Intrusive Features. Feature 36, a pithouse, either overlies or is intruded into by Feature 32. Although it was difficult to be certain in the field, the fact that a floor and hearth were never located for Feature 36 suggests that Feature 32 intruded into it, removing the hearth and most of the floor. Their western walls are both clear, fairly parallel, and 50 em apart. Figure 7.17. Photograph of Feature 32 at the Deer Creek site. 126 Chapter 7 Abandonment and Postabandonment. The house contained no floor assemblage. This, along with the presence of a dense, burned, roof-fall layer lying directly on the floor suggests that the house was abandoned shortly before it burned. It is interesting to note that there were no burned beams or posts recovered, suggesting the possibility that they were intentionally removed prior to the burning. The moderate artifact density within the fill and the contextual analysis further suggests that after the house burned, the house depression was used as an occasional trash dump, although much of the material was washed in. Feature 34 This pithouse was first recorded during the data recovery phase. The feature is situated between the present State Route 87 and the right-of-way fence, and appears to have been compacted during road construction. It is the westernmost house at the site, separated from the other pithouses by approximately 20 m. Two I-m by 2-m control units, one on each side of the backhoe trench, were excavated. Stripping to define the pithouse outline was done initially by hand. Once the house was defined, the overburden within an area of 3-m to 5-m surrounding the walls was removed mechanically by the backhoe. Description. This pithouse is roughly oval in shape with an unknown orientation (Figure 7.18). No entrance was identified; however, from the location of the two hearths (Features 34-1 and 34-2), it may have been oriented toward the south or southeast. The floor area measured 4.0 m by 3.2 m (12.8 square meters). This is one of the smallest pithouses at the site. The fill consisted of a compact gray brown silt with charcoal flecking and very small pieces of burned daub. It contained a moderate artifact density. There were approximately 79 sherds and 99 lithics per cubic meter. The walls were identified by the contrast between the cultural fill and the redder sterile substratum. They were approximately 10 cm high beneath the sheet-trash and road-disturbance layer. The floor was defined by small areas of plaster around the hearths, areas of burning, and thirteen pieces of ground stone. Between these areas, the floor was identified by the contrast between the cultural and sterile strata. Extensive root and rodent disturbance obscured the floor and wall in the eastern portion of the house. The floor assemblage consisted primarily of ground stone. There were nine manos scattered across the floor, as well as three other pieces of indeterminate ground stone, a hammerstone, lithic debitage, and isolated sherds. Three of the manos were clustered around a posthole in the southern portion of the house. They probably were either resting against the post when the house burned or were used to support the post. Similar to several other houses at the site, there were also numerous unworked cobbles on the floor, the function of which is unknown. Three floor features were identified -- two hearths and a very small pit. The hearths are situated in the center of the southern half of the house, 60 cm apart. The small pit, Feature 34-4, is along the wall in the southwest comer. In addition, there are only two postholes, both on the south side of the house. The one with the manos measured 30 cm in diameter and was 17 em deep. The other posthole was 5 em in diameter and 8 em deep. Internal Features. Feature 34-1 is a portion of a hearth that was cut by the backhoe trench. The remaining portion measured 50 cm by 15 cm by 9 em deep. It was probably at least 50 em in diameter before the backhoe trench was cut. There was no evidence for a caliche-plaster coating but the walls were highly oxidized. The fill was an ashy gray silt with no artifacts. No archaeomagnetic samples were recovered. Feature 34-2 is a small hearth 60 cm south of Feature 34-1. It measured 30 em in diameter and was 5 cm deep. The southern half of the feature was badly disturbed by rodents. The hearth consisted of an unplastered oxidized basin containing the same ashy gray silt as Feature 34-1. No archaeomagnetic samples were recovered. The Deer Creek Drainage 127 AZ 0: 15: 52 (ASM) Feature 34 NI42 ". '- NI41 0 ~GS c:S 0 0 NI40 ..:.'' N N J: u Heorlh Feo . 34-1 Z \AI 0 II: I- 0 0 NI39 0 0 0 9· 00 0 0 0 ® NI38 i N • I o ~o 0 em I , ".. NI37 Posthole Pit o Sherd o Lithic ® Mono @ Groundstone c:::J Rock , : : Plaster ,... on N \AI o en on CD N \AI \AI N PROFILE A- Figure 7.1S. Feature 34 at the Deer Creek site. CD N \AI N CD N IIJ 128 Chapter 7 Feature 34-4 is a small pit in the southwest comer of the house. It measured 10 em in diameter by 5 em deep and was covered by a flat-lying rock. The pit contained a sherd and a small mano within a charcoal-flecked fill. Alternatively, given its small size, this may represent a posthole. Intrusive Features. There are no features intruding on this pithouse. The house does not intrude into any other feature. Abandonment and Postabandonment. The artifacts found on the floor, almost all ground stone, may have been left intentionally when the house was abandoned. The house burned, at least partially, as indicated by areas of burned floor, and charcoal and burned daub in the fill. After the house collapsed, it appears from the moderate artifact density and the contextual analysis that the depression filled in naturally with sheet-trash deposits. Some lithic primary refuse is also present within the fill, suggesting possible use of the sheltered depression as an extramural activity area. Feature 36 This feature was first identified as a pithouse in the wall of Trench 10 that was believed to be overlying pithouse Feature 32. The first excavation unit was placed beyond the profile edge of Feature 32 in an attempt to define the house without any interference. Only the bottom 5 em of this unit was screened. Three I-m by 2-m control units were then excavated, two on the east side of the trench and one on the west side. All three units were within the area overlying the fill of Feature 32, and a floor could not be identified. Finally, an area 7 m by 5 m was hand-stripped to define the walls. The 5-cm level above the floor was screened. Because only the western portion of the walls could be defined, and the floor was difficult to follow and absent in most areas, no further work was undertaken. Description. The house could not be fully defined due to the fact that Feature 32 appears to intrude on this feature. As a result, only a 50-em strip of floor and parts of the western and southern walls remain intact (see Figure 7.16). The undisturbed area is situated outside of the walls of Feature 32. This interpretation is supported by the floor assemblage, which was found only within the SO-cm strip, and by the lack of a hearth, which suggests that the construction of Feature 32 removed most of the feature. From what is left of the western wall and a portion of the southern wall it can be surmised that the long axis was approximately 5.6 m long with a northeast-to-southwest orientation. The fact that the remaining walls appear to be parallel to the walls of Feature 32 suggests that both features shared the same southeast orientation, although this cannot conclusively be determined. The walls were only 9 em high below the sheettrash layer, making this one of the shallowest houses at the site. The floor consists of a compacted sterile alluvium with no evidence of preparation. There are, however, small areas of burning. The floor assemblage contained a large sherd disk, lithic debitage, and isolated sherds. The only floor features were a possible pit, five postholes, and three possible postholes. The locations of most of the postholes are somewhat unpatterned. All of the postholes are approximately 10 cm in diameter and 5 em to 10 em deep. The fill of the house consisted of a brown sandy silt with inclusions of reddish brown sterile sediments and black ash. The sterile sediments within the fill may be from backdirt placed there when the pit for Feature 32 was constructed. The artifact density was low-to-moderate with many more lithics than sherds. The ceramic density was 28 sherds per cubic meter and the lithic density was 112 lithics per cubic meter. The fill was fairly homogeneous across the entire area thought to be actually within the feature. Internal Features. The only internal feature was an unexcavated possible pit or rodent burrow. The Deer Creek Drainage 129 Intrusive Features. It is most likely that Feature 32 intruded into this house removing all but the western 50 em. Alternatively, it is possible, although not considered likely, that Feature 36 overlies Feature 32 and the floor was difficult to follow where it was built on the lower pithouse fill. Abandonment and Postabandonment. The small size of the remaining house area makes any interpretation difficult and necessarily tenuous. Burning of the house is indicated by patches of burned floor and deposits of black ash within the fill, although this is not entirely conclusively due to the small area excavated. The low artifact density of the fill suggests that the house filled naturally after it collapsed. At some later time, Feature 32 was constructed, removing most of Feature 36. The fact that both houses appear to be similar in orientation and size suggest that Feature 32 may be a replacement for Feature 36. Feature 59 This pithouse was identified during the data recovery phase. A 2-m by 2-m control unit was first excavated on the east side of the trench. Two extramural surfaces, Features 20 and 66, were identified in this unit, resulting in some mixing of the cultural strata. Only a 50 percent sample of this house was excavated. The remaining half was excavated down to the lower extramural surface, Feature 66. Description. Feature 59 is a rectangular pithouse with rounded comers and a large ramped entryway oriented toward the southeast (Figure 7.19). The pithouse measured 5.9 m by 3.9 m (23.0 square meters). The entrance has an area of 2.66 m (1.9 m by 1.4 m) and opens toward pithouse Feature 9, although the two are not believed to be contemporaneous. The fill of the house was extremely complex. Two extramural surfaces (Features 20 and 66) were encountered within the fill extending over the northwestern third of the house. Although both surfaces contained ash pits there was no evidence for burning, either in the fill above or on the floor. The fill of Feature 59 did show evidence of burning and contained a light brown compact silt with inclusions of charcoal and burned daub. The artifact density within the fill was moderately high, containing approximately 100 sherds and 77lithics per cubic meter. The floor of the pithouse was not prepared or plastered but consisted of a compacted sterile substratum containing burned areas. The floor assemblage was limited but included three mano fragments, a mano, and a hammerstone within the entryway, and a tabular knife, lithic debitage, and isolated sherds scattered throughout the rest of the house. Three floor features were recovered within the excavated sample; a hearth situated directly in front of the entryway, and two small pits. There were also four postholes exposed. Three of the postholes were along the walls of the house ranging in size from 15 to 20 em in diameter and 10 to 15 em deep. The fourth posthole was for one of the main central roof supports, and measured 40 cm in diameter, 51 cm deep, and contained a rock lining. The posthole partially cuts through the hearth of Feature 65, an earlier pithouse into which Feature 59 intrudes. The walls of the house were defined by the distinction between the cultural fill and the sterile substratum. The southern wall of the house intrudes on pithouse Feature 65. Because the depth of the floor of the two houses are identical, the wall was identified by the difference between the fill of Feature 65, which contained large pieces of charcoal and darker fill, from the fill of Feature 59. In addition, the upper portion of the house pit within which the house was constructed was larger than the actual house, forming a lip around the top of the wall approximately 10 em to 40 em wide. The walls within the house were gently sloped with a height of 32 em. Internal Features. Feature 59-1 is a small plastered hearth situated approximately 1.0 m in from the entrance. It measured 20 cm in diameter and 10 cm deep. The small basin is surrounded by a band of oxidized floor 130 Chapter 7 AZ 0: 15: 52 (ASM) Flalures 59 and 65 NI50 NI49 NI48 NI47 , I I , Prob.ble Idoe , c- NI46 NI45 Fl •. 59 1 N NI44 -c· ./FOG. 8 5 - , o • ® @ 50 L..--....J em Potlh~. F.... 8' Pottholl F,o. 15 Unnca.,IItU 111ft a Sh,r. 4 lithic .. MOM c:;, Ro.. C3 Iy,nl. Mo. NI4l1 ., 1ft N 1&1 1 N 1&1 OJ 1ft N ... ! 1ft N 1&1 1&1 : . N PROFILE Fl•. 59 PROFILE F••. 65 Figure 7.19. Features 59 and 65 at the Deer Creek site. The Deer Creek Drainage 131 extending out from it for 15 em. When the archaeomagnetic sample was collected, a lower unplastered hearth was found directly beneath the plastered one, indicating at least one remodeling episode. Feature 59-2 is a small pit approximately 40 em in diameter which slightly overlaps the central posthole. There is a possibility that this pit actually is associated with Feature 65; since the two floors are at the same level there is no way to be certain. The fill of the pit was a loose brown sandy silt. Feature 59-4 is another small pit situated 20 em east of Feature 59-2. The fill consisted of a loose brown sandy silt with a few artifacts. It was oblong-shaped, 45 em by 35 em, and had a depth of 12 em. It is possible that this pit also is associated with Feature 65 instead of Feature 59. Intrusive Features. Features 20 and 66 are two extramural surfaces within the fill of Feature 59. Feature 59 intrudes on Feature 65, an earlier smaller pithouse, leaving very little of the earlier house. Abandonment and Postabandonment. Feature 59 was constructed over much of Feature 65, and in fact the structures share the same floor. It is likely that the depth of the floor of Feature 59 was determined during pithouse construction when the builders reached the compacted floor surface underneath the relatively soft fill. The lack of a substantial floor assemblage and the presence of burned roof fall on the floor indicates that it was abandoned prior to the structure burning. The pithouse depression was then used as an extramural surface or brush kitchen at two different times allowing for a build-up of fill between the two extramural surfaces. After the uppermost surface, Feature 20, was abandoned the shallow depression filled in primarily with sheet trash. The archaeomagnetic date suggests that this was one of the latest structures occupied at the site, dating to sometime during the late Santa Cruz or early Sacaton phase (AD. 900-1050). Feature 62 This feature was first identified in trench profile as a thin band of darker soil at the bottom of the sheet trash, and it was assumed to represent a heavier or more concentrated sheet-trash deposit. When the sheet trash was mechanically removed a plastered hearth was exposed. A small portion of the fill remained north of the stripped area and on the west side of the trench, so these areas were excavated through control units. A I-m by 2-m unit was excavated on the north edge of the house, but due to the backhoe stripping, only the lower 13 em of fill remained. So little of the house was situated west of the trench that the 2-m by 2-m control unit excavated there was mostly a sample of sheet trash. The remainder of the house was hand-stripped to 5 em above the floor (Stratum 19) and approximately a third of the floor (Stratum 20) was exposed. Description. This house was badly disturbed by the backhoe stripping. From what remains it appears to be a rectangular house with rounded comers oriented toward the southeast. The east and north walls are the most clearly defined and the location of the hearth relative to these walls suggests a southeast orientation. The estimated size of the house is 21.3 square meters with dimensions of 5.6 m by 3.8 m. The height of the walls beneath the sheet trash is approximately 25 cm, only evident along the west wall. The fill of the house consisted of a dark gray, sandy loam with a low-to-moderate artifact density. The density is approximately 72 sherds and 48 lithics per cubic meter. The fill and floor were fairly heavily disturbed by rodents and roots. There was no evidence of preparation of the floor. It was defined by the contrast between the dark cultural fill and the red sterile substratum. The only floor artifact was a tabular knife fragment found along the north wall. No postholes were located although only a third of the floor area was exposed. The only floor feature found was the plastered hearth, Feature 62-1. Internal Features. Feature 62-1 is a plastered hearth situated in front of the probable location of the entrance. It is oval- shaped and measures 31 em by 26 em by 7 em deep. The top was scraped by the backhoe stripping, which also produced numerous cracks in the remaining plaster; an archaeomagnetic sample was not recovered. A double line of plaster along the north edge indicates possible remodeling. The double line does not continue all the way around the hearth. The fill was a dark sandy loam similar to the house fill. 132 Chapter 7 Intnlsive Features. Feature 28, a roasting pit, intrudes on the southwest comer of the house. Abandonment and Postabandonment. It is unclear whether this house burned but it appears to have been abandoned prior to its collapse because so little was left on the floor. The dark fill suggests burning but there was no charcoal or burned daub in the fill, nor burned areas on the floor. Possibly the fill was dark only from cultural debris. The site continued to be occupied after the house was abandoned as indicated by the intrusive roasting pit. Feature 65 Feature 65 was first identified during the excavation of pithouse Feature 59. Feature 59 intrudes on the majority of this feature, leaving very little fill associated with it. The entire remaining area was excavated as an irregularly shaped control unit. Description. Feature 65 appears to have been a small rectangular pithouse with rounded comers and at least one extremely well-plastered wall. All that remains of this house are the south wall and short adjacent portions of the east and west walls to the point where they are truncated by Feature 59. The hearth is also present, as well as a short portion of the end of the entryway extending beyond the west wall of Feature 59 (Figure 7.19). The entrance appears to be narrow and ramped, oriented toward the west. The walls are thickly plastered with caliche from the floor up to a height of 33 em where they extend across the presumed prehistoric ground surface (or possibly a bench) for an additional 30 em. This suggests that unlike most of the houses at the site, this was a true pithouse that incorporated the pit walls into the structure, rather than a house-in-pit. It cannot be precisely determined whether the other walls were constructed in this manner, although the fact that adjoining sections of the east and west walls were also plastered suggests similar construction methods. However, the entryway shows no evidence of plaster. The wall plaster begins at the floor but does not extend across the floor. It is heavily burned. The south wall measures 2.7 m long and the distance from it to the hearth is 2.2 m, suggesting that the long axis was probably approximately 4.4 m. This makes an estimated area for the house of 12.1 square meters, one of the smaller houses at the site. The floor of Feature 65 was shared by Feature 59. It was not plastered but was compacted through use. As mentioned above, it is likely that the depth of the floor of Feature 59 was determined during pithouse construction when the builders reached the compacted floor of Feature 65 underneath the relatively soft pithouse fill. There are several areas of burning on the floor, particularly within the portion that was shared by the two houses. Burning was much less in the area outside of the estimated limits of Feature 65. This suggests that Feature 65 was more heavily and completely burned than Feature 59. This is also supported by the amount of burning evident in the fill. Although there was some evidence for burning within the fill of Feature 59, the fill of Feature 65 was much darker in color and contained large pieces of charcoal and burned daub. The artifact density within Feature 65 was very low with approximately 12 sherds per cubic meter. No lithics were recovered. This may not be an accurate measure of the density, however, since it is based on a very small sample from the edge of the house. There were no floor artifacts in the small area that was not intruded on by Feature 59. The only definite floor feature is a plastered hearth, Feature 65-1. There are two floor pits, Features 59-2 and 59-4, described with Feature 59, that could be related to this house instead. Five postholes have been tentatively included with this feature but, except for one, they could actually be part of Feature 59. The one definite posthole is situated in the southeast comer of Feature 65, outside of the walls of Feature 59. The other four postholes have been associated with Feature 65 because they line up with what is thought to be the location of the western wall. The postholes ranged in size from 6 cm to 25 cm in diameter and were 7 em to 15 em deep. No exterior postholes were located although it seems likely, given the fact that the plaster extended out onto the prehistoric surface approximately 30 cm, that they were present. The Deer Creek Drainage 133 Internal Features. Feature 65-1 is a hearth lined with the same thick plaster as the south wall of the house. The plastered collar is intruded into by a main support posthole of Feature 59. The similar type of plaster, its location in front of the entrance, and the later intrusion, suggest that this is the hearth to Feature 65 and it is not associated with Feature 59, even though it is at the same depth as Feature 59-1, the hearth associated with that pithouse. The fill of the hearth was a dark gray compacted ash; some white ash was also present. The feature measured 40 em in diameter and was 13 em deep. An archaeomagnetic sample was recovered. Intrusive Features. Feature 59 intrudes on this feature, removing almost all of the house except for the southern edge. Features 20 and 66, two extramural surfaces within the fill of Feature 59, also overlie Feature 65. Feature 65 does not intrude on or overlie any features. Abandonment and Postabandonment. Feature 65 was the first feature to be occupied in this relatively complex sequence of superposition. It cannot be determined if it had a floor assemblage since so little of the floor remained. It is evident, however, that the house underwent a complete and heavy burning, at least in the remaining portion. After the house burned the same area was used for three successive occupations; first came pithouse Feature 59, which removed most of Feature 65, and then came extramural surface Feature 66, and finally extramural surface or brush kitchen Feature 20. EXTRAMURAL FEATURES Extramural Surfaces Feature 5. This is a relatively poorly defined extramural surface overlying pithouse Feature 6. In profile the surface was originally thought to represent an overlying pithouse. Upon excavation, however, it was found to not contain definable pithouse characteristics, such as a prepared floor, walls, floor features, or an entrance. Instead Feature 5 appears to be an ill-defined use surface, perhaps associated with a large roasting pit, Feature 43. The surface was difficult to follow and not completely excavated. It covered an area of approximately 30 square meters. Most of the excavation of this feature occurred while searching for the boundaries of Feature 6. Several flat-lying sherds were found in possible association with this surface, however, the surface was overlain by a fair amount of sheet trash, so the true relationship of these artifacts to the surface is unclear. Feature 20. This feature is an extramural surface encountered in the fill of pithouse Feature 59 (Figure 7.20). It is somewhat similar to features described as brush kitchens by Haury (1986:229-231) at Roosevelt 9:6 and Hohmann (1985:116) on the Ash Creek Project, both located within the Lower Tonto Basin. It was only identified within the northern section of the house and continued outside of the house wall only in the northwest corner. Once the surface was recognized, a I-m by 2-m unit was excavated to expose the surface and determine how it related to the walls of Feature 59. In the remainder of the feature excavation only the fill from the 5 em level above the surface was screened through Y4-inch mesh. The surface extended over an area 5.3 m by 3.6 m and sloped from the house edges toward the middle of the house where there was an associated ash pit, Feature 54. It was identified by a thin layer of reddish sandy gravel on a slightly compacted, but uneven, surface. The only artifact in direct association was a tabular knife. The fill above the surface contained a light brown sandy silt with charcoal flecking and a moderately high artifact density. There were approximately 200 sherds and 75 lithics per cubic meter in the to-em level above the surface. The feature overlies two pithouses, Features 59 and 65, and another extramural surface, Feature 66, approximately to em beneath this one. An ash pit, Feature 54, is the only feature associated with this surface. The uneven nature of the surface in conjunction with the large size and irregular outline of the ash pit and the lack of evidence for architecture, suggests that Feature 20 is an extramural surface or brush kitchen rather than a pithouse. The size and oxidized nature of the ash pit suggests that it was used for cooking. The slope of the floor toward the hearth within the original pithouse depression would have given some protection from the wind. Unlike Roosevelt 9:6, no actual windbreak was observed (Haury 1986:229-231), although the feature may have served a similar type of function. 134 Chapter 7 AZ 0: 15:52 (ASM) Flaturll 20, 59, 66 q J NI50 NI49 . Feature 20 o / ,- I NI48 / / / " 0/ NI47 (ji9 I 68 ~ / ! :J: :J: U Z III II: ~ u Z II: III ~ .. . NI46 . . . . , .. Feature 59 NI45 i NI44 Probable edge Extrapolated edge NI43 o Pit Sherd ~ Rock T en N 0> o o '" I&J PROFILE I : Fea. 20 I 1- - --.-- - - - -=-.:..--:..;-p:::;:;:R~~ -;.:- Figure 7.20. Extramural surfaces Features 20 and 66, superimposed on pithouse Feature 59 at the Deer Creek site. The Deer Creek Drainage 135 Feature 54. This irregularly shaped ash pit originates on the surface of Feature 20. It measured 55 em by 30 em and was 12 em deep. The fill consisted of a fine white ash with several artifacts but no apparent charcoal. The sides of the pit were heavily oxidized but not plastered. It does not have the appearance of a formalized hearth, due to its large irregular shape, although it seems to have been extensively used. The large size and irregular shape suggest repeated episodes of use and possible remodeling. Feature 66. This extramural surface was also within the fill of pithouse Feature 59, approximately 5 em above the floor and 10 em below the surface of Feature 20. It is much smaller than Feature 20, underlying only the northwest portion and sharing a common surface near its outer edges (Figure 7.20). The surface measured 2.9 m by 2.4 m (approximately 7.0 square meters). All of the fill between the two extramural surfaces was screened through Y4-inch mesh. The surface was defined by a thin red sandy lens overlying a slightly compacted surface, much like Feature 20. The fill between the two surfaces contained a medium brown sandy silt with no evidence of burning and a moderate artifact density. There were 134 sherds and 80 lithics per cubic meter within the 7-cm-thick fill layer. Several isolated sherds were found on the surface. A single feature, Feature 68, originated from this surface. It is a small unplastered but highly oxidized pit filled with ash. Like Feature 20, Feature 66 appears to be an extramural surface associated with some sort of cooking and food preparation activities. It appears to be less extensive than Feature 20, however. Feature 68. This is a small ash pit originating from the surface of Feature 66. The pit measured approximately 40 cm in diameter and was 3 em deep. The bottom and sides were highly oxidized but not plastered. The fill contained a fine white ash with a few artifacts. No charcoal was observed. Feature 72. This feature was exposed by the backhoe during the mechanical removal of the sheet trash layer. It appeared as an irregularly shaped dark stain with a fairly flat bottom containing three secondary features (Features 73, 74 and 75). The reason for identifying it as an extramural surface rather than a pithouse is because of its irregular shape and lack of evidence for architecture. Furthermore, the feature does not appear to be within a pit, since the fill is identical to the surrounding sheet trash, and no pit walls were visible within the trench profile. Although the likelihood that this is an unburned pithouse is considered to be slim, it is pOSSible, because the backhoe stripped away any evidence for walls that would have been present. This is somewhat supported by the presence of a partially plastered hearth (Feature 73) originating from the surface, although plastered extramural hearths, while relatively rare, are known throughout central and southern Arizona. The feature was tested with a control I-m by 2-m unit excavated in one 2-cm level, sifting all fill through Y4-inch mesh. No artifacts were encountered in the fill. The outline was irregular and measured approximately 3.3 m by 3.0 m with a depth below the stripping of 2 cm. The surface was not prepared but identified solely on the distinction between the dark cultural fill and the red sterile substratum. There was no evidence of roof or wall fall in the fill and there were no artifacts. On the extramural surface there were three flat-lying sherds. Three features originated from this surface; two ash pits, Features 74 and 75 and a hearth, Feature 73. Other possible extramural features that were not excavated but also appear to originate at this level are Features 103 and 104. Feature 73. This is a small extramural hearth. It consists of a shallow basin with a small area of plaster at the surface on the south and east sides and oxidation on the walls and bottom. The feature measured approximately 23 em in diameter and was 8 cm deep. Two ash pits, Features 74 and 75, are probably associated with this feature. Feature 74. This feature is a small extramural ash pit or possible hearth. It measured 50 em north-south by 35 em east-west with a depth of 12 em. It has an irregular shape, mostly the result of rodent disturbance. The pit contained a highly compacted gray green ash with some charcoal flecks and a few sherds. The walls of the pit are oxidized. It probably represents the ash pit used in conjunction with the hearth, Feature 73, situated 20 em to the southwest, or it may be a second hearth. 136 Chapter 7 Feature 75. This is another ash pit situated approximately 60 em north of Feature 74. It is small and basinshaped, measuring 28 em in diameter and 5 em deep. The fill consisted of a compacted fine gray ash with some charcoal inclusions but no artifacts. The walls of the pit were not oxidized like those of Feature 74. This feature was situated 1.15 m from the hearth Feature 73, but may be associated with it. Trash Mounds and Trash Areas Feature 38. This is the only feature at the site that appears to have had no other obvious function than as a repository for trash. It was originally thought to be a possible pithouse but this was negated due to the lack of walls and a level floor. A 2-m by 2-m control unit was started in this feature, but because of a large tree stump in half the unit, it was continued as aIm by 2-m unit below the level of the sheet trash. This was excavated to obtain a sample of the fill and to determine whether the feature was a pithouse. The feature fill was excavated in 10-em levels and sifted through Y4-inch mesh. The feature measured approximately 7.3 m north-south by 11 m east-west and was 38 em deep (Figure 7.3). The fill contained a high artifact density and a large number of rocks in a light gray silty loam. The ceramic density was 181 sherds per cubic meter and the lithic density was 139 lithics per cubic meter. The feature is situated east of the pithouses in an area with other smaller extramural features. Its location is similar to trash areas at other sites, which tend to cluster around the site periphery outside of the residential areas. Because most of the trash at the site is believed to have been tossed over the edge of the Deer Creek terrace or deposited in abandoned pithouse depressions instead of within formally defined trash areas, it is possible that Feature 38 originally served another function, perhaps as a borrow pit for adobe mud, although this could not be determined. Pits Extramural pits were the most numerous feature type at the site (see Figure 7.3). Twelve were excavated and an additional 50 (including possible pits) were plotted on the map after the sheet trash was stripped mechanically from the site. Within the excavated sample, all pits smaller than 1.5 m in diameter were totally excavated as a single unit and screened through Y4-inch mesh. Pollen and flotation samples were collected. Due to the relatively large size of Feature 63, the only pit larger than 1.5 m in diameter, only half of the pit was excavated. Metric data from these features are presented in Table 7.3. Feature 33. This small pit was identified in the backhoe trench. Estimating from the remaining portion, it probably measured around 30 cm in diameter and was 20 em deep. The pit originated from the sheet trash layer and contained a dark gray fill with charcoal flecks and very few artifacts. There was some root disturbance. The walls of the pit were defined by the contrast between the dark cultural fill and the light brown sterile substratum. There was no evidence for in situ burning. Feature 45. This small pit is located directly outside the wall of Feature 18, a pithouse. It was described in the description of Feature 18 since it is felt to be possibly related to the pithouse. A phallic argillite censer was recovered from the pit. Feature 56. This is a small pit measuring 33 cm in diameter and 12 cm deep. The pit was visible beneath the sheet-trash layer as a semicircular outline of a thin oxidized daub-like material, and it was thought at first to be a crematorium. The fill contained a cluster of rocks and artifacts as well as charcoal in a dark gray matrix. The walls and bottom showed no evidence of burning below the upper edge. The pit is situated at the edge of the cemetery area west of pithouse Feature 11 where there are several other extramural pits and roasting pits. The Deer Creek Drainage 137 Table 7.3. Metric data from sampled extramural pits at AZ 0:15:52. Feature Number Feature Type Percent Excavated N-S (m) E-W em} Depth CMBDl Volume em3) Comments 33 45 Pit Pit/posthole 100 100 0.30 0.30 0.15 0.30 9.86-10.06 10.34-10.53 0.005 0.009 56 58 63 Pit Pit Trash pit Bell-shaped pit 100 100 0.40 0.40 200 0.35 (I) 0.65 (B) 0.50 0.28 0.60 0.30 0.35 0.45 0.58 0.45 0.33 0.40 155 0.35 0.65 0.35 0.28 0.63 0.30 0.35 0.45 0.40 0.45 10.71-10.83 10.66-10.82 10.71-10.96 10.76-11.17 0.008 0.013 0.403 0.086 10.62-10.74 10.62-10.67 10.87-11.05 10.64-10.76 10.61-10.66 10.72.10.82 10.60-10.68 11.02-11.18 0.011 0.002 0.011 0.006 0.003 0.010 0.010 0.016 64 74 75 76 77 79 80 81 119 MBD Ash pit Ash pit Burned pit Burned pit Pit Pit Pit Pit 60 100 100 100 100 100 100 100 100 100 Prob. assoc. with F. 18 = Meters Below Datum Feature 58. This is a small basin-shaped pit located approximately 70 em to the west of Feature 56. The pit measured 40 em in diameter and 12 em deep, with no evidence for burning. The fill consisted of a dark brown sandy silt containing sherds (too small to collect), lithic debitage, and a metate fragment. The only rocks in the pit were three unworked cobbles at the top. The function of the pit is unknown. Feature 63. This large pit is located in the same area as Features 56 and 58. It is roughly oblong, although fairly irregular in shape, and measured 2.0 m by 1.55 m and was 25 em deep. Because of its size the pit was sampled. The top 11 em of the entire pit were excavated to define it. Then the pit was bisected and only the north half was excavated. The fill contained a moderately high amount of artifacts including sherds, Iithics, ground stone, shell, and a small amount of cremated bone, within a dark gray matrix containing fire-cracked rocks. On the bottom of the pit was an inverted trough metate with a mana below it. The mana was in the trough of the metate. Both pieces had ground argillite on their grinding surfaces. The function of the pit is unknown; the high artifact density suggest the use of this feature as a trash pit, although the mano and metate with ground argillite suggest that the function may have been initially different or changed through time. Feature 64. This is an extramural bell-Shaped pit. It measured approximately 35 em in diameter at the top and 65 em at the widest point within the pit. It was 41 cm deep. The rim of the pit was oxidized but there was very little evidence for burning within the pit. Possibly the ground surface here was burned before the pit was constructed. The fill contained a light tan sandy silt with very few artifacts. There was a large rock embedded in the north wall. This pit may have been used for storage and then filled in naturally with alluvium. It is somewhat uncommon to find extramural bell-shaped pits, although they have been recorded from other areas of the southwest. It is possibly that the pit was within a ramada or some other type of ephemeral structure obscured by the sheet trash or other disturbance factors. Feature 76. This burned pit was situated within the cemetery area and thought initially to represent a crematorium, so all fill was screened through lIS-inch mesh. No bone was recovered. The walls and bottom of the pit were oxidized and very compact. The fill contained dark silty loam with charcoal inclusions 138 Chapter 7 throughout and a few artifacts; sherds, lithics, and a shell bracelet fragment were recovered. The function of this pit is unclear and it is not known whether it is contemporaneous with the cemetery. Feature 77. This feature is a small burned pit in the northern part of the site. It is situated 10 em northwest of a roasting pit, Feature 121, and may be related to it. It measured 30 em in diameter and was 12 em deep. The walls were oxidized and the fill contained a large amount of charcoal, a few fire-cracked rocks, and two sherds. The bottom was obscured by rodent disturbance. Feature 79. This small pit is located in the same portion of the site as Features 77 and SO. It measured approximately 35 em in diameter and was 5 em deep. The fill contained charcoal, ash, and a few fire-cracked rocks within a dark gray matrix. No artifacts were recovered. The pit walls exhibited no evidence of in situ burning. This suggests that the fill came from elsewhere, possibly from Feature 121, a roasting pit situated 1.15 m to the northeast. Feature 80. This is a small pit in the same area as Features 77 and 79. It is very similar to Feature 79 and is located 50 em southeast of it. The pit measured approximately 45 em in diameter and was 10 em deep. The fill contained a few fire-cracked rocks and sherds within a dark gray ashy matrix. Like Feature 79, the walls of the pit were not oxidized, suggesting that the fill came from another feature, such as Feature 121. Feature 81. This is another small pit in the same area as the preceding pits. The pit was irregular in shape, and measured 58 cm by 40 em by 8 em deep. The fill contained a dark gray matrix with a high ash and charcoal content. A cluster of plainware sherds from the same vessel was found in the top of the northwestern portion. There were several burned rocks of varying sizes embedded in the walls and bottom. It appears that burning occurred within the pit, although its function is still unclear. Feature 119. This is a small pit within the cemetery area. There was no evidence for in situ burning nor was any human bone recovered. It measured 45 cm in diameter and was 16 em deep. The fill consisted of a dark gray ashy matrix containing a few artifacts. The contemporaneity of this feature with the cemetery features is unknown. Roasting Pits An excavated sample was recovered from ten roasting pits, while another four (including possible roasting pits) were identified and plotted on the site map after the site was mechanically stripped of the sheet-trash layer (see Figure 7.3). Metric data on the excavated or sampled roasting pits are presented in Table 7.4. Four of the features (Features 17, 43, 47, and 60) are clusters of fire-cracked rocks, mostly in pits, that do not contain strong evidence for in situ burning. The walls were not oxidized and the fill did not contain high quantities of charcoal. These four roasting pits were totally excavated. Four other roasting pits (Features 28, 86, 118, and 121) contained burned fill and all but Feature 121 were oxidized. These features were sampled by bisecting the outline exposed by the stripping and excavating half of the fill with an emphasis on recovering flotation, radiocarbon, and pollen samples. The final two roasting pits, Features 15 and 44, were large areas of fire-cracked rocks on or near the present ground surface showing little evidence of burning. Feature 15 may represent an Apachean mescal or roasting pit; the function of Feature 44 is unclear although it may be Apache-related as well. A sample control unit was excavated in each of these features. Feature 15. This feature is a large scatter of fire-cracked rocks that was partially visible on the original ground surface. Two backhoe trenches were excavated across it, one north-to-south and one east-to-west, to identify the area of deepest deposits and to determine its overall shape (Figure 7.2). A sample 1-m by 2-m test unit was excavated from the surface to the bottom in 10 em levels, screening through ¥.i-inch mesh. The fill consisted of a light tan, sandy silt containing abundant fire-cracked rocks with very few artifacts and almost no charcoal flecking. The fire-cracked rock scatter measured approximately 12 m north-south by 9 m east-west and was 40 em deep at the deepest point visible in the backhoe trenches. The depth varied considerably across the feature, however, although smaller pits within the general scatter could not be identified. There was no The Deer Creek Drainage 139 Table 7.4. Metric data from roasting pits at AZ 0:15:52. Feature Number 15 17 28 43 44 47 60 86 118 121 Feature Type Percent Excavated Apachean roasting pit 2 Roasting pit 100 Roasting pit 50? Roasting pit 100 100 Roasting pit? Roasting pit debris 100 Roasting pit 100 Roasting pit 50 Roasting pit 50 Roasting pit 50 Depth (MBO) Volume (m3) N-S (m) E-W (m) 120 0.72 1.75 1.50 4.1 9.0 0.40 1.75 1.00 5.2 11.30-11.70 10.44-11.08 10.33-10.97 10.26-10.69 10.18-10.32 22.61 0.10 1.01 0.34 234 0.90 0.40 0.90 1.00 1.10 0.90 0.38 0.90 1.00 1.35 10.78-10.85 10.66-10.90 10.79-11.08 11.27-11.71 10.64-11.10 0.03 0.02 0.12 0.23 0.35 Oxidation + + + MBD = Meters Below Datum visible oxidation and little charcoal or ash indicative of in situ burning. The appearance of the pit, along with the partial alluviation, suggests that the pit might have been related to the Apachean occupation of the site and functioned as a mescal pit. No Apache ceramics were recovered from this area, however, and no charred remains were identified in the flotation samples. A single uncarbonized marginal tooth from an agave leaf was identified, although it is unclear if this is related to the function of the feature. No pieces of charcoal large enough for radiocarbon analysis were recovered. Feature 17. This feature is a small roasting pit situated in an area west of pithouse Feature 11, where numerous extramural features were recorded. It was cut by a backhoe trench but may have been around 72 em in diameter and 64 cm deep. The top appears to have been obscured by the sheet-trash layer. The fill consisted of a gray brown, sandy silt with numerous fire-cracked rocks and a few artifacts. The sides of the pit were not oxidized but the fill showed evidence of burning. Due to the lack of oxidation it is possible that this was not a roasting pit but a trash pit containing roasting pit fill. Feature 28. This large roasting pit was first identified in trench profile. It measured approximately 1.75 m in diameter and was 64 em deep. The fill contained abundant fire-cracked rocks, which increased in size with depth, as well as charcoal flecks and oxidized sediments within a dark gray ashy matrix. The walls of the pit were moderately oxidized. No artifacts were recovered. Approximately 50 percent of the feature was sampled. It intrudes on the south side of pithouse Feature 62. Feature 43. This roasting pit is intrusive into pithouse Feature 6 and probably originated from an overlying extramural surface that was obscured by Feature 44, a possible Apachean mescal pit. The overlying extramural surface may be part of Feature 5, another extramural surface, although due to disturbance factors the relationship among all of these features was never clearly established. Feature 43 is irregularly shaped but measured approximately 1.5 m by 1.0 m and was 43 em deep. The bottom of the pit cut through the floor of Feature 6. The shape of the pit at the bottom suggests that three separate pits may have been present and intruded on each other. The fill contained numerous fire-cracked rocks and artifacts, including several pieces of ground stone, within a dark brown matrix. The pit walls were not oxidized, nor was there much charcoal within the fill. As a result, it is possible that this feature served as a dump for roasting debris and was not a true roasting pit. 140 Chapter 7 Feature 44. This was a scatter of fire-cracked rocks overlying Feature 6, a pithouse, and Feature 43. The scatter was slightly below the present ground surface, and was found when the fill of Feature 6 was handstripped to define the pithouse walls. It appears to be similar to Feature 15, although smaller, and may also represent an Apachean mescal or roasting pit given its shallow depth and the recovery of several Apache sherds from this area. The feature consisted of a dense scatter of fire-cracked rocks, many of which were broken pieces of ground stone, with little evidence for in situ burning. A 2-m by 2-m control unit was excavated over the edge of the feature. The feature measured approximately 4.1 m north-south by 5.2 m eastwest and was 14 em deep. The fill contained a moderately high artifact density, primarily composed of lithic debitage within a tan sandy silt substrate. The fire-cracked rocks all appeared to be scattered on a level surface rather than in a defined pit. The level surface may be part of what was defined as Feature 5, an extramural surface, although this is unclear due to the difficulty in following and joining these surfaces. Beneath the rocks there was no evidence of oxidation. Feature 47. This feature is a cluster of fire-cracked rocks first identified in the fill of pithouse Feature 21 and the sheet trash above it. No pit outline was visible, nor was there any indication of burning, suggesting that this represents a single dumping episode from a roasting pit rather than an actual roasting pit. The feature measured approximately 90 em in diameter and was 7 em deep. Ceramics and lithics were recovered but are probably associated with the sheet trash and pithouse fill rather than directly with the fire-cracked rock cluster. Feature 60. This feature is located west of pit house Feature 11 in an area where numerous extramural features were recorded. It is a small pit filled with fire-cracked rocks and charcoal flecks within a dark gray matrix. The pit measured approximately 40 cm in diameter and was 24 em deep. The walls of the pit were not oxidized, although the fill is similar to that from well-defined roasting pits. This lack of oxidation also occurred in several other features identified as roasting pits, such as Feature 17, located 5 m to the south. Feature 86. This roasting pit was exposed during the mechanical extramural stripping. It is located within the southern periphery of the site on the southern edge of the extramural area west of pithouse Feature 11. It is a shallow roasting pit measuring approximately 90 cm in diameter and 29 em deep. The fill contained primarily medium-sized fire-cracked rocks and charcoal flecks within a dark gray ashy matrix. No artifacts were recovered from the half of the feature that was excavated. The walls of the roasting pit were oxidized. Feature 118. This feature is a large roasting pit originally exposed when the sheet-trash layer was removed by the mechanical stripping. It is situated in the southeastern part of the site, south of the cemetery area. The feature measured approximately 1.0 m in diameter and was 44 cm deep. The fill contained fire-cracked rocks and charcoal within a dark gray ashy matrix. One sherd was recovered from the half that was excavated. The walls of the feature were heavily oxidized and were several centimeters thick. Cobbles were found embedded in the walls, extending through the oxidized layer into the sterile substratum. Feature 121. This roasting pit is located in the northeast portion of the site. It is slightly oblong, measuring 1.1 m north-south by 1.35 m east-west with a depth of 46 cm. The fill contained abundant fire-cracked rocks, some animal bone, and a few sherds within a dark gray matrix. The walls of the pit were not oxidized but, like several other roasting pits at the site, the fill suggests it was used either as a roasting pit or the feature functioned as a dump for roasting debris. Extramural Hearths Feature 61. This is a small unplastered extramural hearth. It was circular in shape, with vertical walls and a basin-shaped bottom, and measured 42 cm in diameter by 20 cm deep. The hearth was oxidized only along the southern rim and contained burned artifacts and ash. The little oxidation suggests that it was not used very extensively. The Deer Creek Drainage 141 Rock Concentrations Two features (69 and 83) were identified as rock concentrations. Neither of these was excavated and their function and nature are unknown. MORTUARY FEATURES Twenty-one mortuary features were recovered at the site, including 13 crematoriums (or primary cremations), 6 secondary cremations, and 2 inhumations. The majority of these were found within a cemetery area in the southwest portion of the site (Figure 7.3). This area appears to date to the Gila Butte phase occupation, while some of the isolated mortuary features may be related to the smaller later occupations. Due to the extensive stripping undertaken at the site it is believed that all mortuary features were recovered. The osteological analysis of the recovered human remains is presented in Appendix C of Volume 3. Crematoriums and Primary Cremations The crematoriums (or primary cremations) were some of the most interesting and potentially significant features recorded at the site. This is due to their relative uniqueness, being unlike most crematoriums or primary cremations recorded in the Hohokam area of the Phoenix and Tucson basins. They consist of small, rectangular, daub-lined, burned pits, some with comer posts, containing small amounts of cremated human bone and charcoal within the fill (Figure 7.21). Some contain mortuary offerings, while others do not. Only three other Colonial period (ca. AD. 750-950) sites are now known to contain similar features; Ushklish (Haas 1971) situated in the Upper Tonto Basin approximately 4 miles to the southeast, and Buh Bi Laa and East Fork Village (Halbirt and Dosh 1991), situated in the White Mountains some 80 miles to the east. Although some similarities exist between these features and what have been called primary cremations in the Phoenix Basin and Gila Bend areas of the Hohokam (Saul 1988:430; Wasley and Johnson 1965), this type of mortuary pattern usually has been attributed to Classic period populations. Thirteen crematoriums were identified and totally excavated. Each was excavated as a single unit with the fill being removed in 20-em levels or smaller natural layers. All fill was screened through lI8-inch mesh, saving all decorated sherds, plainware sherds larger than the size of a quarter-dollar, all other artifacts, and all fragments of bone. Pollen and flotation samples were also recovered. Due to their relative unique nature, the crematoriums are described in some detail. Metric data from these features can be found in Table 7.5. A summary of their characteristics is presented here. Only one crematorium, Feature 37, was located outside of a 25-m by 25-m (625 square meter) cemetery area. They are all fairly rectangular, of similar size, and all but two (Features 52 and 70) are oriented within 13 degrees of east-west (Figure 7.22). Five of the crematoriums have been truncated by a backhoe trench or another crematorium, so their lengths are unknown (Figures 7.23 and 7.24). Using only the crematoriums for which we have the full length, the mean length is 1.39 m with a standard deviation of 19 em. This includes Feature 37, the crematorium that was not within the cemetery and is significantly longer than any of the others. Without this feature the mean length is 1.34 m with a standard deviation of 11 cm. The widths are equally uniform with a mean of 59 em and a standard deviation of 12 em. The depths also vary little, with a mean of 36 cm and a standard deviation of 6 em. The small size of the standard deviations strongly indicates the uniformity of these features. These measurements are also similar to those from the site of Buh Bi Laa, where the crematoriums averaged 1.5 m in length, 50 em wide, and 60 em deep (Halbirt and Dosh 1991). All thirteen crematoriums have a lining of a burned daub-like substance on the walls (Figure 7.25). It is most evident near the tops of the crematoriums, which facilitated the identification of the feature outline. The quantity and quality of this layer varies considerably, from a few small patches at the top to the entire lining 142 Chapter 7 Figure 7.21. Photograph of crematorium Feature 46. of the walls. It does not have the appearance of being heavily burned, although the warping of ceramic offerings from some of the pits suggests fairly intense burning. None of the crematoriums have the daub lining extending across their bases, however, which consist of the sterile substratum, with light oxidation in some cases. An archaeomagnetic sample of this lining was collected from Feature 71. The dating of this sample, AD. 725-855, is within the Gila Butte phase and consistent with the majority of the archaeomagnetic dates recovered from the site. Three of the features contained rounded comer protrusions that appear to have been for posts. Features 71 and 85, approximately 1.5 m apart, were the only two to have evidence of four comer posts (Figures 7.26 and 7.27). Feature 31 had two comer posts in the west end (Figure 7.28). Possibly these posts were used to support a funeral pyre. The crematorium features at Ushklish also contained comer posts. Two had four comer posts and the third had just two, at the northeast end (Haas 1971). Comer posts were found at the site of Buh Bi Laa as well (Halbirt and Dosh 1991). The upper fill of the crematoriums generally consisted of a very light-colored sandy silt, much like the surrounding sterile substratum, suggesting the deliberate filling in of these features with relatively sterile soil. The lowest levels of the features generally contained the concentrations of charcoal and bone. There are The Deer Creek Drainage 143 AZ 0:15:52 (ASM) Feature 48 NI14 J: U Z 1&.1 ...a: -A' NI13 ... . Ox i dized soil 0 Rock -'X' axs:s:a Daub lining Cluster of cremated bone Root disturbance r oI 25 , cm N PROFILE Figure 7.22. Crematorium Feature 48 at the Deer Creek site. 144 Chapter 7 AZO : 15: 52 (ASM) Features I, 70 NI17 e Feature 70 I / / I / NI16 i I I I 20cftl " N u Z L&I / I I I ........ _-.:/ :z: I Features 1,70 Mlled 0: .... Feature I / I - Oaub lining ( ) Pit @ Stone bowl 1 N210 ------ -- -"- ...... ...... ....... o " ", N209 "- O-Pol ishing stone ~-Sherd cluster N20B a l: o Z LLI II: I- GS-c:() N207 GS 00.2 o o 000,' o D If. .... / I o-Hommerstone I ():\\." PossIble Hearth " Area 0' .. ''''', N206 ..... , a Probable edge N205 ... I ....... o CD LLI LLI I N cm PROFILE Figure 7.33. Feature 1 at the Hilltop site. 170 Chapter 7 Feature 6 This feature was initially identified in profile in Trench 9 during the testing phase. The first stage of data recovery consisted of setting up and excavating a 2-m by 2-m control unit on the west side of the trench. The Stratum 10 fill sediments were excavated in two levels, one 20 em deep, the other 10 em deep. By the end of the second Stratum 10 level it became clear that only about half the control unit fell within the house. Therefore, the size of the unit was reduced to 1 m by 2 m for the Stratum 19 level. All fill sediments from the control unit were screened through Y4-inch mesh, and all artifacts except plainware sherds smaller than a quarter-dollar in size were collected. A composite flotation sample was collected from the Stratum 19/20 level, and a composite pollen sample was collected from the Stratum 20 level. Because of the nature of the fill sediments (see below), the pit walls were easier to define from below rather than in plan view. Thus, noncontrol Stratum 10 deposits were excavated without screening. Once the basic outline of the house had been defined, the remaining Stratum 19 deposits were removed and sifted through Y4-inch mesh screen. All floor (Stratum 20) artifacts and features were point provenienced and recorded separately. Composite flotation samples were collected from the Stratum 19/20 levels, and a composite pollen sample was collected from Stratum 20. Description. Feature 6 is an oval-shaped pithouse located about 2.5 m north of Feature 1 (Figure 7.34). The house has an estimated area of 12.5 square meters, with a maximum length of 3.8 m and a maximum width of 3.3 m. The house pit was cut into the pediment surface to an average depth of 20 em. The pit walls were nearly vertical, and there were no indications of either plastering or burning on them. A straight-sided, ramped entryway faced to the east; its maximum length was 0.78 m and its maximum width was 0.57 m. The fill sediments consisted of a medium brown, fine-grained, clayey silt with numerous flecks of charcoal and caliche. Occasional patches of an orange sandy silt also were present throughout the fill, and concentrations of ash were found on and just above the floor in the northern half of the house. Artifact densities were lowto-moderate, averaging 60 artifacts per cubic meter. The degree of root and rodent disturbance was also moderate. The floor was defined as a compacted cultural surface overlying a calcic horizon. The compaction was continuous but uneven; the floor varied in depth by as much as 10 em to 12 cm in places. Occasional scorched areas were noted, especially in the northern half of the structure. The floor assemblage consisted of three manos, several pieces of burned daub and fire-cracked rock, and a 30em-long charred beam. Most of these materials were found near the walls; the central floor area appears to have been kept relatively clean. The only floor features identified were four postholes and a small floor pit (Feature 6-1). Three of the postholes were arranged in a roughly linear alignment extending from the southwest corner of the entryway to the north-central portion of the house. The largest of these had a diameter of 15 em and a depth of 10 cm; it may have held one of the two central support posts for the house. The other central posthole appears to have been destroyed by the backhoe. An external posthole was found outside the house, 20 cm from the northeast corner. The function of this external posthole is unclear. Internal Features. Feature 6-1 is a basin-shaped floor pit located 25 em from the south wall. The maximum dimensions of the pit were 33 em east-west by 20 em north-south, with a maximum depth of 10 em below the house floor. The fill sediments consisted of a dark gray, loosely compacted, silty matrix. Two complete manos and a single plainware sherd were recovered from the fill. Intrusive Features. No features intruded into Feature 6. Abandonment and Postabandonment. Based on the limited floor assemblage, it appears that the house was largely cleaned out at the time of abandonment. The structure then appears to have burned, based on the charred beam in direct contact with the floor and the presence of scorched areas on other parts of the floor. The Deer Creek Drainage 171 AZ 0: 15: 53 (ASM) Feature 6 N2 16 N215 • N214 N213 N212 Posthol. • o M Rock .. '" .. '" . ., '" .,'" '" . ... .. '" PROFILE Figure 7.36. Pithouse Features 14 and 15, with possible wing wall, Feature 20 at the Hilltop site. 176 Chapter 7 through Y4-inch mesh, and all artifacts except plainware sherds smaller than a quarter-dollar in size were collected. A composite flotation sample was collected from the Stratum 19/20 level. The remaining fill sediments were excavated in three stages. First, following the excavation of Feature 14, the fill sediments on the east side of the trench were hand-stripped and Feature 20, a rock alignment built in the fill of Feature 15, was exposed. Feature 20 was mapped, recorded, photographed, and removed. The rest of the fill on the east side of the trench was then excavated and the floor of Feature 15 exposed. Finally, the fill sediments on the west side of the trench were removed and the floor exposed. Noncontrol fill units were not screened, but grab collections were made. All floor artifacts and features were point provenienced and recorded separately. Description. Feature 15 is a very large pithouse (the largest at the site) located on the northeast side of the limestone knoll (Figure 7.36). The house is sub rectangular in shape. It has an estimated area of 34.8 square meters, with a maximum length of 7.1 m and a maximum width of 4.9 m. The house pit was cut into the pediment surface to an average depth of 30 em, and in several places (e.g., the northeast comer) the pit cut into the calcic horizon. The walls were nearly vertical, and there were no indications that they were plastered or burned. A rounded, slightly ramped entryway faced to the east; it measured roughly 2 m east-west by 1.8 m north-south. Two postholes were located along the northern wall of the entryway, suggesting that it was covered. The fill sediments consisted of a light brown, compacted silty matrix that contained numerous flecks of charcoal and caliche. A thin (2-cm to 3-em thick) ash lens was found directly overlying the floor in places, especially in the eastern half of the structure. Artifact densities throughout the fill were generally low, averaging 35 artifacts per cubic meter. The floor follows the calcic horizon for the most part. In the southern quarter of the house, though, a layer of compacted silts overlies the calcic horizon. The only portion of the floor that shows evidence of burning is around two of the hearths (Features 15-1 and 15-2). The floor assemblage consisted of four manos, all located near the outer edges of the structure. Floor features identified include 13 postholes, three hearths (Features 15-1, 15-2, 15-3), and three floor pits (Features 15-4, 15-6, 15-7), one of which (15-6) may be another hearth. The outer postholes averaged 10 em to 13 em in diameter and 10 cm to 15 em in depth. The one central posthole that was identified had a diameter of 25 em and a depth of 22 cm. Internal Features. Feature 15-1 is a small, unplastered hearth located about 0.75 m west ofthe entryway. The hearth was basin-shaped in appearance, and had a diameter of 20 cm and a depth of 6 cm. The pit walls and bottom were well oxidized. The fill sediments consisted of an ashy matrix interspersed with occasional charcoal flecks and nodules of caliche. Feature 15-2 is a small, unplastered hearth located about 1.5 m west of the entryway. The teeth of the backhoe slightly impacted the western half of the feature. The hearth was basin-shaped in appearance, with a diameter of 15 em and a maximum depth of 12 em. Both the walls and bottom of the hearth were well oxidized, and indications of burning extended out another 20 cm from the pit edges. The fill sediments consisted of an ashy matrix with occasional flecks of charcoal. No artifacts were recovered from the fill. Feature 15-3 is an ephemeral hearth or firepit located about 1.1 m from the western wall. The feature was basin-Shaped in appearance, with a diameter of 50 em and a depth of 10 cm. The walls of the pit were lightly oxidized. The fill sediments consisted of an ashy matrix. A complete rna no and one piece of tabular rock material were the only artifacts recovered from the fill. Feature 15-4 is a floor pit located about 1.7 m southwest of the entryway. The feature was basin-Shaped in appearance, and had a diameter of roughly 50 em and an average depth of 15 em. The walls were lightly oxidized. Due to time constraints, only about three-quarters of the pit was excavated. The fill sediments The Deer Creek Drainage 177 consisted of a dark brown silty matrix with occasional pockets of ash and charcoal, and a low density of artifacts. Feature 15-5 is posthole that contained a mano fragment in the fill. The posthole measured 14 em in diameter and 11 em in depth. Feature 15-6 is a small basin-shaped pit located about 2.5 m west of the entryway. The diameter of the pit was about 33 em and the average depth was 10 em. The pit walls were very lightly oxidized, suggesting a minimal burning episode took place inside it. The fill sediments consisted of a dark brown silty matrix intermixed with ash and occasional flecks of charcoal. No artifacts were recovered from the fill. Feature 15-7 is a small ash pit located in the approximate center of the house. The pit had a diameter of 15 em and a maximum depth of 10 em. The pit walls showed no signs of burning. The fill sediments consisted of a fine-grained ashy matrix. No artifacts were recovered. Intrusive Features. At least three features intruded into Feature 15. First, it appears that the southern wall of Feature 9, a pithouse, destroyed part of the northern wall of Feature 15. Second, the eastern wall of Feature 15 was impacted slightly by the construction of Feature 20, a cobble alignment that may be associated with the use of Feature 14, a pithouse that partially overlies Feature 15. Because Feature 14 was built largely in the fill sediments of Feature 15, its construction does not appear to have significantly impacted the lower house. Abandonment and Postabandonment. Given the general lack of floor artifacts, it appears that the house was cleaned out at the time of abandonment. The structure may have remained standing for some time after its abandonment, based on the lensing of silts (Stratum 12) evident between the roof/Wall fall and floor levels. The lensing also is evident just above the roof/Wall fall level, which suggests that alluviation continued after the structure collapsed. There are no indications that burning was the cause of this collapse; rather, the structure appears to have decayed gradually over time. Given the low density of artifacts in the fill, it does not appear that significant amounts of trash were dumped into the depression created by the collapsed structure. Based on this evidence along with the stratigraphic evidence discussed above, it seems likely that Feature 15 was one of the first houses occupied at the site. MASONRY ROOMS Feature 5 and Feature 16 This feature was first recorded by archaeologists from the Arizona State Museum in 1972 (ASM site files). It was subsequently rerecorded during the State Route 87 survey by Archaeological Research Services, Inc. (Stone 1986) and during the Rye Creek testing phase (Elson and Swartz 1989a). A narrow test trench was hand-excavated through the feature during the testing phase. The remaining portion was excavated during the data recovery phase. As noted previously, for purposes of recording, the area inside the inner row of cobbles was designated Feature 5, whereas the area between the inner and outer rows was designated Feature 16 (see Figure 7.37). Data recovery began by hand excavating a 50-cm-wide test trench along the east-west axis close to the southern wall of the feature. All fill sediments from this trench were screened through Y4-inch mesh. Once the floor level (Stratum 20) in the test trench had been exposed, excavations expanded outward to expose the floor in other parts of the feature. A grab sample of artifacts was made from the Stratum 10 fill levels outside the test unit; however, all fill from the Stratum 19 level was screened through Y4-inch mesh. A composite flotation sample was collected from the Stratum 19/20 level, and a composite pollen sample was collected from the floor (Stratum 20). The remaining portion of the feature was then excavated in two recovery units; the first (Stratum 10) averaged 5 em in depth, the second (Stratum 11 roof/Wall fall) averaged 10 em in depth. Fill sediments from this portion were not screened, but grab collections were made. 178 Chapter 7 o ow 8' o ~ ::::. -~ ~"'@:-. ~; o D :: :".: .... .0. F 5-1 H~ Vme Ohs o Feature 5 AZ 0'15'53 o , {J OJ 50 • , It' CENTIMETERS o Wall rock - Feature 16 Upright wall rack - Feat ure 5 o Rack F 5-1 Feature number H Hearth Mortar ,;;~ Burned floor '.;': ': '. Burned plaster Past hole o Sherd • Lithic m Mana me Metate hi Hammerstone - e o o A Feature 16 8 o r-I- -----Feature 5 Pit AI -~ r---------------------- Feature 8 ~----- Feature 16 -----------------~ 5------- Figure 7.37. Masonry structures Features 5 and 16 at the Hilltop site. 8' The Deer Creek Drainage 179 Description. Features 5 and 16 make up a fairly unique, three-walled, dry-laid cobble masonry room (Figure 7.37). During the fieldwork this was believed to be a single structurally related feature, although separate collections were made from each component. It is now thought that the two features are unrelated, although this is still not entirely clear. Feature 16 contains a maximum interior dimension of approximately 6.0 m north-south by 4.0 m east-west. The walls form a rough semicircle and are made of unshaped quartzite river cobbles that range in size from 15 em to 40 em in diameter. The cobbles were resting on the ground surface; no fill was encountered when this surface was tested. Feature 5 is situated approximately 1.5 m inside of the outer walls of Feature 16. It consists of a single row of upright cobbles. These cobbles range in diameter from 10 em to 20 em, and they are embedded in the sterile substrate approximately 15 em below the level associated with the outer wall rocks of Feature 16. In this sense, Feature 5 is actually a masonry pitroom. Occasional patches of plaster were found on the walls (but not on the walls of Feature 16). Although no formal entryway was identified, the structure opens to the east. The fill sediments within Feature 5 consisted of an orange brown, silty matrix interspersed with small flecks of charcoal and burned daub. Artifact densities were low-to-moderate, averaging about 90 artifacts per cubic meter. Parts of the structure, especially near the northwest corner, showed signs of root disturbance caused by a nearby juniper tree. Portions of the floor were plastered, with the degree of preservation strongly correlated with the degree of burning. The best preserved portion was in the southwest corner of the inner structure, where patches of oxidized floor plaster were covered by a thin ash lens. The floor assemblage consisted of two flat-lying sherds, several pieces of lithic debitage, a hammers tone, and a metate fragment. The metate fragment covered a partially plastered hearth (Feature 5-1). Two postholes were identified, one near the western wall of the inner structure, the other just east of the where the masonry walls end. Internal Features. Feature 5-1 is a somewhat irregularly shaped, plastered hearth located along the open side of the feature. The hearth is 20 em long, 15 cm wide, and 20 em deep. No archaeomagnetic samples were recovered. Intrusive Features. No obvious features intruded into either Feature 5 or Feature 16. Abandonment and Postabandonment. Before the abandonment/postabandonment sequence of these two features can be determined, it is necessary to have some idea of how the features were related to each other both temporally and functionally. Did the inner structure (Feature 5) come first, or did the outer structure (Feature 16)? Or were the two features contemporaneous and thus part of the same structure? Unfortunately, the architectural evidence does not provide a definitive answer to these questions. The results of the contextual analysis (see Chapters 11 and 14, Volume 2), though, are highly suggestive in this regard. They indicate that the inner structure was probably occupied first, based on the fact that the lower fill levels (Stratum 19) contained secondary refuse deposits, whereas the upper fill levels (Stratum 10) contained primary refuse. From this it would appear that the inner structure was occupied, abandoned, and filled in with secondary trash, and that these trash deposits were then overlain by primary refuse associated with the use of Feature 16. Although no clear-cut activity surface could be identified with the primary refuse deposits, during the testing phase a "floor" was defined in the trench profile that turned out to be 15 em above the plastered floor; it may be that this upper "floor" was actually an activity surface associated with the later occupation. 180 Chapter 7 EXTRAMURAL FEATURES Thirteen extramural features were recorded and excavated during the data recovery phase, including five pits, one roasting pit, a rock alignment or wing-wall, and six possible crematoriums. Summary information on the pits and roasting pit is presented in Table 7.9, and Table 7.10 presents summary information on the possible crematoriums. Table 7.9. Metric data from pits and roasting pits at AZ 0:15:53. Feature Feature Percent Excavated Type Length (m) Width (m) Depth (MBO) Volume 2 Large pit 100 1.05 035 11.09-11.20 0.021 11 Small pit 100 0.41 0.38 11.12-11.20 0.007 12 Small pit 100 0.47 035 11.10-11.15 0.004 13 Roasting pit 100 1.00 0.87 10.87-11.11 0.109 18 Small pit 100 0.35 035 10.62-10.77 0.010 19 Small pit 100 0.63 0.60 10.62-10.73 0.022 MBD = Meters Below Datum Table 7.10. Metric data from possible crematoriums at AZ 0:15:53. Feature Percent Excavated Length (m) Width (m) Depth (MBO) Volume Comments Backhoe removed 1/3 of pit 4 100 0.75 035 1034-10.56 0.030 17 100 1.08 0.45 10.44-10.67 0.063 21 100 0.60 0.50 1038-10.61 0.036 Feature largely destroyed by backhoe 22 100 1.20 0.40 1030-10.40 0.025 Backhoe cut through center of pit 23 100 0.40 0.25 10.45-10.52 0.004 24 100 0.50 0.40 10.46-10.68 0.023 MBD = Meters Below Datum Backhoe removed 1/2 of pit The Deer Creek Drainage 181 Pits Feature 2. Feature 2 was first exposed in profile in Trench 2 during the testing phase. At that time, the feature was recorded as a "possible pithouse," but, further investigation during the data recovery phase revealed it to be an extramural pit. The pit was irregular in shape, with maximum dimensions of 1.0 m north-south by 0.3 m east-west, and an average depth of 10 em; it appears that the backhoe destroyed the western 40 percent or more of it. The feature was excavated as a single recovery unit. The fill sediments, which consisted of a gray brown silty matrix, were not screened but grab collections were made. Artifacts recovered from the pit include one plainware sherd and one piece of lithic debitage. Feature 11. Feature 11 is a small basin-shaped pit that was first exposed in plan view during the hand-stripping around Feature 2. The pit had a diameter of 40 em and an average depth of 8 em; it intruded 10 em into the northern wall of Feature 12. The walls of the pit showed slight indications of oxidization. The fill sediments consisted of a dark brown, silty matrix with numerous pieces of charcoal and pockets of ash interspersed throughout. Artifacts recovered from the pit include one plainware sherd and one piece of lithic debitage. Feature 12. Feature 12 is a small basin-shaped pit located due south of Feature 11. As noted above, Feature 11 intruded into the northern wall of Feature 12, indicating that it came later. Despite this slight disturbance, though, the size of Feature 12 can be estimated with some confidence; its maximum dimensions were 45 em east-west by 35 em north-south, and the average depth was 5 em. The fill sediments consisted of a loose, gray brown silty matrix. Although a few pieces of charcoal were noted, the amount was far less than was the case with Feature 11. Also, in contrast to Feature 11, the pit walls showed no signs of oxidation. Three plainware sherds and one piece of lithic debitage were the only artifacts recovered from the pit. Feature 18. Feature 18 is a small basin-shaped pit that was first exposed during hand-stripping east of Feature 9. The diameter of the pit is 35 cm, and it has an average depth of 15 cm. The fill sediments consisted of a light gray, ashy-silty matrix with a low density of artifacts (three pieces of lithic debitage). Feature 19. Feature 19 is a shallow basin-shaped pit located just outside the presumed entrance of Feature 9. The pit had a diameter of 65 cm and a maximum depth of about 15 em. The fill sediments consisted of a light gray silty matrix with a low density of artifacts (one piece of lithic debitage). Roasting Pit Feature 13. Feature 13 is a small rock-filled roasting pit located on the east side of the limestone knoll. The feature was first identified in plan view during the hand-stripping around Feature 2. Once it had been entirely exposed, it was excavated as a single recovery unit. No pit outline could be discerned, but most of the rocks were clustered tightly together over a roughly 1.0 m by 0.75 m area. The majority of rocks were fire-cracked. The fill sediments consisted of a fine-grained silty matrix with numerous charcoal inclusions. Artifacts recovered from the fill include one plainware sherd and three pieces of lithic debitage. Rock Alignment Feature 20. Feature 20 is a roughly 5-m-Iong rock alignment that was exposed during the excavation of Feature 15 (Figure 7.36). The feature consists of a loose arrangement of unshaped limestone cobbles, stacked two to three courses high in places. The cobbles varied in size from 20 em to 60 em in length and 5 to 20 em in thickness, and they ranged in elevation from 15 em to 35 em above the floor of Feature 15, and 2 to 15 em above the floor of Feature 14. The rock alignment curved slightly to the northwest, and there was a roughly 6O-em gap (an entryway?) near the middle of it. It is not altogether clear how Feature 20 is related both temporally and functionally to other features at the site. Although it clearly postdates the occupation of Feature 15, there are some indications that it may be 182 Chapter 7 associated with the use of Feature 14. Support for this idea comes from three lines of evidence. First, Feature 20 practically abuts the northeast corner of Feature 14. Second, most of the rocks from Feature 20 were found at about the same depth as the upper wall from Feature 14, which presumably corresponds to the prehistoric ground surface. Finally, the entryway of Feature 14 faces the open space that the rock alignment appears to bound. Although no extramural activity surface could be identified, this could easily be due to the nature of the fill sediments from Feature 15. Possible Crematoriums Six features that appear to be crematoriums were found clustered together near the northern edge of the ridge below the limestone knoll. With one exception (Feature 23), all of these features were oval-shaped pits dug into the calcic horizon to a depth of between 20 em and 30 em (Figure 7.38 and Table 7.10). Although the evidence for burning within most of these pits was minimal, all of them had at least several surfaces where scorching was evident. The identification of these features as crematoriums is based on the presence of similar, although much more clearly defined, features at AZ 0:15:52, located less than 200 m to the southwest, and the fact that small amounts of cremated bone were typically present. Feature 4. Feature 4 is an oval-shaped pit first exposed in profile in Trench 11. Although the backhoe destroyed the southern third of the pit, the remaining portion measured 0.75 m east-west by 0.35 m northsouth. The pit cuts through the calcic horizon to an average depth of 22 em. The walls and bottom of the pit were burned, and the fill sediments were a light brown silty matrix with small flecks of charcoal and caliche. Artifact densities were generally low, but a single piece of cremated human bone was recovered from the lower fill deposits. Feature 17. Feature 17 is the largest of the possible crematoriums. It measured 1.08 m east-west by 0.45 m north-south, with a maximum depth of 25 em. The walls and floor of the pit showed signs of moderate-toextensive burning. The fill sediments consisted of a light brown silty matrix with flecks of charcoal and caliche interspersed throughout. A moderate amount of root disturbance also was evident. Artifact densities were low except for sherds from two plainware vessels that were recovered from near the center of the pit. A single piece of cremated bone was found about 20 em south of the pit. Feature 21. Feature 21 is a pit that was largely destroyed by the backhoe. The portion that remained intact measured 50 em east-west by 60 em north-south and it cut through the calcic horizon to an average depth of 20 cm. The pit walls were burned in places. The fill sediments consisted of a light brown silty matrix with occasional flecks of charcoal and caliche and pieces of burned daub. Artifact densities were generally low, but a large sherd was found resting on the southern wall of the pit. Two small quartzite cobbles were also noted in the fill, along with a small amount of cremated human bone. The degree of root disturbance was moderate-to-high. Feature 22. Feature 22 is an oval-shaped pit that extends about 30 em on either side of the backhoe trench. Assuming the backhoe removed the center of the pit, the maximum dimensions were 1.2 m east-west by 0.4 m north-south, with a depth of roughly 10 cm. Only slight evidence of burning was noted on the walls of the pit. The fill sediments consisted of a light brown silty matrix. No artifacts were observed in the fill. Feature 23. Feature 23 is not so much a formal feature as it is a scorched area directly overlying the calcic horizon. The scorched area measured 4S em east-west by 25 cm north-south. No artifacts were recovered. Feature 24. Feature 24 is a pit, the eastern half of which was destroyed by the backhoe. The remaining portion measured 40 cm east-west by 50 cm north-south, with an average depth of 10 em. The feature had been extensively disturbed by roots. The fill sediments consisted of a light brown silty matrix with occasional flecks of charcoal and caliche. Although the artifact density was low, several pieces of cremated human bone were recovered. The Deer Creek Drainage 183 AZ 0: 15: 53 t o , 50 , , ! , , CENT I METERS N F4 o (0) o Feature number Rock Burned sod Sherd ........ ........ N 21 8 +E 84 -c::: N 218 +E86 ~ ~ F 23 c::: Q.) " h.. F 22 Figure 7.38. Crematorium area at the Hilltop site. ARTIFACT SUMMARIES Artifact totals are summarized by feature and stratum in Table 7.11. Note that these totals reflect only the major artifact classes. Also note that only artifacts in clear association with a particular feature are included here; thus. artifacts from surface contexts (Stratum 0). mixed contexts (Stratum X). or disturbed contexts (Stratum SO) are not included. 184 Chapter 7 Table 7.11. AZ 0:15:53 artifact totals by feature and stratum. Fea. Str. 1 09 1 10 1 19 1 20 Pln 108 79 Red Buff B/W 7 1 4 1 1 3 2 2 09 50 40 4 4 09 50 1 3 5 5 5 09 10 19 o o o o o o o o 4 1 4 3 6 55 6 6 6 09 10 19 20 30 56 54 1 o o o o o o o o o o o o o o o o o o o 8 8 09 10 63 44 2 2 2 9 48 39 24 18 3 9 9 9 09 10 19 20 30 11 50 12 50 13 50 14 14 10 19 13 14 15 15 15 09 10 19 13 12 16 11 17 50 18 50 19 50 6 9 Total 42 26 1 o 1 o o o o o o o o o o o o o 1 o o o o o o o o o o o o o o 764 24 11 o 3 o 1 o o o o o o o 1 o o o o o o o o o o o o o o 1 o o o o o o o o o o o o o 2 o o o o 5 Oth Dec 1 o o o o o o o Flk Debit. 11 14 22 1 44 1 Tool Core Tool 2 o o 2 1 5 3 1 4 o o o 2 4 2 1 o o o o o 2 1 1 o o o o o o o o o 1 o o o o o o o o 5 o o o o o o o o 1 1 o o 10 2 o 4 o o o o 3 o o o o o o 2 504 25 22 32 1 o o o o o o o o o o o o o o o o o o o o o o o 10 o 8 44 29 19 39 12 o o 60 32 56 15 32 7 1 2 8 5 o 5 11 o o o o Oth Mano o 1 o 1 4 o 1 2 2 1 o o o o o o 1 3 o 2 1 1 o o o o 9 o o o o o o 2 Met. GS 1 8 6 o o o o o o o o 1 o o 1 o o o o o o o o 2 o o o o o o o 1 o o o o o 6 o 2 6 o 1 o o o o o o 1 o o 1 2 1 o o 3 o o o o o o o 2 o o 36 Shell o o o o o o o o o o o o o o o o o o o o o o o o o o o o o 1 o o o o o Total 132 102 81 42 97 2 12 3 9 54 36 76 100 68 2 3 133 82 109 56 60 45 1 2 4 3 22 20 16 19 15 18 3 4 1432 The site produced a total of 1,432 artifacts from 16 primary features. Ceramics account for 56.3 percent of this total, chipped stone for 38.5 percent, ground stone for 5.2 percent, and shell for 0.1 percent. Nearly 95 percent of the ceramics recovered from the site were plainwares, 3 percent were redwares, and the remaining 2 percent were decorated wares. Hohokam buffwares were the most common decorated ware found in clear association with features; they account for 61.1 percent of the sample shown in Table 7.11. It bears mentioning, though, that buffwares were far less common in nonfeature contexts; in fact, they account for only about 30 percent of the decorated assemblage from the site as a whole (see Chapter 12, Volume 2). A similar pattern holds true for the whitewares, except in reverse. As a group they account for 27.8 percent of the The Deer Creek Drainage 185 decorated wares from features (Table 7.11), but for the site as a whole they make up almost 63 percent of the assemblage. This discrepancy is perhaps most easily explained as the result of visibility factors (whitewares tend to be more visible than buffwares in surface contexts), but other unknown factors may also be involved. Other decorated ceramics recovered from the site include a Show Low Black-on-red sherd from the upper fill deposits of Feature 5, a Gila Polychrome sherd from the upper fill of Feature 1, and an indeterminate San Juan redware sherd from the surface. Debitage accounts for 91.5 percent of the chipped-stone assemblage, flake tools account for 4.5 percent, and core tools account for 4.0 percent. Manos outnumber metates by a ratio of 5.3:1. The miscellaneous groundstone assemblage consists mainly of polishing stones and tabular knife fragments. The single piece of shell recovered is a small fragment of Laevicardium from the fill of Feature 15. POLLEN AND FLOTATION SUMMARIES Pollen Data Four samples were analyzed from this site: three from pithouses (Features 1, 6, and 9) and one from a crematorium (Feature 4). There was insufficient pollen to characterize the crematorium. The remaining samples included 24 different taxa. Nine taxa were found in all three pithouses. They included Artemisia (sagebrush),Ambrosia-type (ragweed and related species), High Spine Compositae (sunflower family), Chenoam, Gramineae (grasses), Eriogonum (wild buckwheat), Pinus (pine), Quercus (oak), and Zea (com). Of the economic taxa identified, Zea (com) was the only one found in all three houses. Cruciferae (mustard family) was found in two pithouses (Features 1 and 6). The only other economic type, Cylindropuntia (cholla), was found in Feature 1. Flotation Data Two samples, totaling 8 liters of sediment, were analyzed from this site. One of these samples was from the lower fill-floor levels of Feature 1, a pithouse; the other sample was from the lower fill-floor levels of Feature 5, a masonry pit room. The pithouse is tentatively dated to the Sacaton phase, the masonry pit room is dated to the early Classic period. Agave accounts for 54.5 percent of the 2.75 relative plant parts recovered from Feature 1. Cheno-ams were the next most common taxa, accounting for 18.2 percent of the sample, followed by Gramineae (grasses), Descurainia (tansy mustard), and Hordeum (barley) at 9.1 percent each. Of note, no Zea mays (com) remains were recovered. The sample from Feature 5 produced 2.75 relative parts, all of which were Cheno-am seeds. SITE CHRONOLOGY No archaeomagnetic dates were recovered from the site. The diagnostic decorated ceramic assemblage includes only 10 temporally sensitive sherds, which were scattered across the site with no apparent spatial patterning (see Chapter 12, Volume 2). The decorated assemblage is composed primarily of Sacaton Red-on-buff (AD. 950-1150)(n=4). This, along with a single Black Mesa Black-on-white sherd (AD.1ooo-1135) (and another Black Mesa sherd recovered during the testing phase), suggests the primary occupation of the site was sometime between AD. 1000-1150. In addition, two Kiatuthlanna Black-on-white (AD. 850-950) sherds and a Santa Cruz Red-on-buff sherd (AD. 850-950) suggest an earlier use of the site area. A Deadman's Black-onred sherd (AD. 800-1000) and two Kana-a Black-on-white sherds (AD. 825-1000) recovered from the surface of the site during the testing phase could be associated with either occupation. 186 Chapter 7 Excavations within and around Feature 5, the masonry structure, recovered a Show Low Black-on-red (AD. 1030-12(0) sherd from the overburden above the structure, as well as several redwares from the structure floor. This, in conjunction with the masonry architecture, suggests a possible early Classic period occupation (AD. 1150-13(0). Other Classic period ceramics recovered from the general site area (but not within the Feature 5 area) include: a single Gila Polychrome (AD. 1250-1400) sherd recovered during the mitigation phase, and single Tusayan Corrugated (AD. 1000-1300), Tonto Corrugated (AD. 1200-1400), and Salado Red Corrugated (AD. 1200-1400) sherds recovered during the testing phase. These sherds provide additional documentation for Classic period use of the site area, although their association with Feature 5 is unclear. SITE INTERPRETATION The Hilltop site appears to have been a either a fieldhouse or farmstead site that was used on a seasonal basis over a relatively long period of time. Although no com remains were recovered from the flotation samples, the recovery of Zea pollen suggests an agriculturally based subsistence strategy. Further support for this idea comes from the fact that the site is situated directly above a large expanse of arable land near the confluence of Deer and Rye creeks. The five pithouses and numerous extramural features located on top of the ridge are thought to be associated with a Preclassic occupation; the masonry features at the base of the ridge are associated with a Classic period occupation. Based on stratigraphic and depositional evidence, it seems highly unlikely that more than one or two of the pithouses were occupied at anyone point in time. A tentative sequence of construction and occupation is as follows: Features 1 and 15, the two largest houses, appear to have been built and lived in first, followed by Features 6 and 9. Feature 14 appears to have been the last pithouse in use; it may have been occupied in association with Feature 20, an enigmatic rock alignment. The exact relationship of the other extramural features (including the possible crematoriums) to this sequence of pit house occupation is uncertain. The Preclassic occupation may be related to similar events occurring at the Deer Creek site (AZ 0:15:52), situated less than 200 m to the south, which also appears to have a small seasonal occupation during the late Santa Cruz and Sacaton phases. The Classic period occupation of the site appears to have been fairly brief and of low intensity. Feature 5, the three-walled inner structure, was probably used first, based on the results of the contextual analysis. Feature 16, the three-walled outer structure, appears to have then been used as a foundation for a wind break or brush structure. Evidence for primary lithic core reduction was found associated with the later structure. CHAPTER 8 THE CLOVER WASH DRAINAGE Clover Wash is the smallest named drainage within the project area, and is actually a tributary of Rye Creek. It is currently classified as an intermittent stream, flowing seasonally or during periods of high precipitation, although it is much smaller than Rye, Deer, or Hardt creeks, which are also classified as intermittent streams. It is unknown whether Clover Wash contained a significant flow during the prehistoric period. The majority of sites along this wash are actually situated on the ridge between Clover Wash and Rye Creek, or between Clover Wash and Deer Creek, and could have utilized these more major drainages for water and other resources. Two data recovery phase sites are included within this drainage area: the Clover Wash site (AZ 0:15:1(0) and the Redstone site (AZ 0 :15:91). Both of these sites are small, Preclassic period farmsteads. An additional three sites (AZ 0:15:93, AZ 0:15:94, and AZ 0:15:95) within the drainage area were investigated during the testing phase and are described in Chapter 10. A map of the Clover Wash drainage area is presented in Figure 8.1. . .-:~:. 188 Chapter 8 Figure 8.1. Location of sites within the Clover Wash drainage area. THE CLOVER WASH SITE AZ 0:15:100 (ASM) [AR-03-12-06-704 (TNF)] Deborah L. Swartz Site AZ 0:15:100 is a small farmstead situated on the north bank of Clover Wash (Figures 1.3 and 8.1). The surface artifact scatter measures 50 m north-south by 80 m east-west (4000 square meters) and straddles the proposed right-of-way. An area of approximately 1,125 square meters lies outside of the right-of-way to the east (Elson and Swartz 1989a:Figure 4.19). This area was recorded and mapped but no collections or subsurface testing was undertaken. The site area within the right-of-way contained 27 cultural features. Five pithouses, 11 extramural pits, 2 roasting pits, a rock cluster, an inhumation, and a large unknown extramural feature were sampled. Six additional extramural pits were recorded but not excavated. Given the distribution of surface artifacts and subsurface features, it is believed that the majority of cultural features were within the right-of-way (Figure 8.2). The site is situated on the Mazatzal pediment at an elevation of 3,090 feet above sea level. It lies approximately 1 km west of the confluence of Clover Wash and Rye Creek. This juncture is the location of Rye Creek Ruin (AZ 0:15:1), the largest site in the area (Figure 1.3). Root-plowing in the 1960s disturbed the soil to a depth of 40 em below the surface of the site, which produced a fairly open vegetation cover of small acacia, small mesquite trees, and grasses. Excavations during the data recovery phase yielded 4,138 artifacts. These consist of plainware, redware, and decorated ceramics, lithic tools and debitage, ground stone, shell, and animal bone. The decorated ceramics date from AD. 750 to 1200 with a likelihood of two components. The early component dates from AD. 750 to 850 and the later component dates from AD. 1000 to 1100. EXCAVATION METHODS The site was originally recorded by the Forest Service and reevaluated by Stone (1986:30). He described it as a low-density artifact scatter with extensive surface disturbance. Testing Phase The testing assessment found that the surface artifact scatter was denser and more diverse than originally recorded by Stone. A total surface collection was undertaken by dividing the portion of the site within the right-of-way into four quarters. Each quarter was totally collected as a unit. Due to the extensive surface disturbance caused by root-plowing this was considered to be sufficient for horizontal control. The surface collection resulted in the recovery of 656 artifacts, which calculates to a surface artifact density of 0.3 artifacts per square meters. Decorated ceramics included Gila Butte Red-on-buff and Black Mesa Blackon-white. The site area was then divided into 20-m grid units to facilitate laying in the backhoe trenches. Five northsouth backhoe trenches were excavated at lO-m intervals across the portion of the site within the right-of-way. A total of 264 linear meters of trench were excavated extending over an area 60 m north-south by 40 m eastwest. Eight cultural features were recorded. They were identified as four pithouses, one possible pithouse, two pits, and a possible pit. These features were below the root-plowed zone, although disturbance was noted in their upper levels. A vertical datum was established at N155 E100 to record the features (Elson and Swartz 1989a). 190 Chapter 8 AZ 0: 15: 100 (ASM) N200 NI80 NI60 Datum o!. ,/ ....- -.. F. F.G NI40 10 >. / o ,- I lE ..... ~ ............... ? / ~ \ ...... ..... , 01 "\ " ~ \ NI20 \ \ I F.7 I I I I I T7 NIOO - - - N 80 T9 --- - T3 T2 T8 T4 I T5 I / / / ---.." ./ Si1e boundary within riQht-ol-way Site boundary ouilide riQht-ol-way o 10 ! Backhoe trenche, (TI-91 1-28 / t meters Feature, o to W N o o o N o v w W w Figure 8.2. Overall site map of the Clover Wash site (AZ 0:15:100 [ASM]). o \0 W The Clover Wash Sites 191 Data Recovery Phase The first procedure during the data recovery phase was to excavate additional backhoe trenches at five-meter intervals between the testing trenches. Four 9O-m-long trenches were excavated. They were extended to the north, beyond the original artifact-scatter boundary, to confirm the northern site boundary (Figure 8.2). No features were recorded beyond the limits of the features recorded during the testing phase. Four additional features were recorded in the new trenches. They were identified as one pithouse and three small pits. Once the features were recorded and the site area defined, the entire area (1,225 square meters) was mechanically stripped to remove the root-plowed zone. This revealed outlines of four features that were not cut by the backhoe trenches and outlines of the previously recorded features. These new features include one small pit, two roasting pits, and a rock cluster. Pithouses were generally excavated by initially placing a 2-m by 2-m control unit in the center of the exposed outline. These units were excavated to the pithouse floor in arbitrary 2O-cm levels or smaller natural layers to collect a controlled sample of the fill. Of the five pithouses, three were sampled and two were completely excavated. The sampled houses were not overly trash-filled and did not contain floor assemblages; their size and orientation were determined from the outline of the cultural stain. Complete excavation was undertaken at the other two houses since their orientation was not apparent from the stripped surface. The pithouse fill (Stratum 10) outside of the 2-m by 2-m control unit was hand-stripped without screening. A grab sample collection of artifacts was recovered. At 5 em above the floor the fill was screened (Stratum 19). This method quickly exposed the plan and orientation of the structure while also recovering any floor assemblage (Stratum 20). In all sampled houses the hearth was exposed and archaeomagnetic samples were recovered. Extramural pits and roasting pits were generally sampled by bisecting the feature and excavating half. Several of the small or unclear pits were totally excavated. A rock feature of unknown function also was excavated completely, as was the single inhumation. Summary Twenty-seven features were recorded at the site (Figure 8.3). Five were pithouses, of which two were totally excavated and three were sampled. The majority of the features were extramural pits. Seventeen pits were recorded; 6 were sampled, 5 were totally excavated, and 6 were not sampled. The single inhumation was totally excavated, as was the rock cluster. The two roasting pits and the large unknown extramural feature were sampled. The area around all of the features was mechanically stripped prior to any hand excavation. This was done to remove the root-plowed disturbance. The stripped area measures 35 m by 35 m for a total area of 1,225 square meters. No hand-stripping was undertaken at this site because of the root-plowing. The data recovery phase lasted from July 6, 1989, to August 2, 1989 expending approximately 166 person-days. PITHOUSES Metric data on the characteristics of the five excavated pithouses are presented in Table 8.1. 192 Chapter 8 ,-------------I AZ 0: 15: 100 (ASM) I I NI40 -0 I . K ID $) F.28 W ~ .~'4 F. 13<) NI30 'aJ.>' Roostino pit I'I \ I I F. 15 Rock clu ster \ I i NI20 I F. I~ ~ II F. 7 1_ _ _ _ _ _ _ _ NIIO ~~ @ Edge of str i pped Probabl. tdCJe Unexcavoted ar.as or~ _ _ _ _ _ _ _ _ _ _1 , 0 5 I meters Pit · Postholes 0 I I I I I I I N Hearth 0 0 0 0 C\I 0 to W ~ W W Figure 8.3. Area of data recovery at the Clover Wash site (AZ 0:15:100 [ASM]). The Clover Wash Sites 193 Feature 1 The feature was originally identified during the testing phase. Mechanical stripping of the site prior to excavation revealed an ill-defined dark stain. A 2-m by 2-m control unit excavated on the east side of the backhoe trench was disturbed by an intrusive pit, so a second 2-m by 2-m unit was excavated on the west side of the trench to obtain an undisturbed sample of the pithouse fill. The orientation and outline of the feature were unclear so the whole feature was excavated. A second pit intrudes on the northwest wall of the house. Description. Feature 1 is a clearly defined rectangular pithouse with rounded comers and a large ramped entrance (Figures 8.4 and 8.5). The entrance opens to the southwest with the long axis of the house perpendicular to it. The floor area measures 16.7 square meters and the entrance measures 3.8 square meters. The house was excavated into the red sterile substratum to a depth of 25 em below the root-plowing disturbance. The fill of the pithouse consisted of fine-grained, dark organic sandy silt with a moderate artifact density. Ceramic density within the fill averaged approximately 150 sherds per cubic meter. The lithic debitage density averaged approximately 43 lithics per cubic meter. Directly above the floor was a thin band of black, burned roof fall containing abundant charcoal and some oxidized daub. Four burned beams were recovered from this layer. The floor was not prepared and was identified by the contrast between the dark cultural fill and the red sterile substratum. Table 8.1. Metric data from pithouses at AZ 0:15:100. 8 ...-; of! ...... fil ~5: e .. a.8 11 ~ p..Z -] .. !: ~~ -a.. ~ ..,of! 0 ~~ Ii t'I ..,S s s ~ 'E .- ... ~ .8 .., s !j '~ a :s;:3~a tIl~ .. . . ~.. ~ .. ... 'il... 8 .. ~. .ss (§ 8 Ii: ~ 1I 0 ',c eo .!! "i>. "i>. :I: ~ 100 4.9 3.4 16.66 10.55-10.90 5.8 197 44 2250 + 3 82 6.9 3.7 25.53 10.92-11.25 8.4 142 19 9~ + + - 4 60 6.0 4.4 26.40 11.02-11.36 9.0 62 17 900 + + - 0 6 '1 3.7+ 3.2+ 11.84+ 10.54-10.58 0.5 356 62 105 + 12 45 6.8 5.1 34.68 10.98-11.49 17.7 100 17 2800 + 2 Remnant The floor assemblage consisted of three mano fragments, a metate fragment, and a large piece of ground argillite. The only floor features were a plastered hearth (Feature 1-1) and three small floor pits that were not excavated. No postholes were encountered. Internal Features. Feature 1-1 is a poorly plastered hearth situated 80 em back from the entrance. There is evidence for at least two remodeling episodes; the two remodeled interior hearths are within the boundaries of the larger original hearth. No visible plaster remains on the original hearth. The second hearth lies within the first one but has only fragmentary plaster. The last hearth is within the second and also has patchy plaster. The total extent of the hearths measures 36 cm in diameter and 13 cm deep. An archaeomagnetic sample was recovered. 194 Chapter 8 AZ 0:15:100 (ASM) Feature I NI40 ,< ~ ... .-t't I- "I"-r- }-~ . f- ,. ~,~~ .\ ,.. Feature 27 ~ -i \\- 0 -\ ... ~ a J: u Z a ILl a: f- a a ~ \<'~j1l!,4 O'","d~~!." . ' ) ...;r\) MFO a a 04-Mano 0 NI3S 1 NI34 so o '--------' cm - - - - Probable edoe • Posthole ~~ ~ N ... w Intrullve pit ~ Unekcovoted ~Sherd @ Meta!e 00 ;;:; -I -i -I I- a D~Monos ... t- 0 ~ A ~ .r NI36 ~~ I- 0 NI38 NI37 }-. ~ ..c...a...L..l-' a ' ~ ~ (~~J NI 39 I- 6 fraomen! Rocks '" ILl PROFILE Trench 4 Figure 8.4. Feature 1 at the Clover Wash site. floor features ... ILl The Clover Wash Sites 195 Figure 8.5. Photograph of Feature 1 at the Clover Wash site. The three unexcavated small floor pits do not have feature numbers but range in diameter from 25 em to 40 em. Their function and depth are unknown. Intrusive Features. Feature 27 is a large pit that intrudes on the northeast portion of the house. Feature 28 is a small intrusive pit that cuts through the northwest wall of Feature 1. Abandonment and Postabandonment. The continuous layer of charcoal and dark burned fill lying directly on the floor suggests that the house burned shortly after abandonment. The floor assemblage is mostly ground stone and may represent articles that were intentionally left behind. The burning, then, would have been unintentional after the house was abandoned. The two episodes of remodeling on the hearth support the idea that the house was used over several years. After the house burned the feature was moderately used as an area for the dumping of secondary trash, as determined through the contextual analysis. The two later intrusive pits indicate that occupation of the site continued after the abandonment of this feature. Feature 3 Feature 3 was first identified during the testing phase. Mechanical stripping of the site prior to excavation revealed a large dark stain with no clear pithouse outline. Because of the low artifact density in the fill, two 2-m by 2-m control units were excavated. Approximately 82 percent of the house was excavated. An area 2 196 Chapter 8 m by 3 m was not excavated due to the low artifact density in the fill and the lack of a floor assemblage. The southern two meters of pithouse fill were excavated to floor without screening. The remainder of the house was excavated to 5 em above the floor without screening and the bottom 5 em (Stratum 19) were screened. Several intrusive pits obscured the surface outline. Description. This is a large rectangular pithouse with rounded comers and a large ramped entry oriented toward the east (Figure 8.6). The feature was excavated into the red substratum 22 em below the level of the root-plowing disturbance. The floor covers an area of 25.5 square meters and the large entryway adds another 7.8 square meters. The fill of Feature 3 was a gray brown, silty sand containing a moderate number of artifacts overlying a layer of burned roof fall. The roof fall contained large chunks of burned daub and charcoal in a compact gray brown, silty loam. The fill contained an average of approximately 92 sherds and 15 lithics per cubic meter. The north, south, and west walls of the house were defined by the distinction between the dark burned roof fall and a lighter alluvial fill. The lighter fill was determined to be the fill of Feature 16, a large pit. The floor, which was directly beneath the roof fall, is distinguished by the compact red sterile substratum, except between the hearths where there was a small area of plaster. The sterile substratum indicates that the pithouse cut entirely through the pit (Feature 16) except on the north side of the house where the floor is built on the fill of the pit, making it difficult to define. The floor of the entryway was built on the white calcic horizon. The floor assemblage consisted of five manos, one hammerstone, and one core from the entryway area, and a piece of ground argillite, a hammers tone, and a core from the pithouse area. There are seven small postholes around the periphery of the floor. Several contained in situ burned post remnants, and one large central posthole to the south side of the hearth also contained a post. A corresponding large posthole in the northern section of the house may remain under the unexcavated portion. Seven small postholes are also scattered throughout the pithouse suggesting they were used as roof supports. The only other floor features are two well plastered hearths (Features 3-1 and 3-2) situated in front of the entrance and a large floor pit (Feature 3-3). Internal Features. Feature 3-1 is a partially plastered basin-shaped hearth which measures 30 em by 26 em and is 16 em deep. The fill contained a loose dark brown, silty loam with charcoal inclusions. Cracked plaster lined the southern half of the hearth, extending halfway down the basin. The remainder of the hearth was unplastered. This is probably the earlier of the two hearths. Feature 3-2 is a well-plastered hearth situated directly in front of the entryway. It probably is the later of the two hearths as suggested by its complete nature. There is a well plastered collar enCircling the entire hearth and extending out from it apprOximately 15 em. The hearth measures 24 em in diameter and is 16 em deep. It was lined with continuous plaster across the entire basin. The fill was a gray brown, silty loam with charcoal inclusions at the top and a white ash layer covering the bottom 11 cm. An archaeomagnetic sample was recovered. Feature 3-3 is a large floor pit situated in the northwest comer of the pithouse. It measures approximately 60 em in diameter and 24 em deep but has been extensively disturbed by rodent activity. The fill contained a gray brown, silty loam with abundant large pieces of charcoal and several artifacts. Intrusive Features. Four pits without feature numbers and an infant burial (Feature 25) intrude on the walls of this house. Feature 3 intrudes on Feature 16, an extremely large pit that nearly surrounds Feature 3. The pithouse walls are cut into a light-brown alluvial fill that was determined to be the fill of Feature 16, however, the floor across much of the house is built on the sterile substratum meaning that the pithouse cuts all the way through the fill of the pit. The only exception is on the north side of the house where the floor is built on the pit fill. The Clover Wash Siles 197 AZ 0: 15: 100 lASM) Feature 3 ---- ... NI32 , NI31 J " -r -r ..,... NI30 ~-r, / -.-.-.- . -.-~ 1/ Fea.25 J y. Edge of "good" X Y : floor '(~ ;vY • .. . ' . ,.,""T'T'.,. . . . i "-Argillite (!)b?-M NI29 A- 0 ~(J-O r. I A Y I- ;J f- Feo. '" NI 25 r ~ 20 \'( y Y I Feo. 4-2 )oj< o Hommerstanes o NI 24 I a Sherd J: U Z lJ.J a: I- ,..., J: N I 23 u z lJ.J A' ~ Burned beam a: I- NI22 - - - Probable edge Extrapolated edge i NI 2 1 • o Posthole Pit ~~"; Intrusive pit N ~~ • •:i.:•....•::j., ...:!. :' . .,. . •. :i~. ;i:~., :.~ ; :'.Ij~ i. ;:I.> :;:r.~:. :~ :.~ i: ·.:~.:>~·.: ~'.: i '>,:~.!;:~. :.:m;.:. :i.: .::.';:;. :.'::';. . :f:. :. :i.i. :i::i:.:: .. :;I,::;t.:i .• .. :I:.::... :....':'.. .•.. :.•.i':.::.'::'i: ..I ..:';:.;.::'.:'i... ::: .. ..:':.'.:':.::':;:I:; .. .. :i.. :I,.:'. i:::::i. :!i.:..::I.:..,..... :.:. :·..i. . ::!.-..:i ... :;:>:•::.. . :•,.:. ::> .. .. : :. :,::: i.;: : : : · . : : ..; :·.;' .... ' ...•..... :!.:i. :i. !.:'.••..•.•. :i: ... ;..• :l.: •:. :I,::I •.. •. . OOt .. _. •. .i.:.•_:i:•... :J:I::.:!.: •... . ij..;.,.:.;:i;...:.':•. ::;>•. . . ..: .: ;•:. ....•.:;.>r .. . . .. ::...•. ~~ 230 Chapter 9 ~ o 30 60 I Figure 9.1. Location of sites within the Rye Creek drainage area. THE ROOTED SITE AZ 0:15:92 (ASM) [AR-03-12-06-1111 (TNF)] Deborah L. Swartz AZ 0:15:92 is a large agricultural and possible habitation site situated across a small unnamed wash to the south of the Cobble site (AZ 0:15:54), the largest pueblo site in the project area (Figure 1.3 and 9.1). The majority of the site was extremely disturbed by root-plowing in the 1960s (see Figure 2.2). The undisturbed features are all at the base of a cobble terrace or in the present right-of-way for State Route 87. Neither of these areas were root-plowed, although road- construction disturbance is evident within the right-of-way. Ten checkdams and a cobble concentration that was probably a small masonry pueblo were recorded in the rootplowed area east of the right-of-way fence line. The area within the present right-of-way west of the fence line contained a pithouse, a ramada, two inhumations, and an additional checkdam. All of the features, except for six of the checkdams, are within the proposed right-of-way. The overall site area, approximately 20,762 square meters, was divided into two loci (Figure 9.2). Locus A, the larger, measures 110 m by 180 m (19,800 square meters). It contains all of the intact cultural features. Because the upper 40 em of this locus were extremely disturbed by root-plowing, Locus A includes a large area with a moderately dense artifact scatter but no subsurface features. Because subsurface features are present in areas of lower surface density that were not root-plowed, it is probable that any subsurface features within this area were destroyed by the root-plowing. Judging from the relatively high density and diversity of the surface artifact assemblage, it is likely that pithouses were present, and the site may have contained a fairly substantial Preclassic period pithouse village or hamlet. Unfortunately, there is no way to assess the extent or nature of this settlement. Locus B is approximately 100 m to the south and is connected to Locus A by a very low-density artifact scatter. It measures 35 m in diameter (962 square meters) and it too has been root-plowed. Within Locus B there is a cobble concentration that at one time probably was a one-to three-room pueblo but because of the rootplowing disturbance, surface room outlines are no longer visible. Hand-trenching during the testing phase indicated that no subsurface cultural material remained intact. The proposed right-of-way runs across the eastern side of Locus A and bisects Locus B. The site is situated along both the second terrace of Rye Creek and the Mazatzal Pediment at an elevation of 3,040 feet above sea level. The unnamed wash that runs along the north side of the site joins Rye Creek approximately 0.5 km to the east. Rye Creek Ruin (AZ 0:15:1), the largest site in the area, is approximately 1 km to the southeast along Rye Creek (Figure 1.3). The root-plowing during the 1960s produced a moderately dense vegetation cover of mesquite, acacia, prickly pear, and grasses. The data recovery phase produced 1,721 artifacts. These include plainware, a single redware, and decorated ceramics, lithic debitage and tools, ground stone, shell, and animal bone, including a bone awl. The decorated ceramics have a wide temporal range from AD. 800 to 1325. The majority of the site probably dates to between A D. 950 and 1150 with possibly an earlier component dating from AD. 850 to 950. Diagnostic ceramics recovered from the cobble structure suggest that this feature dates to the early Classic period between AD. 1000 and 1325, and possibly to the period between AD. 1200 and 1325. EXCAVATION METHODS The site was originally recorded by Archaeological Research Services, Inc. (Stone 1986) as a moderately dense artifact scatter associated with eight checkdams. Stone recorded three checkdams within the right-of-way and 232 Chapter 9 ·;..::::,:·t , -----------, X ••• N 16 0 / I I~ '"'. " ' iTI !!! ~ Xi i=1-1- 1 ~ I- F,I,3 I ~F I~ N 12 0 \ " ...... ""- FI4PithouseCi= ==========:J'T2 (!2 F?S 40 '_.". _." " U"l ~ NI ~., . - ........... ' - •• _.. ,... -.--.._--- ... Locus A T3 .... -- ----" /,,,,\ FI5 Ramada? , 1=== = = = = = = ==:======== fF5 ......... ..... ' .... \ I I / " Art i fact Scatter Bou ndary Outs i de Ri ght ·o f ·Way I I I fj J N I OO I I I I I F6 F7 I .-/ "_" / . / N 80 ____ ;;V t.. F8 " 'l~;f~"/~ " ~==~I N60 AZ 0: I 5 : 92 (ASM) ,.-...... Inhumation 0 N40 . - Extent of Di sturbed Masonry Rubble I Chick Dam Locus Boundary Wi th i n Ri Qht ' of - Way N20 = Backhoe Trench (T I -17) = NO Hand Trench FI-17 ~ Features Center L i ne Stat i on 345 "*--+ 520 Fence Wash -" .-' ".,- TI2 (Til 540 ~~ \ 560 FI r n\ '.!!/ o, 20 i Meters , \ \ I I Locus B I '\:-' "",Artifact Scatter Boundary _ _ ... ,; Outside Right·of·Way r') 580 o (Xl UJ o .,.o o o o o N (J:) (Xl N .,.o o(J:) N N UJ UJ UJ UJ UJ N UJ UJ UJ N UJ o o o Figure 9.2. Overall site map of the Rooted site (AZ 0:15:92 [ASM). The Rye Creek Drainage 233 five more outside of it. The cobble concentration in Locus B was not recorded until the testing phase (Elson and Swartz 1989a). Testing Phase During the testing phase, in addition to the cobble structure that was disturbed by root-plowing, two unrecorded checkdarns were identified, one within and one outside of the right-of-way. Also, the associated artifact scatter was determined to be larger than Stone had recorded. The testing phase began by establishing a grid at 20-m intervals across the portion of the artifact scatter that lies within the right-of-way. A systematic sample of the surface artifacts were collected from 20 grid units (20 m by 20 m) in Locus A The sample consisted of 40 percent of the ceramics and 10 percent of the lithics. Because of the small size of Locus B, the entire portion of the locus within the right-of-way was collected as a single unit. All ceramics and lithic tools were recovered. In Locus A, 10 east-west backhoe trenches were excavated at 10-m or 1S-m intervals across the portion of the locus within the right-of-way (Elson and Swartz 1989a:Figure 4.25). The S61linear meters of trench that were excavated in Locus A extend over an area 100 m by 80 m. In Locus B, 42 m of trench were excavated in two trenches spaced 10 m apart, to the west of Feature 1, the cobble concentration (Figure 9.2). Hand-trenching and stripping were undertaken within Feature 1 to identify any wall alignments and to evaluate the impact of the root-plowing. The surface collection resulted in the recovery of 700 artifacts from the 20 grid units (7,200 square meters). This calculates to a sherd density of 0.16 sherds per square meter and a lithic density of 0.32lithics per square meter. During the subsurface testing of Feature 1 an additional 23 artifacts were recovered. The diagnostic ceramics recovered from the site included one Sosi Black-on-white, one Holbrook Black-on-white variety A or B, and one Pinedale Black-on-red. It is the only site within the project that yielded White Mountain Redwares from the surface collection, all of which were recovered from the disturbed cobble structure (Feature 1) (Elson and Swartz 1989a:69). The testing results indicate that root-plowing significantly impacted both loci at the site. The only features recorded were ten checkdarns and Feature 1, the disturbed pueblo. The trenches across Locus A indicated that the root-plowing had destroyed all subsurface cultural features. Data Recovery Phase The data recovery phase was begun by excavating additional north-south trenches on S-m centers in Locus A, west of the right-of-way fence along State Route 87 (Figure 9.3). This area had not been root-plOWed so it was felt there was a possibility of undisturbed buried deposits. Two features tentatively identified as pithouses, and two inhumations were recorded in the trenches, although the surface had been disturbed by construction of State Route 87. Once the features were identified, the disturbed layer of approximately 30 ern was mechanically stripped off from over the features. No clear feature outlines were exposed. The two features identified as pithouses were begun by excavating a 2-m by 2-m control unit in 10-ern levels or smaller natural layers. All sediments were sifted through v..-inch mesh, and sherds larger than the size of a quarter-dollar and all other artifacts were collected. The remainder of the feature-fill was removed without screening to S ern above the floor (Stratum 19), which was then screened. Through excavation, one of the features, Feature 1S, was determined to be a ramada rather than a pithouse. This feature was fairly ephemeral so it is difficult to determine whether it was totally excavated. The other feature (Feature 14) is a pithouse and was excavated by exposing the floor in the southern 60 percent of the house and defining the outline of the walls in the northern portion of the house. 234 Chapter 9 AZ 0: 15: 92 (ASM) If I ~o~'I\/ · ,... I I NI60 ./ x TIS I I .; \ l,fea. 16 Inhumations i Fea. I3 0 Il: w ~ NI40 TI3 I Ii 0 x ",,' Fea. 14 Pithouse xl'-. ;} l· .. I. '. I '. \ 1',,', CI) I Fea. 15 Ramada I NI30 T . l' w I- x ~ o .: TI4 ,.. x I / / I tl. x I I I /" J( Fea.17 Checkdam I x NI20 0 .... UJ x Tl7 TI6 I NI50 I I I / 0 0 0 UJ Q UJ UJ 0 ° n-r T22 Q:: 0 lu f- ~x 1/ N260 Ii x I N250 X ;:! CI) • T21 ty I ~ I • F~ Ie I ITS HI6 X N240 I I 'T20 I, J Ie ITiS 0 I meters (TS-24) = oiqfo~ • ~ * o N UJ o o ,.., N UJ N N UJ Figure 9.7. Locus B at AZ 0:15:54. 5 I i N Backhoe trench Rocks Posthole Hearth Inhumation 250 Chapter 9 Summary The data recovery phase lasted from July 31, 1989, to August 11, 1989, expending 43 person-days. Three masonry features and the trash mound were sampled, and the inhumation was totally excavated. The other four features were determined to be too disturbed to define, and further work was not undertaken. MASONRY STRUCTURES Excluding the extremely disturbed and undefinable areas of masonry rubble, three structures (Features 5, 8, and 9), all within Locus B, were excavated. Metric data on the characteristics of these features are presented in Table 9.2. Table 9.2. Metric data from excavated structures at AZ 0:15:54. ..8 Q ale;: e .. .... 11: )1::;; B~ ~ P.Z ~ .. . ~j .. e " ...of! ~ . ~ ... ~ of! 0 Q.Q. Ci~ ...e ..e ~ ...e ...e ~ 'E .- J! ~ tIlCi ~ .. 0 .~= 5 1 3.8+ 3.2+ 1216+ 8.19-8.49 3.65+ 166 95 1 8 1 3.2 1 1 8.09-8.43 1 160 31 1 9 50 4.0 3.8 15.2 8.53-8.92 5.93 284 4.8 66°1 MBD 011 'j§" a . ~ 'ii Ii :.:J Ci .... .. ~ i. . .. 8 = fi: £ al C. ~ .. JS f! :l ::t: "e e 8 Disturbed through road construction Disturbed through road construction = Meters Below Datum Feature 5 Feature 5 was first identified during the testing phase as a possible circular masonry structure in Locus B, west of State Route 87. In this area there was relatively heavy alluviation so only a few rocks were exposed on the surface. The area was shovel scraped to better define the feature. It appeared to have an approximate diameter of 5.5 m (23.75 square meters) with the cobbles being fairly discrete and clustered, although they did not form a clearly defined shape (Elson and Swartz 1989a:75). During the data recovery phase Trench 22 was excavated through the southern side of the concentration. A cobble alignment was visible in the trench wall so the area was again hand-stripped, this time much deeper; at least one layer of cobbles was removed before the walls of the structure were defined. Only the southern and western walls were present. The northern and eastern walls could not be readily found, although time constraints precluded extensive stripping. Therefore, because the whole structure could not be defined, the feature was sampled. A 2-m by 2-m control unit was placed along the west wall and excavated to the floor, sifting the fill in natural layers through V4-inch mesh screen. A 50-cm by 4O-cm unit was then excavated without screening, north of the unit along the west wall to determine whether the cobbles in this area were the surviving northern wall. It was determined that they were wall fall resting on fill. The Rye Creek Drainage 251 Description. Feature 5 is what remains of a cobble masonry structure or pitroom. A 3.8-m segment of the south wall and a 3.2-m portion of the west wall were exposed (Figure 9.8). The area of the structure is a minimum of 12.16 square meters. The orientation of the structure is unknown because there was no evidence of an entry in the two exposed walls. The walls are constructed of relatively tabular river cobbles ranging in size from 25 em to 65 em long, 20 em to 30 em wide, and 10 em to 20 em thick. They appear to have been selected for their shape and size, but were not intentionally shaped. The roughly tabular cobbles were placed horizontally in the wall with the long axis running the direction of the wall. In addition, the side of the cobble facing inward toward the room is fairly flat. Adobe-mud mortar was found between the cobbles. The amount of rubble within the structure suggests that the walls may have stood two or three courses high. The west wall was excavated subsurface approximately 15 em; a single vertical course was resting on the excavated pit wall for a total wall height of 35 em. The pit wall beneath the rocks consisted of compacted sterile sediments, similar to the floor. In the south wall, the western end was excavated into the substratum with a course of cobbles on top, while at the eastern end the cobbles are at floor level. This suggests that to compensate for the natural ground slope from west to east, only the western portion of the structure was excavated in a pit to create a level floor. The natural slope in conjunction with the surface disturbance may also explain the absence of the eastern wall. The fill of the structure contained wall rubble in the upper levels with a moderate artifact density. The fill beneath the wall fall was a medium brown sediment with charcoal inclusions. The artifact density decreased significantly near the floor. The ceramic density overall was relatively high, averaging 166 sherds and 103 lithics per cubic meter. The floor consisted of a compact surface with no evidence of plaster or preparation. There was an area on the southern portion of the floor with red pigment but no floor assemblage. No floor features were exposed within the excavated portion of the structure. Internal Features. No internal features were encountered in the portion of the structure that was excavated. Intrusive Features. No intrusive features were identified. Abandonment and Postabandonment. Feature 5 had no floor assemblage so was probably abandoned before it collapsed. There is little evidence that it burned and the depth of fill below the masonry rubble suggests that it filled naturally prior to its COllapse. The moderate- to high-density of the artifacts in the fill suggests some secondary trash deposition and that the site area continued to be occupied after the structure was abandoned. Feature 8 This feature was first identified during the testing phase as a possible masonry structure. Three rough alignments of cobbles, as well as an amorphous area of scattered cobbles to the east of the alignments, were exposed by shovel-scraping. The overall dimensions of the cobble scatter were 6.0 m north-south by 5.0 m east-west (30 square meters). The longest alignment was 2.5 m long-running northeast-southwest (Elson and Swartz 1989a:75). The arrangement of the cobbles suggested that they may have been disturbed. During the data recovery phase, additional hand-stripping determined that the alignments identified during the testing were not walls. A short distinct row of small rocks however, was identified. These were exposed to reveal the top of a double row of upright slabs. A 2-m by 2-m unit was excavated on the north side of the wall. It was excavated in 20-cm levels to 5 em above the floor (Stratum 19), which was then excavated separately. These levels were sifted through V-!-inch mesh screen. The fill along the wall, to the east and west of the unit, was removed without screening. This was undertaken to follow the wall to see if other walls were present. No comers or any other walls were encountered. 252 Chapter 9 AZ 0: 15:54 (ASM) Feature 5 50 0, N269 O~Metate fragment IIIII/; ~ Probed I cm down to floor N 0 Rock S ,,,--' Wall fall (rock) Rock impression N268 2 x 2 meter control uni t N267 N266 N265 TRENCH 21 It) (\J (\J w w (\J (\J w f'-. (\J (\J w Figure 9.8. Feature 5 at the Cobble site. CXl (\J (\J w I en (\J (\J w The Rye Creek Drainage 253 Description. Feature 8 is defined by one wall of double vertical slabs with a mortar core. The wall runs eastto-west and is 3.2 m long, 0.40 m wide, and 0.34 m high. The west end of the wall has the appearance of an intentional end, as if this were a three-sided structure or the entryway was at this comer. The western rock on the north side is turned sideways, perpendicular to the wall. The east end has been truncated by numerous cobbles that from their shape do not appear to come from this structure (Figure 9.9). The slabs forming the north side of the wall were unshaped but all of a similar size and tabular shape (approximately 34 em long and between 7 and 12 em wide). The slabs on the south side are less tabular. This suggests that the interior of the structure was to the north. There were two large flat-lying rocks on top of the uprights that may be all that remains of a second course. This construction pattern would then be similar to Feature 9 (see below). The only unit excavated was placed on the north side of the wall because this was believed to be within the structure. The fill contained a dark brown, compact silt with a moderate-to-high artifact density and several cobbles in the upper portion. The ceramic density was approximately 160 sherds per cubic meter. The density of the lithic debitage was approximately 25 lithics per cubic meter. The floor was distinguished by the difference between the darker cultural fill and the sterile substratum but was never clearly defined. Excavation continued to the bottom of the wall where a floor pit was exposed but not excavated. There was no floor assemblage. Internal Features. One pit was exposed but not excavated, and no feature number was assigned. It measures 70 em in diameter and is situated against the wall. Intrusive Features. The cobbles that truncate the east end of the wall may be related to a later masonry structure that was badly disturbed by the surface modifications. On the other hand, they may have been pushed there during the surface disturbance and may not represent a former cultural feature. Abandonment and Postabandonment. It is difficult to determine with any certainty the function and nature of the structure. It seems from the small area that remains, that the structure was abandoned prior to its collapse since there is no floor assemblage in the 2-m by 2-m unit that was exposed. The wall fall was in the upper portion of the fill which suggests that the structure partially filled with trash before and while the walls were collapsing. It appears that this wall is all that remains of the structure. What is uncertain is when the disturbance on the eastern end occurred. It could have been a later prehistoric structure that truncated the east end of the wall, or it could have been caused by the modem construction of State Route 87. Feature 9 This feature was initially identified as a subsurface hearth or burned pit (Feature 4) in the profile of Trench 8 during the testing phase. No associated floor or feature outline was visible in the trench profile, although there were large rocks above the feature on the surface that appeared to be randomly scattered. During the data recovery phase but prior to any hand-excavation, the surface was mechanically stripped to identify a masonry wall or pithouse outline if possible. A tentative cobble alignment was identified. A 2-m by 2-m control unit was excavated over the hearth and the possible wall. Fill sediments were removed in 20-cm arbitrary levels or smaller natural layers and sifted through ~-inch mesh screening. The finalS em of fill above the floor (Stratum 19) was treated separately, as were artifacts found in direct floor contact (Stratum 20). The remainder of the feature south of the backhoe trench was excavated without screening down to 5 em above the floor, which was then screened. In the northern portion of the feature, a 20-cm level was excavated without screening to define the walls, although the unit was not excavated down to floor. The area outside of the wall was stripped to the prehistoric ground surface to expose any exterior postholes. 254 Chapter 9 AZ 0 : 15:54 (ASM) Feature 8 N260 TR ENCH 21 1/; -0 !~>-~ CJ Q N259 T""',m,, m N258 ~c:::::::. c::::::> vO "m'/~/& UNEXCAVATED / TO FLOO R / A VD~ / v/ ''\ if / {::5l./ / 0', I oO~g;Mani'~Jr\O 'oj c:::l~C)\ v.. .... l>O C5 O Oc:>.oovV O S i N , Sherd 50 0 I I cm 'e." I:Q + ~ i.. i. "i! ~ ~ ::x:II + 3 + 2 ."a J!l a 8 + -7 + + 7 100 MBD = Meters Below Datum The walls were made of unmodified quartzite river cobbles. These cobbles were almost certainly obtained from either the terrace that the site is situated on, or from the nearby Rye Creek floodplain; they are ubiquitous in both locales. The northern and southern walls consisted of a double row of rocks that averaged 45 cm to 60 cm in length and 20 em to 40 cm in height. The average width at the base of these walls was 1 m, with rocks generally stacked two and three courses high. The western wall also consisted of a double row of rocks stacked two and three courses high, but the rocks tended to be smaller than those in the northern and southern walls; their average length was 25 to 50 em and their average height was 20 to 30 cm. The average width at the base of the western wall was 60 cm. The eastern wall consisted of a single row of rocks, one course high. The average length of these rocks was 50 cm, the average width was 30 em, and the average height was 20 cm. Interestingly, nearly all the inner wall rocks were underlain by 15 em to 25 cm of cultural fill deposits, whereas most of the outer wall rocks were directly on top of sterile alluvium. This suggests that either cultural fill was brought in to build up the wall after the house pit had already been dug, or that the house was built on earlier trash deposits. The second of these alternatives is considered the most parsimonious explanation. The Rye Creek Drainage 269 N 132 +EIIS Feature 23 CJ 00 o Feature 24 ,0- Sterile alluvium I cobbles <> / -:.. () / . O..L ~ Com pact f i II / &"'I~/~~o~ DO 0 ,~~ 111• '. ,. ilom b& F 0 ~F 1-2 gs _\"-~ " N I27 +E 120 Feature I AZ 0:15 : 55 o" A , .. , METER _ !f\N a F I-I o H o A a Os k m me o A Wall rock Rock Feature number Extramural pit Hearth Sherd Lithic Ground argillite Ground stone Tabular knife Mono Metote Plaster Possi ble extent of feoture A' Floor I _._- - - - - - - Floor 2 W/////////~ Figure 9.16. Feature 1 at the Boone Moore site. 270 Chapter 9 The western wall abuts both the northern and southern walls, indicating that it came later. How much later is not altogether clear. The simplest explanation is that the room was built during a single construction episode; however, the possibility that the feature was a two-or-three walled structure that was subsequently remodeled cannot be entirely ruled out. The eastern wall may also be a later addition based on its apparent abutment to the northern wall; the relationship between the eastern and southern walls is uncertain because of wall collapse in this part of the structure. Most of the intact wall rocks were arranged so that the interior surface had a plumb-vertical facing. Occasional chinking stones were present, either in the form of tabular rock fragments wedged between the partially collapsed upper courses, or as rounded cobbles (roughly 15 em in diameter) wedged between the exterior basal course. A possible entryway was indicated by a 5O-cm wide gap in the eastern wall near the southeastern comer. There were no indications that the wall was plastered. The fill sediments from the structure consisted of a medium brown, loosely compacted, sandy silt. Small angular gravels and rocks were present throughout the fill, but very little charcoal or other evidence of burning was observed. The density of artifacts in the fill was moderate-to-high, averaging almost 300 artifacts per cubic meter. An estimated three courses of wall fall were removed during excavation, suggesting that the original wall height might have been at least twice the current height, perhaps around 1.0 to 1.5 m high. The wall fall was largely restricted to the Stratum 10 level. The depth of the house floor averaged 20 em below the basal course of wall rocks. The floor was unplastered except for a small patch just west of the test trench. Along the northern and western walls, the floor was defined as a compacted cultural surface overlying a concentration of small- to medium-sized cobbles. Throughout the rest of the house, the floor was built on top of a thin layer (3 em to 5 em) of sheet trash. The floor assemblage consisted of several flat-lying sherds and pieces of lithic debitage, two mano fragments, a slab metate, and a small piece of ground argillite. The only floor features encountered were three hearths (Features 1-1, 1-2, 1-3). No internal postholes were observed; however, a large triangular-shaped niche in the northeastern corner of the room may have served to support an external post. A similar niche may also be present in the southwestern corner of the house. Internal Features. Feature 1-1 is a plastered, basin-shaped hearth located approximately 1.2 m west of the presumed entryway. Although the test trench "nicked" the eastern edge of the hearth, the estimated diameter was 21 cm and the maximum depth was 20 em below the floor. The entire hearth was plastered except for a rodent-disturbed area near the bottom. The fill sediments consisted of a fine-grained ashy matrix. Archaeomagnetic samples were taken from the best-preserved portions of the feature. Feature 1-2 is a hearth remnant located in the southwestern portion of the house, about 30 em inside the southern wall. The feature had an average diameter of 25 em and a maximum depth of 10 em. It had been badly disturbed by rodent activity, and only about a third of it remained intact. The only evidence of plaster was from around the lip on the northeast side; nonetheless, the entire hearth was probably plastered when in use. The fill sediments consisted of an ashy-silty matrix. The feature was too poorly preserved for archaeomagnetic samples to be collected. Feature 1-3 is a possible hearth remnant located about 0.8 m west of the presumed entryway. The feature was not a formal hearth so much as it was a small (10 cm diameter), shallow (5 em) depression with evidence of oxidation around the edges. A small patch of plaster was identified adjacent to the test trench (see Figure 9.16). Intrusive Features. No features intruded into Feature 1, but at least a portion of the room appears to overlie earlier trash deposits. Abandonment and Postabandonment. Given the minimal floor assemblage, it would appear that the room was purposely abandoned. The fact that the wall fall was mainly concentrated in the Stratum 10 fill levels further suggests that the structure remained standing for some time after its abandonment. There is no evidence that The Rye Creek Drainage 271 the structure burned. The results of the contextual analyses suggest that the feature was filled with a mixture of primary and secondary refuse. Feature 18 This feature was first identified in profile in Trenches 3 and 5, and in plan view as a result of the mechanical stripping. A 2-m by 2-m control unit was then excavated adjacent to Trench 3, in the northwest corner of the structure. Because only 5 to 10 em of fill remained, no Stratum 10 level was excavated. Data recovery began with Stratum 19, which was excavated down to a poorly defined floor (Stratum 20). All fill sediments from the control unit were screened through ¥.i-inch mesh, and all artifacts except plainware sherds smaller than a quarter-dollar in size were collected. Composite flotation samples were collected from Stratum 19/20 levels, and a composite pollen sample was collected from Stratum 20. Following the completion of the control unit, an exploratory trench was excavated along the northern wall, and a possible second floor was found about 10 em below the Stratum 20 level from the control unit. Consequently, the rest of the feature was excavated in two levels. First the upper floor (Stratum 20) and then the lower floor (Stratum 20A) was exposed. Subfloor testing failed to reveal any additional surfaces. Although fill sediments from noncontrol units were not screened, grab collections were routinely made. All artifacts and features from floor contexts were mapped and recorded separately. Description. Feature 18 is either a badly disturbed masonry structure or a bounded extramural activity area. This assessment is based on the fact that only short segments of the northern and southern walls could be identified. Although, it is possible that the rest of the structure, if in fact one existed, was destroyed during the mechanical stripping, it is more likely that it was destroyed at some earlier point in time, perhaps during the occupation of the site, or perhaps through activities associated with the construction of State Route 87. Figure 9.17 shows the upper floor of the structure, and Figure 9.18 shows the lower floor. The walls were made of unmodified quartzite river cobbles, similar to Feature 1, but only 3.2 m of the northern wall and 1.2 m of the southern wall remained intact. Assuming the walls extended at least to the two backhoe trenches, a reasonable assumption given the location of the intact wall rocks, the estimated area of the room is about 20 square meters. The wall rocks varied in length from 35 em to 50 em, in width from 20 cm to 35 em, and in thickness from 10 em to 20 em. All of the rocks were lying directly on the sterile alluvium, and all were arranged so that the interior surface had a plumb-vertical facing. The northern wall consisted of a single course of large cobbles; the southern wall consisted of a single course of large cobbles reinforced by smaller upright chinking stones. There were no indications that the walls were plastered. Although no obvious entryway could be identified, it appears that the structure faced to the south, based on the location of the hearth just inside the southern wall. The fill sediments consisted of a light brown, sandy silt matrix with occasional small gravels. Some charcoal flecking was observed, but it tended to be minimal. The density of artifacts in the fill was fairly high, averaging 330 artifacts per cubic meter. No wall fall was observed in the fill, suggesting, perhaps, that the wall was dismantled shortly after abandonment (see below), or that it was not a full standing wall to begin with. Two floors were exposed during the excavations. Both were poorly preserved, leading to some question as to whether or not they were actual occupation surfaces. The upper floor (Figure 9.17) was defined as an uneven surface that contained large quantities of artifacts, including several whole or partial reconstructible vessels. This surface, which was located about 5 cm below ground surface, was not compacted to any significant degree, nor was there any evidence of plastering or burning. The lower floor (Figure 9.18) extended another 5 em to 10 cm below the upper floor. It was defined largely on the basis of several flat-lying sherds in association with a plastered hearth; it, too, showed no evidence of plastering or burning. Over 50 artifacts were recorded on the two floors. The majority of these were associated with the upper floor, including five reconstructible vessels, two manos, four tabular knives, one stone ax, one metate fragment, one 272 Chapter 9 Feature 18 AZ 0:15:55 G.ii) o cb gd hS Wa II rock ~ Rock RV: ( _ , . Recanstructabt t e vesse t F is., ntrusive leature o Feature numb Sherd er .. Lithic a Axe ~ Human bone Core Hearth (H'" ... f m ml mel ps sh k gsl / // t C remated b Stone disc one ~ round stone ammerstone Mana Mono I ragment M etate Ira Polish' gment s i n g stone N T~~~~~~rkh~lmati t e nI e G round sto ne I ragmen! Disturbed area (in level below) b· Osh o o o o .. b o '0 o QSocb .b o o ~ m mel D~ o C)Qs RV-2 NI26 +EI09 mlCT G t\o gsl ~ ', '..-..I Oml NI24 +EI05 Om ~D) /~-".~I~~ . ,. '.9O'~b g' :~ Figure 9.17. Upper floor of Feature 18 at the BooneMoore site. o • 'UP The Rye Creek Drama o - 273 ... .. • ... o c , - .... , \ I , / \ Om \ , ,I ~odobe ,, , Feature 20 I NI26 EI09 -f o ~ Ph~Phfloor rock set in ;' • II xx omf NI24 -#'05 B OFIS-2 sub floor rocks -@c:J , Feature 18 AZ 0'15 :55 .~ • t om gs .. o 0 ~~w~ A~~~ - t ! , , '0 I CENTIMETERS Wall rock ber FIB-I Feature num Hearth H Pi t mf Feature 20 ~ Figure 9. 18. Lower fl oor of Feature 18 at the Boone Moore site. Mono Mono fragment Ground stone Sherd Lithic a //" Disturbedk:~eroCk x Fire-croc ~ "-, ,Ph Rock . ve feature Intrusi Post hole 274 Chapter 9 lithic core, two pieces of chipped stone debitage, one piece of specular hematite, and several pieces of animal bone. Several sherds and pieces of lithic debitage and ground stone were recovered from the lower floor. The only floor features observed were a hearth (Feature 18-1), a sherd-lined floor pit (Feature 18-2), and a single posthole, all associated with the lower floor. Internal Features. Feature 18-1 is a lightly plastered hearth located about 50 em from the southern wall. The hearth was circular in shape with a flat bottom; it had a diameter of 40 em and a maximum depth of 25 em. The fill sediments consisted of a fine-grained silty matrix, with only occasional flecks of charcoal and pockets of ash. Because of poor preservation, archaeomagnetic samples were not collected from the feature. Feature 18-2 is a sherd-lined floor pit located directly adjacent to Trench 3. The pit had a diameter of 34 em and a maximum depth of 25 em. The edges were lightly oxidized in places, but most of the sherds, which were embedded in the pit wall, did not appear burned. The fill sediments consisted of a light brown, silty matrix with small gravels and a low density of artifacts. Several pieces of charcoal were also observed. The function of the pit is unknown. Intrusive Features. A small roasting pit (Feature 20) intruded into the north-central portion of the house. The roasting pit originated in the fill levels and extended roughly 5 em below the lower floor of the house. Abandonment and Postabandonment. Any reconstruction of the abandonment/postabandonment sequence of Feature 18 must somehow account for the highly disturbed nature of the architecture in conjunction with a floor assemblage that appears to be largely intact, at least with respect to the upper floor. The most likely scenario, it is suggested here, is that only the lower floor was associated with the occupation of the structure, and that the upper floor was an activity surface which post-dates the abandonment and subsequent dismantling of the structure. Support for this idea comes from several lines of evidence. First, floor features were only associated with the lower floor, whereas most of the floor artifacts (including all the reconstructible vessels) were found on the upper floor. Second, there was one instance where the upper floor (but not the lower floor) continued beyond a dismantled section of the wall, suggesting that the dismantling occurred prior to the upper floor's use. Finally, the feature fill sediments contained primary refuse but no wall fall, a pattern that is difficult to explain if the walls were still standing when the primary refuse was deposited. COBBLE-UNED ADOBE PITROOMS Feature 5 This feature was first identified in profile in Trench 4 during the data recovery phase. Handstripping then revealed the outline in plan view. A I-m by 2-m control unit was set up on the east side of the trench and excavated in 20-em arbitrary levels down to the Stratum 19 level. The Stratum 19 level was then removed as a single unit and the floor (Stratum 20) exposed. All fill from the control unit was screened through Y4-inch mesh, and all artifacts were collected except plainware sherds smaller than a quarter-dollar in size. Following the completion of the control unit, the remaining Stratum 10 fill sediments on the east side of the trench were excavated as a single recovery unit. The Stratum 10 fill sediments on the west side of the trench were then excavated in the same manner. During the excavation of the Stratum 19 level, two floors were discovered. This led to a distinction between Stratum 20, the upper floor, Stratum 21, the lower floor, and Stratum 49, the fill between the two floors. Each of these levels was treated separately. Fill sediments from noncontrol units were not screened, but grab collections were made. Composite flotation samples were collected from the Stratum 19/20 and 49/21 levels, and composite pollen samples were collected from the two floors. Description. Feature 5 is a cobble-lined adobe pitroom located near the western edge of the site. The house was oval-to-subrectangular in shape, and covered an estimated area of 22.6 square meters (Figure 9.19). The pit for the house was dug into the alluvial terrace to an average depth of 36 em. The walls of the pit were The Rye Creek Drainage 275 AZ 0 : 15: 55 (ASM) Feature 5 N I21 N I 20 o N I1 9 9~- N 0 I NI04 em '0 I •••• , Extrapolated edge ....-- , : Snerd cluster -" G Rock NI03 CD '" lAO '"'" lAO 0 0 lAO (3 W N Q lAO Figure 9.21. Cobble wall showing reuse of Feature 6 at the Boone Moore site. The Rye Creek Drainage 281 Description. Feature 9 is a badly disturbed pithouse located near the site's northern boundary (Figure 9.22). The house was subrectangular in shape, with maximum dimensions of 5.2 m in length and 4.7 m in width and an estimated area of 24.4 square meters. The house pit was cut into the alluvial terrace to an average depth of 20 em, although this figure should be treated as an absolute minimum given the amount of ground disturbance in the area. A poorly defined entryway faced either to the east or southeast. The fill sediments consisted of a compacted, medium brown silty matrix intermixed with small gravels. Artifact densities were generally low, averaging only 86 artifacts per cubic meter. Small quantities of animal bone also were recovered from the fill. The floor of the house was defined as a compacted cultural surface overlying the sterile alluvium. The compaction was minimal, however, which in combination with poor preservation made defining the floor difficult. The only evidence for a prepared surface was a small patch of plaster near the hearth. There were no indications that the floor burned. A single sherd and a scraper were the only floor artifacts recovered; both came from about 1 m south of the hearth. A small fragment of human bone was recovered near the southeast corner of the structure. The only floor feature clearly associated with the occupation of the house was the hearth (Feature 9-1). Several unexcavated floor pits and the homo remnant are believed to be intrusive features. Internal Features. Feature 9-1 is a plastered hearth located about 1.5 m from the presumed entryway. The hearth was basin-shaped in appearance; it had a diameter of 27 em and a maximum depth of 14 em. The northern portion of the feature had been disturbed by root activity. The fill sediments consisted of a medium brown silty matrix with only small amounts of ash and charcoal. No artifacts were observed in the fill. An archaeomagnetic sample was recovered. Intrusive Features. Feature 29, a remnant homo, intruded into the northern third of the house. Several other floor pits that were not excavated also appear to have intruded into the house (see Figure 9.22). Abandonment and Postabandonment. Little can be said about the abandonment process because of the heavy post-abandonment disturbances. Based on the results of the contextual analysis, the fill sediments consisted of either sheet trash or low density, secondary refuse. Feature 11 This feature initially was identified in profile in Trench 7 during the data recovery phase. Hand-stripping revealed the outline in plan view, and a 1-m by 2-m control unit was excavated adjacent to the backhoe trench. The Stratum 10 fill sediments from the control unit were excavated in 20 cm arbitrary levels. Stratum 19 (5 em above floor) and Stratum 20 (floor) levels were treated separately. All fill from the control unit was sifted through ¥.i-inch screen, and all artifacts except plainware sherds smaller than a quarter-dollar in size were collected. Composite flotation samples were collected from the Stratum 19/20 level, and composite pollen samples were collected from the Stratum 20 level. Following completion of the control unit, the western half of the house (i.e., the portion west of the trench) was excavated as a single recovery unit down to Stratum 19. The Stratum 19 level was then removed and a well-plastered floor exposed. The same procedures were followed in excavating the eastern half of the house. Because it was anticipated from the trench profile that human bone would be encountered, all fill sediments, even those from noncontrol units, were screened through Y4-inch mesh. In addition, all identifiable fragments of human bone were point provenienced (horizontal and vertical grid coordinates). and trend and plunge readings were taken for all long bone fragments. Floor artifacts and features were recorded separately. 282 Chapter 9 AZ 0: 15:55 (ASM) Feature 9 N I 38 NI37 Fee. 29 Horno NI36 q- :r u z W Q: I- NI35 ;Q N I 34 B .: -Plaster Hearth Fee. 9-1 a i l>-Scraper NI33 N - - - - Proboble edge Black horno lining W a NI32 CD en w en en w o o o w w Figure 9.22. Feature 9 at the Boone Moore site. Unexcavated pit Sherd C\J o w The Rye Creek Drainage 283 Description. Feature 11 is a subrectangular pithouse located near the eastern edge of the terrace (Figure 9.23). The house pit cut through a thick cobble bar to an average depth of 50 cm. The house had an estimated area of 18.9 square meters, with maximum dimensions of 4.6 m north-south and 4.1 m east-west. The pit walls had a nearly vertical cut, and in several places were coated with a thin veneer of plaster. A small, poorly defined entryway faced to the west. This contrasts with all other pithouses at the site, which were oriented to the east. The entryway of Feature 11 may have been stepped; a 20- cm high, to-cm wide plastered remnant was observed near its northern edge. The fill sediments consisted of a medium brown, compacted silty matrix with gravel and cobble inclusions. Patches of bright orange, culturally sterile, sandy silts were observed in the general vicinity of the walls throughout the lower fill levels. Presumably, these deposits are the result of walVroof melt. The density of artifacts in the fill levels was mOderate-to-high, averaging 287 artifacts per cubic meter. Relatively few pieces of charcoal were observed. The floor of the house was covered by a nearly continuous layer of caliche-based plaster. The plaster averaged 3 cm to 5 cm in thickness; it was generally thinner near the walls and thicker near the hearth. At least two discrete plastering episodes could be discerned based on differences in the layering of the plaster. There were no signs of burning or oxidation on the floor except for a small area around the hearth. Strata 19 and 20 contained large quantities of disarticulated and partially articulated human bone. Although this bone was widely dispersed spatially, two dense concentrations were noted, one near the northwest corner and the other just east of the backhoe trench in the center of the house. Both concentrations were roughly aligned in an east-west direction. The vast majority of bone consisted of long bone fragments with the proximal and distal ends removed; several of the long bones also had rodent tooth marks on them. Three patellas were found, indicating the remains of at least two individuals (one male adult and one female adult, see Appendix C). Two sets of partially articulated bones were also recovered, a forearm and a lower leg, suggesting that at least one of the bodies was interred while it was still "fresh." In addition to the human bone, the floor assemblage consisted of several flat-lying sherds and lithics, two grinding slabs, one shaft straightener, and one mano. Floor features identified include a hearth (Feature 11-1), nine postholes, and a floor pit (Feature 11-2). Internal Features. Feature 11-1 is a plastered hearth located approximately 1 m due east of the entryway. The hearth was basin-shaped in appearance, and had an average diameter of 20 cm and a maximum depth of 18 cm. The plaster inside the hearth was well preserved, but the eastern two-thirds of the collar had been badly disturbed by rodent activity. The hearth fill sediments consisted of a medium brown silty matrix with few artifacts and very little charcoal or ash. Pieces of plaster, however, were observed in the upper fill levels. Archaeomagnetic samples were collected from the preserved portion of the feature. Feature 11-2 is a shallow, irregularly shaped pit located near the center of the house. The pit extended below the house floor to an average depth of 5 cm. Although the backhoe trench removed a portion of the pit, the remaining portion measured 60 cm in length and 20 cm in width. Several human bone fragments were recovered from the fill sediments, and an articulated limb (a lower leg) was found on the house floor just to the east of the pit (see Figure 9.23). It is possible that this feature represents a disturbed burial pit. If so, though, it implies that nearly the entire house floor was exposed at the time of interment, as the majority of bone was recovered from direct floor contact, not from the pit. Moreover, even making allowances for the portion of the pit removed by the backhoe, it is unlikely that the pit was large enough to have held a full-sized adult, even if the body were in a flexed or semi flexed position. The most likely scenario, it is suggested here, is that the pit was later than both the house and the interments, and that the bones recovered from the pit were ones that had been redeposited from the house floor. 284 Chapter 9 AZ 0 : 15 : 55 (ASM) Feature II NI04 ~Iaster MOlale {l)° blan~ NI03 0 °Gl ~oo COJO GS NI02 o °a/ Hearlh NIOI Rodenl dillurbanco .., :J:: U Z .... a:: ... NIOO 0 0 0 ° N99 • i Pa.,halo OPI! N98 o Shord o L1lhlc N 0 I GS Ground "0 . . B Human bone :",: :.- Hee~tfI 50 I cm ..... 1.... plosl .... o '" ....'" PROFILE - A' Figure 9.23. Feature 11 at the Boone Moore site. The Rye Creek Drainage 285 Feature 11-3 is a large posthole that held one of the main support posts for the house. The feature was cylindrical in shape, with an average diameter of 40 em and a maximum depth of 83 cm. The size of the feature is especially impressive when one takes into consideration the nature of the cobble terrace that it cut through. The pit walls were dark red in color with many cobble inclusions. The fill sediments consisted of a medium brown, silty matrix that contained few artifacts and very little charcoal. Of note, a human tooth was recovered from the lower fill deposits. Feature 11-4 is another central posthole that was similar in most respects to Feature 11-3. The posthole was slightly elliptical in shape and had an average diameter of 40 cm and a maximum depth of 58 cm. The fill sediments consisted of a medium brown, silty matrix that contained few artifacts. Several small fragments of human bone also were recovered, including a tooth. Feature 11-5 is a posthole located near the southern comer of the entryway. The posthole was cylindrical in shape, and had an average diameter of 30 cm and a maximum depth of 24 cm. The pit walls contained several cobbles and an imbedded hammers tone that may have been used to help support a post. The fill sediments consisted of a medium brown, silty matrix with a low density of artifacts. Approximately 2 liters of fill were collected as a flotation sample. Intrusive Features. Feature 11-2 is the only feature that appears to have intruded into the pithouse. As discussed previously, it cut through the floor on the east side of the trench. Given that it is one of the few places where the floor is not plastered, it seems reasonable to suppose that the pit postdates the occupation of the house. Abandonment and Postabandonment. There are several lines of evidence that bear on the abandonment/postabandonment sequence of Feature 11. First, the general lack of floor artifacts suggests that the structure was cleaned out prior to abandonment. Second, the fact that most of the human bone was found either directly on or just above the floor suggests that the bodies were interred at the time of abandonment, or very shortly thereafter. Third, the presence of wall melt in the fill deposits above the floor suggests that the structure was still standing at the time of the interment. The dispersed arrangement of bone on the house floor further supports this idea, as the floor must have been exposed at the time the bone was deposited. Rodents were probably the principal agent responsible for dispersing the bone, although cultural factors may also have been involved (e.g., the excavation of Feature 11-2). The depression formed by the collapsed house was subsequently used for secondary trash disposal. Feature 19 This feature was first exposed in profile in Trenches 4 and 6 during the data recovery phase. Hand-stripping revealed the outline in plan view, and a 2-m by 2-m control unit was set up adjacent to Trench 6 in the approximate center of the house. Excavation of the control unit began with the removal of a roughly 10-cmthick Stratum 10 fill level, followed by Stratum 19 (5 cm above floor) and Stratum 20 (floor) levels. All fill sediments from the control unit were screened through 1/4-inch mesh, and all artifacts except plainware sherds smaller than a quarter-dollar in size were collected. Following the completion of the control unit, the remaining fill sediments (Stratum 10) were excavated as a single recovery unit. This upper fill was not screened, but grab collections were made. Stratum 19 was then excavated and screened through 1/4-inch mesh. Because of the tremendous number of floor sherds, they were mapped and recorded in "sherd groups" rather than as individual artifacts. All other floor artifacts and features were mapped and recorded separately. Composite flotation samples were collected from the Stratum 19/20 level, and composite pollen samples were collected from Stratum 20. Description. Feature 19 is a shallow pithouse located near the eastern edge of the terrace (Figure 9.24). The house was rectangular in shape, with an estimated area of 25.8 square meters; its maximum length was 5.6 m and its maximum width was 4.6 m. The pit for the house cut through the cobble terrace to an average depth 286 Chapter 9 AZ 0: 15: 55 (ASM) Feature 19 OJ NIII :0 10 \000 I ° I I Possible poslhole 0 yr') I o IlsherdclUllerG \ NIIO o ~' I I I A_ NI09 / 0 :E: a: o " II -o> Shtrd Clusler E ~ .i::'......:: .'. 0 .. .' '--~) ,.... ,---- -~ - - 0 .ctd> l- / ...... I " , Cp ;"'/ o , 0' 0 " Sherd Lilhlc @ cJ Rock Sherd Cluller 0 • o, Mono / ,'0 \ 0 I 0 0 Sherd Clusler C , \1 Poslhole t:; I" 0 ~ArQlilite 0 ,J' I I I 50 , em i N NIO!l ! __ 0 ,," " oeer.,1';t<'cf1e 0 mandible _--/ ;';' ~, COo'0 ' ~ .,AJ "- ..... ;' - - ... I Sherd'O'o, ,DoClu"er B / ,," ct1 / Plasler coliar 08 \ /0 , 0 0 ,I" 0 ~\ I OShtrd I \ Clusler F I .... ~--,'/ \ \oJ 0 \oJ \oJ 0 0 ,0 I Z a: • t500 0 I- l.t NI06 ..:0_" Heorlh FeD . ~9-1 11 "- 0 Jloo 00 00 Q'~'~o{)oooo090 U z I lIo I :" \ Bg oori~~g " 10 u 1'1; NI07 , :E: I I g I~ Nloe 0 0 - ___0_ 80 goo O· ID I I I I \ 00 ',__ "" v I 0 0 00 0 ~ogO o 0 0 000 0 \./0 0 ooo~o~ o o ',00 0 00 c:H 00° 0 0 0 I 0 1 I 0 Clu sler 0 0 \0 .., \oJ PROFILE Figure 9.24. Feature 19 at the Boone Moore site. The Rye Creek Drainage 2i37 of 25 cm. Remnant sections of an adobe wall were evident just inside the house pit. In some places the adobe wall, which averaged 5 cm in thickness, abutted the house pit; in other places it was separated by a narrow (5m to lO-cm wide) trench. Several postholes were observed within the adobe wall. A narrow, stepped entryway faced to the southeast. Although it was partially cut by Trench 6, the portion that remained had approximate dimensions of 70 cm in length and 50 em in width. The step was a direct extension of the adobe wall on the south side; it measured B cm in height. The upper fill sediments consisted of a medium brown, silty matrix intermixed with cobbles of various sizes; occasional flecks of charcoal were also observed. The lower fill sediments contained much more adobe wall melt, especially in the final 2 cm to 4 cm above floor. Artifact densities throughout the fill were moderate-tohigh, averaging 255 artifacts per cubic meter. The floor was defined as a compacted cultural surface overlying the sterile cobble bar. The only plaster observed was from around the hearth. Several large concentrations of flat-lying sherds covered parts of the floor. Most of these sherds were "sealed" by adobe wall melt. In addition to the various concentrations of sherds, the floor assemblage consisted of two manos, one lithic core and a few pieces of debitage, a piece of ground argillite, and a deer mandible. Floor features consisted of two large central postholes, several smaller outer postholes, and a plastered hearth (Feature 19-1). Internal Features. The only secondary feature from which either artifacts or samples were recovered was Feature 19-1, a plastered, basin-Shaped hearth located about 1 m from the entryway. The hearth had a diameter of 35 cm and a maximum depth of 20 cm. A plastered apron extended 2 em to 10 cm out from the hearth. The fill sediments consisted of an ashy-silty matrix with occasional flecks of charcoal. Several large plainware sherds and a piece of lithic debitage were recovered from the fill. Archaeomagnetic samples were collected from the plaster lining of the hearth. Intrusive Features. No features intruded into Feature 19. Abandonment and Postabandonment. The results of the contextual analysis suggest that Feature 19 was filled with secondary refuse deposits. Moreover, it seems likely that much of this refuse was deposited shortly after the structure was abandoned, based on the number of artifacts that were covered by wall melt. The fact that the floor contained little in the way of de facto or primary refuse further suggests that the house was cleaned out at the time of abandonment. Finally, there is no evidence that the house burned. MORTUARY FEATURES Six burials, all inhumations, were identified during the data recovery phase. Numerous instances of scattered human bone also were recorded, perhaps the most significant of which was the large concentration found in Feature 11. In the following section, only the formal burials are discussed. Basic descriptive information on the burials is presented in Table 9.5. More detailed information on the bone and artifacts recovered is presented in Volume 2 and Appendix C of Volume 3. Feature 3 Feature 3 is an adult inhumation, probably a female, that was located near the eastern edge of the terrace. The feature was first identified in profile in Trench 2. The outline of the burial pit was then exposed in plan view by hand-stripping. All fill sediments from the pit were screened; the upper 10 cm were screened through 1/4-inch mesh, the lower 20 cm was screened through liB-inch mesh. The exposed burial was mapped, recorded, and photographed. 288 Chapter 9 Table 9.5. Metric data from burials at AZ 0:15:55. Feature Number Percent Excavated Length (m) Width em} De(1th (MBD) Volume (m3) Comments 0.50 100 1.0 10.90-11.20 0.15 Backhoe removed lower 1/3 of body 3 Feature Tvoe Inhumation 7 Inhumation 100 1.6 0.40 11.45-11.62 0.11 Backhoe removed upper skull 8 Inhumation? 100 0.75 0.26 10.79-11.29 0.10 Backhoe removed western 1(2 of pit 17 Inhumation 100 0.98 0.72 10.80-11.45 0.46 FJqn;ed in 00ckh0e trench 21 Inhumation 100 0.75 0.50 11.22-11.38 0.06 23 Inhumation 100 0.85 0.85 10.46-10.55 0.07 MBD No clear pit outline defined = Meters Below Datum The backhoe removed the lower third of the body, basically, from the thighs down to the feet. The portion of the burial pit that remained measured 1 m in length and 50 cm in width, and it cut into the cobble terrace to an average depth of 30 cm. The fill sediments consisted of a medium brown silty matrix intermixed with numerous small cobbles. Some rodent disturbance was also noted. Artifact densities were low-to-moderate, characteristic of general sheet trash at the site. The body was extended in a supine position. The head faced to the east. Most of the bone was in a poor state of preservation; this may partly account for why so few small bones were recovered. Of note, no cranial bones were found, nor were any grave goods observed. Feature 7 Feature 7 is an adult male inhumation, located near the southern edge of the site. The feature was first identified in Trench 6 during the data recovery phase. A section of the cranium was exposed in profile, along with a largely intact Puerco or Escavada Black-on-white restorable vessel. A combination of machine- and hand-stripping was used to remove the overlying sheet-trash deposits and expose the outline of the burial pit in plan view. All fill sediments were screened through lIS-inch mesh. The exposed burial was mapped, recorded, and photographed. Although no clear pit outline could be discerned, there were slight Sedimentary differences between the overlying sheet trash deposits and the lower fill sediments around the skeleton. The overlying sheet trash consisted of a compacted, reddish brown, sandy-silty matrix, with numerous small rock and cobble inclusions in the upper 50 cm to 70 cm. Artifact densities in the sheet trash levels were low-to-moderate. The lower fill sediments consisted of a medium brown, sandy silty matrix that was more loosely compacted than the sheet trash deposits; the lower fill also contained more charcoal flecks and fewer small rocks and cobbles than the sheet trash. The Rye Creek Drainage 289 The body was extended in a supine position with both arms laid straight along the side of the body. The extended length of the skeleton was 1.6 m, and it was about 40 cm wide. As can be seen in Figure 9.25, both arms were laid straight along the side of the body. The right hand overlay the right hip, with the palm facing downwards; the position of the left hand could not be determined. Both legs were slightly flexed at the knee. The left foot appears to have overlain a 20-cm-Iong, 12-cm-wide rock; consequently, only a few bones remained in place. Two sets of grave goods were associated with the burial. First, an upright Puerco or Escavada Black-on-white bowl was recovered from just to the right of the head, slightly overlying the right shoulder. Second, an Olivella shell bead necklace was found around the neck of the body. Interestingly, the beads were strung together in an upright position, rather than end-to-end. No evidence of the material used in stringing the necklace was found. Feature 7 AZ 0 : 15:55 N925 , , , 0 +E1I3 , , 20, CENTIMETER S = RV~ Bon e Reconsl ruc loble ves sel Ro ck l' N I , I <2 ========= C::C N91.5 +E113 Figure 9.25. Feature 7 at the Boone Moore site. 290 Chapter 9 Feature 8 Feature 8 is an infant inhumation that was located just north of Feature 6. The feature was first identified in profile in Trench 4, and then exposed in plan view by hand-stripping. All fill was removed as a single recovery unit and screened through 1/4-inch mesh. The exposed feature was mapped, recorded, and photographed. It is estimated that the backhoe removed the western half of the burial pit. The remaining portion measured 75 cm north-south by 26 cm east-west, and was basin-shaped in appearance with a depth of about 50 em. The fill sediments consisted of a reddish brown silty matrix with occasional flecks of charcoal. Artifact densities were generally low, characteristic of sheet-trash deposits in other parts of the site. Although no articulated bones were observed, numerous fragmentary remains were recovered, including several teeth and cranial bones. In addition, 2 redware bowls and 15 to 20 shell beads were recovered in the fill of the pit. Feature 17 Feature 17 is a probable child inhumation located about 1.5 m north of Feature 11. The feature was first identified in profile in Trench 7 during the data recovery phase. Hand-stripping then revealed the basic outline in plan view. Because of the depth of the burial pit (65-80 cm), the fill sediments were removed in two recovery units. The upper 30 cm were excavated and sifted through l/4-inch screen; the lower 35 to 50 cm was excavated and sifted through l/8-inch screen. When it was discovered that the pit extended further west than originally thought, the exposed bones and grave goods were mapped, photographed, and removed before the rest of the feature was excavated. As it turned out, no additional burial remains were encountered. The burial pit measured 98 cm east-west by 72 cm north-south, and was cut into the cobble terrace to an average depth of 75 cm. The fill sediments consisted of a light, brown orange silty matrix with gravel and cobble intrusions, which closely resembled the surrounding sterile alluvium in composition. Hardly any charcoal was observed in the fill, and it does not appear that burning took place inside the pit. Three redware reconstructible vessels, two bowls and a jar, were recovered from near the approximate center of the burial pit. Of note, an intact redware jar (RV 2) was found below the bottom of the backhoe trench, whereas one of the other vessels (RV 1) was exposed in the sidewall of the trench, roughly 20 cm higher up. A turquoise pendant was recovered from the area around the reconstructible vessels. The only bones observed in situ were several cranial fragments found near one of the vessels. Several other small bones, including a molar, were recovered from the same general area during screening. Feature 21 Feature 21 is a subadult inhumation that intruded into the southeastern comer of Feature 6 (Figure 9.26). The feature was initially thought to be a reconstructible vessel on the floor of the house, but in exposing it human bone was noted. The burial was then excavated as a single recovery unit, and all fill sediments were screened through 1/8-inch mesh. The burial pit was somewhat irregular in shape; it had a maximum length of 75 cm, a maximum width of 50 cm, and a maximum depth of 5 cm below the house floor. In addition to cutting through the floor, the pit partially intruded into the eastern adobe wall. The body was extended in a supine position, with the head oriented to the northeast. The skeleton was well articulated and relatively well preserved, except for the skull, most of which was missing. Three restorable redware vessels, all bowls, were located in the area where the skull should have been, and several cranial The Rye Creek Drainage 291 Feature 21 AZ 0: 15:55 o 10 I , , • , I CENTIMETERS = Bone RV- Reconstructoble vessel t N RV-2 I RV-3 under RV -I Edge N 104 +EIO I -Figure 9.26. Feature 21 at the Boone Moore site. fragments were recovered from under one of them (RV 3). This vessel was stacked underneath another (RV 1); a third vessel (RV 2) was situated next to it (RV 3) at the same depth. No other grave goods were found. Feature 23 Feature 23 was not so much a formal burial as it was a concentration of human bone from just outside the northwest corner of Feature 1. The bone was found during hand-stripping of the area. Since no burial pit outline could be discerned, data recovery consisted of exposing the bone, mapping and photographing it, and then removing it. All fill sediments from around the bone were screened through lI8-inch mesh. The bones encountered include several cranial fragments, a rib, and a possible scapula, all apparently from a child. The bones were not articulated. Although no artifacts were found in clear association with the bones, it is noteworthy that an unworked deer antler was found 10 cm to 20 cm north of the main bone concentration at the same elevation. 292 Chapler 9 EXTRAMURAL FEATURES Roasting Pits Feature 20. Feature 20 is a small roasting pit that intruded into Feature 18. The feature was first encountered and exposed in plan view during the excavation of Feature 18. The southern half was excavated as a single recovery unit, and all fill sediments were screened through l/4-inch mesh. The roasting pit was oval in shape, with maximum measurements of 0.8 m north-south by 0.5 m east-west and a maximum depth of 21 em (Table 9.6). The pit was lined with a 1-cm-thick lens of charcoal and capped by a dense concentration of tightly packed, fire-cracked, quartzite cobbles. The walls of the pit were only lightly oxidized. Artifacts recovered from the fill included several sherds, two hammers tones, and a few pieces of lithic debitage. Table 9.6. Metric data from pits and roasting pits at AZ 0:15:55. Feature Number 2 10 13 20 22 MHD Feature Type Extramural pit Extramural pit Extramural pit Roasting pit Intrusive pit Percent Excavated 100 100 100 50 100 Length (m) Width (m) 0.50 1.03 0.15 0.80 0.95 0.30 0.73 0.15 0.50 0.70 Depth (MHO) 10.65-10.95 10.80-11.17 10.90-10.95 10.32-10.56 10.80-11.27 Volume (m3) O.otl 0.070 0.0003 0.012 0.078 = Meters Below Datum Extramural Hearths Two extramural hearths were located approximately 5 m southeast of Feature 6 during the mechanical stripping. Both hearths were completely excavated. Feature 27. Feature 27 is a plastered, basin-shaped hearth. A rodent burrow ran through the middle of it and Trench 3 removed the western 20 percent. The remaining portion measured 30 em in diameter and had a maximum depth of 10 cm. A single sherd was the only artifact recovered from the fill. Feature 28. Feature 28 is a plastered, basin-shaped hearth. The northwest quarter of the hearth had been disturbed by a rodent burrow, and some of the upper plaster may have been scraped off by the backhoe. Nonetheless, the plaster that remained was well preserved. The hearth had a diameter of 20 em and a maximum depth of 5 em. No artifacts were recovered from the fill. Homo Feature 29. Remnants of a badly disturbed homo were found during the excavation of Feature 9 (see Figure 9.22). Although only patches of the black organic lining remained, the feature was assigned a number and recorded separately. A roughly 1-m-long segment of the lining on the eastern edge was preserved, as was a 10-em-long section along the backhoe trench. Because of the disturbed nature of the feature, no samples were collected. The Rye Creek Drainage 293 Pits Seven pits were identified during the data recovery phase. Four of these (Features 2, 10, 13, and 22) were excavated in their entirety (Table 9.6). The other three (Features 24, 25, and 26) were mapped but not excavated. Only the pits that were excavated are discussed here. Feature 2. Feature 2 is a small extramural pit located near the northern tip of Trench 2. The pit was basin· shaped in appearance, with an estimated diameter of about 50 em and a maximum depth of about 30 em. The entire pit was treated as a single recovery unit. The fill sediments consisted of a dark gray silty matrix with small gravel inclusions. Numerous pieces of burned daub were found near the surface, but there was no evidence of burning inside the pit. Artifact densities were generally low; most artifacts were recovered from near the surface. Feature 10. Feature 10 is an irregularly shaped pit located roughly 1.5 m south of Feature 6, a cobble-lined adobe pitroom. The pit was first observed in profile in Trench 4, and then exposed in plan view through handstripping. Its maximum dimensions were 1.03 m north-south by 70 em east-west. The fill sediments consisted of a loosely compacted, mottled silty matrix intermixed with numerous flecks of charcoal. Artifact densities were moderate, with a wide range of artifact classes represented, including sherds, lithics, ground stone fragments, both human and animal bone, and several pieces of fire-cracked rock. Despite the presence of firecracked rock, it does not appear that burning took place inside the pit, as the walls were not oxidized. Feature 13. Feature 13 is a small pit that was almost completely destroyed by Trench 2. Only about a 15-cmlong, 5-cm·wide stain remained on the east side of the trench. The remaining portion of the pit was excavated as a single recovery unit. No artifacts were recovered. Feature 22. Feature 22 is a trash pit that intruded into Feature 5, a cobble-lined adobe pitroom. The pit was noted in profile in Trench 4, but it was not clearly defined until the upper house floor was exposed. The maximum length of the remaining portion of the pit was 95 em and the maximum width was 70 em. The bottom of the pit extended roughly 1 em to 2 cm below the lower floor of the house. The edges of the pit show signs of oxidation, suggesting that burning took place inside it. Further support for this idea comes from the fill sediments, which consisted of a dark gray ashy matrix containing numerous flecks of charcoal. Artifact densities in the fill were generally low, but at least five charred deer mandibles were recovered. ARTIFACT SUMMARIES Artifact totals are summarized by feature and stratum in Table 9.7. Note that these totals reflect only the major artifact classes. Also note that only artifacts in clear association with a particular feature are included here; thus, artifacts from surface contexts (Stratum 0), mixed contexts (Stratum X), or disturbed contexts (Stratum 80) are not included. The site produced 12,244 artifacts from 19 primary features. Ceramics were by far the most common class of artifacts recovered, accounting for 82.2 percent of the overall assemblage. Chipped stone is the next most common artifact class, accounting for 14.1 percent of the overall total, followed by ground stone (2.8 percent) and shell (0.9 percent). Plainwares and redwares combined account for 99.3 percent of the ceramic assemblage, with plainwares slightly outnumbering redwares by a ratio of 1.15:1. Decorated wares account for the remaining 0.7 percent of the assemblage. In terms of the decorated sample, no buffwares were identified, and whitewares outnumber other decorated wares by a ratio of almost three-to·one. The whitewares were about evenly split between Cibola and Little Colorado whitewares; the only Tusayan Whiteware recovered was a Black Mesa or Sosi Black-onwhite sherd from the fill of Feature 11. Roosevelt Redwares account for just over half of the other decorated wares recovered. Two Roosevelt Redware types were identified: Pinto Black-on-red and Pinto Polychrome. St. John's Black-on-red was the only other decorated type that was clearly associated with a feature. 294 Chapter 9 Table 9.7. AZ 0:15:55 artifact totals by feature and stratum. Pln Red 10 11 19 20 169 32 54 17 2 50 3 5 oth Buff B/W 104 0 10 44 0 0 4 0 25 5 50 16 10 09 10 11 19 20 30 195 421 11 365 9 12 222 419 6 6 6 6 09 10 11 19 20 1161 179 418 93 7 7 09 50 8 2 o 8 8 09 50 96 40 86 9 9 9 09 10 19 5 2 102 27 63 10 50 39 29 11 11 11 11 11 11 09 10 11 19 20 30 10 329 28 359 42 20 17 50 18 17 18 18 18 19 20 30 346 26 33 288 o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o 2 o o o o 19 19 19 19 19 09 10 19 20 30 24 74 373 76 7 13 71 517 20 50 3 4 21 50 2 6 22 50 23 50 15 24 50 3 27 50 o Fea. Str. 5 5 5 5 5 6 Total 64 5 375 71 21 37 629 108 191 60 6 21 5 o 230 12 383 43 18 105 23 382 6 3 5349 3 4653 o o o o o o o o o o o o Dec Debit. 0 163 0 0 0 o 81 2 o o 6 1 1 18 48 6 7 160 1 o o 2 6 o o o o 1 o o 10 1 5 o o 1 o o o 3 o 3 2 o o 2 2 o o 1 o o o o o o o o o 48 o 1 1 1 o o o 1 2 o o 1 o o o o o o o o o o o o o o o o o o o o 1 26 2 131 49 66 o 4 1 o 9 5 Flk Tool Core Tool Oth Met. GS Shell Total 7 0 2 0 0 6 0 000 0 0 0 011 3 4 456 3 3 2 1 1 0 185 27 o o o o o o o o o o 5 2 4 1 11 1 1 8 13 15 3 5 19 o 1 1 8 5 9 o o 3 o 7 o o 2 4 6 3 o o o o 3 2 o o o o o o 1 1 o o o 5 2 o o o o o o o o 1 28 12 Mano 1 o 1 2 7 6 1 1 o o o o o o o o o o o o o 1 o o 2 o o o 68 3 6 2 160 o o 4 o 18 4 6 7 o 10 2 2 10 1 6 o o o 329 23 2 4 6 6 12 3 o o o 14 1 7 115 4 1 o o o o o o o o o o o o o o 17 1537 136 o o o o o 2 2 o o o o o o 51 o 1 o 1 5 o 1 2 4 4 o o o o o o 97 1 2 o o o 9 o o o o o o o o o o o o 21 1 23 o o 1 7 2 7 o 2 o 1 o o 4 o o 6 49 o 16 1 1 o o o o 6 o 1 o o 13 48 36 27 466 935 18 978 82 61 107 1966 350 703 160 20 49 55 1 193 89 19 o o o o 1 o o o o o 13 202 44 76 17 690 45 954 108 50 43 45 15 6 1 o o o 2 o o o o o o o o o o o o o o o o o o o o 224 111 1050 181 60 38 165 1018 468 14 10 8 5 18 4 12244 The Rye Creek Drainage 295 Debitage accounts for 89.2 percent of the chipped stone assemblage, flaked tools account for 7.9 percent, and core tools account for the remaining 3 percent. Manos outnumber metates by a ratio of 4.6-to-1, but overall they combine to represent only 34.5 percent of the ground stone assemblage. Tabular knives and tabular knife material account for almost 60 percent of the miscellaneous ground stone assemblage, and polishing stones account for another 11.6 percent. AZ 0:15:55 produced more shell than any of the other sites investigated; over 60 percent of the sample, though, came from two burials (Features 7 and 8), mostly in the form of beads. POLLEN AND FLOTATION SUMMARIES Pollen Data Eight pollen samples were analyzed from this site, but the one from Feature 17, an inhumation, contained insufficient pollen to be characterized. The remaining samples were taken from a masonry structure (Feature 1) two pitrooms (both floors of Feature 5, and Feature 6), two pithouses (Features 11 and 19), and an extramural pit (Feature 10). Thirty-four different taxa were represented in the samples. Ten of these were found in all the samples: Ambrosia-type (ragweed and related species), High Spine Compositae (sunflower family), Cheno-am, Gramineae (grasses), Boerhaavia-type (spiderling), Sphaeralcea (globe mallow), Pinus (pine), Quercus (oak), Juniperus (juniper), and Zea (com). Among the economic taxa identified, Zea (com) was found in all the samples. Platyopuntia (prickly pear) was found only in the sample from Feature 6, a cobble pitroom. Cylindropuntia (cholla) pollen was found in two pitrooms (both floors of Features 5 and Feature 6) and a pithouse (Feature 19). Cereus (saguaro and related taxa) pollen was found in four samples: the lower floor of Feature 5 (pitroom), Feature 6 (pitroom), Feature 19 (pithouse), and Feature 10, (extramural pit). Onagraceae (evening primrose family) pollen was found in three samples: the upper floor of Feature 5; a pithouse (Feature 19); and an extramural pit (Feature 10). Samples from the upper and lower floors of Feature 5, a pitroom, contained Salix (willow) pollen and Cruciferae (mustard family) pollen was found in the sample from a cobble pitroom, Feature 6. Cyperaceae (sedge) pollen was found on the floor of pithouse Feature 19. Prosopis (mesquite) pollen was found on the bottom of an extramural pit, Feature 10. Flotation Data Nine samples, totaling 34 liters of sediment, were analyzed from this site. Five of the samples were from the lower fill-floor levels of pithouses and pitrooms (Features 5, 6, 11, and 18), two were from internal hearths of pithouse Features 11 and 19, one was from an extramural pit (Feature 22), and the other was from a small roasting pit (Feature 20). All of these features are thought to date to the early Classic period. Agave dominated the botanical sample from the site, accounting for 91.7 percent of the 27.25 relative plant parts recovered. Zea (com) was the next most common taxon; it accounts for 5.5 percent of the sample, followed by two other taxa (Cactaceae [unidentifiable cactus] from Feature 5 and Echinocereus [hedgehog cactus] from Feature 6) that account for the remaining 2.8 percent. No other economic taxa were recovered. Agave was found in all analyzed samples, while com was recovered from Features 6, 11, 19, and 22. SITE CHRONOWGY Six archaeomagnetic samples and approximately 40 diagnostic decorated sherds were recovered from the site. Unfortunately, the archaeomagnetic dates are far from conclusive; the Southwest Master Curve doubles back on itself during the late Preclassic and early Classic periods (Eighmy and McGuire 1989), producing a large set of statistically acceptable alternative dates for each sample. These data are given in Table 9.8. Given the fact that the seventh-century dates for each sample are unacceptable from a ceramic viewpoint, as is the one sixteenth-century date, the only relatively unambiguous date is from pithouse Feature 9 (AD. 1100-1200) 296 Chapter 9 which clearly dates to the late Preclassic period. The dates from the remainder of the features fall within the late Preclassic or early Classic periods (based on the CSV curve. The VA curve, which is not as recent as the CSV curve, is a little less ambiguous since two features appear to date solely to the early Classic period). As a result, relative ceramic dating is the main method for dating the site and its features, although this is not overly conclusive either. The majority of the ceramics are either temporally mixed, in poor context, or both (see Chapters 11 and 12, Volume 2). Three features contained diagnostic sherds associated with their floors or Stratum 19. Pithouse Feature 11 had a partially reconstructible Snowflake Black-on-white (AD. 1050-1150) vessel on its floor; masonry structure Feature 18 had a Reserve or Tularosa (AD. 1100-1300) sherd in Stratum 19 and a Holbrook Black-on-white (AD. 1100-1250) sherd on the floor; and pitroom Feature 5 had a single sherd of Pinto Polychrome (AD. 1250-1350) in Stratum 19. In addition, a Puerco or Escavada Black-on-white (AD. 1000-12(0) reconstructible vessel was recovered from Feature 7, an adult inhumation. Table 9.S. Archaeomagnetic dates recovered from the Boone Moore site (AZ 0:15:55). Feature Feature Type CSU588 (AD.)- UA1982 (AD.)b Feature 1 Masonry pitroom 630-685 920-1045 1160-1305 950-1050 1150-1350 Feature 5 Cobble/adobe pitroom 630-680 980-1050 1060-1100 1155-1335 1100-1350 Feature 6 Cobble/adobe pitroom 630-680 925-1110 1150-1330 950-1050 1100-1350 Feature 9 Pithouse 630-690 915-1035 1530-1615 850-1050 Feature 11 Pithouse 630-675 980-1115 1150-1325 1100-1300 Feature 19 Pithouse 630-685 920-1115 1150-1410 1515-1560 950-1400 ·CSU588 - Colorado State University Southwest Master Curve (Eighmy et al. 1982). bUA1982 - University of Arizona Southwest Master Curve (Sternberg 1982). The fill deposits are extremely mixed; St. John's Black-on-red (AD. 1175-1325), Pinto Black-on-red (AD. 12501350), and Pinto Polychrome ceramics, consistently co-occur with earlier types such as Puerco Black-on-white (AD. 990-12(0), Black Mesa or Sosi Black-on-white (AD. 1000-1150), Holbrook Black-on-white (AD. 10501150), Walnut Black-on-white, Variety A (AD. 1100-1250), Padre Black-on-white (AD. 1100-1250), Leupp The Rye Creek Drainage 297 Black-on-white (AD. 1200-1250), Reserve or Tularosa Black-on-white (AD. 1100-13(0), and Snowflake Blackon-white (AD. 1100-12(0). Holbrook, Reserve or Tularosa, Walnut, and Snowflake black-on-white sherds comprise almost 80 percent (n=23) of the diagnostic whiteware assemblage, and overlap between AD. 1100 and 1200. Holbrook may begin as early as AD. 1050 and Walnut and Reserve or Tularosa may extend as late as AD. 1250 or 1300. The polychrome ceramics (n=12) date between AD. 1250 and 1350, although St. John's may start as early as AD. 1175. These data, in conjunction with the archaeomagnetic dates, suggest a primary occupation spanning the time between AD. 1100 and 1300. The data are not precise enough, however, to ascertain whether occupation of the site was continuous or periodic. Pithouse Feature 9, dating to AD. 915-1035, appears to represent an earlier component, perhaps related to the pithouse occupation at the Compact site (AZ 0:15:90) just across State Route 87. No diagnostic ceramics that could be unambiguously related to this occupation were recovered. SITE INTERPRETATION The Boone Moore site was a farmstead or possibly a small hamlet (since the portion destroyed by State Route 87 is unknown) that was occupied on both a permanent and seasonal basis during the late Preclassic and early Classic periods (see Chapters 26 and 28, Volume 3). Based on the architectural, contextual, and chronometric evidence, it seems unlikely that more than two or three houses were lived in at anyone point in time. The expectation, based on the architectural evidence, is that the pithouses (Features 9,11, and 19) were occupied first, followed by the adobe pitrooms (Features 5 and 6), and then the masonry structures (Features 1 and 18). The results of the contextual analysis generally support this sequence. They indicate that all of the pithouses were filled with secondary refuse (although there is primary refuse directly above the floor of Feature 11), whereas the masonry structures contained mainly primary refuse. The two adobe pitrooms fall somewhere in between, with Feature 5 containing a mixture of primary and secondary refuse and Feature 6 containing a mixture of secondary and transformed secondary refuse. The main point of clarification provided by the archaeomagnetic dates is that Feature 9 appears to date to the late Preclassic period, in contrast to the other structures that date to either the late Preclassic or early Classic periods. With respect to site function, the archaeobotanical data suggest that the site's inhabitants practiced a mixed subsistence strategy; however, the data also suggest that agave was of critical importance. At no other site of comparable size did agave constitute such a large percentage of the macrobotanical sample. Moreover, the site produced the largest sample of tabular knives and tabular-knife material recovered from the project area. Hunting also appears to have been important to the site's inhabitants, based on the number of formal chippedstone tools (bifaces, scrapers, projectile points) and faunal remains recovered. Of note, the site contained one of the largest faunal assemblages recovered from the project area, and artiodactyls and lagomorphs dominated the faunal assemblage (see Chapter 21, Volume 2). Most of the houses were arranged around a large, open area or possible plaza. With two exceptions, all entryways faced inwards towards the open area. The exceptions were Features 1 and 19, which were oriented so as to look out over the Rye Creek floodplain. This pattern may be related to the fact that Features 1 and 19 are the two houses closest to the edge of the terrace; temporal considerations may also have been involved. The plaza-like arrangement of the other houses is suggestive of a more formalized settlement pattern than is usually the case at seasonal sites; the number of burials also is considered atypical for a seasonal site. It may be, therefore, that the site was reused on a fairly regular basis over a relatively short period of time, and that some of the occupations may have been more permanent, as the sedentism analysis in Chapter 26 (VOlume 3) suggests. Given that a similar pattern is evident at AZ 0:15:90, and that the two sites may actually be part of the same prehistoric community (see the following site description), it would appear that there was longterm continuity in land-use patterns along the Rye Creek drainage. THE COMPACT SITE AZ 0:15:90 (ASM) [AR.03.12.06-1107 (TNF)] Doug/as B. Craig The Compact site, AZ 0:15:90 (ASM), is a small farmstead located on the first terrace above Rye Creek (Figures 1.3 and 9.1). The site covers an estimated area of 2,850 square meters, approximately 90 percent of which lies within the project right-of-way. Data recovery efforts focused on three pithouses, an homo, and several extramural pits within the right-of-way. A fourth pithouse that straddled the right-of-way was sampled. The excavated features produced 1,675 artifacts, most of which date to the late Preclassic period (ca., AD. 1000-1150). The site lies on a narrow flat segment of the terrace at an elevation of 3,050 feet above sea level. Vegetation on the site consists of a sparse cover of mesquite, acacia, prickly pear cactus, and grasses. There is a steep drop-off to the north where the terrace rises above the Rye Creek floodplain, and a sharply upsloping cobble terrace forms the site's western boundary. The eastern boundary corresponds to the present State Route 87 roadcut. Given that the road is the only thing separating 0:15:90 from 0:15:55, it seems likely that the two sites represent a single prehistoric community, and that an unknown, but significant, portion of the site has been removed. EXCAVATION METHODS AZ 0:15:90 was first recorded in 1986 during a survey by Archaeological Research Services, Inc. (Stone 1986). No surface features were identified, and fewer than 10 artifacts were observed. Based on this evidence (or lack thereot), Stone (1986:17) concluded that the site, if in fact one existed, was largely destroyed during the construction of the road. Testing Phase The testing phase consisted of first systematically surface collecting the entire site and then excavating three narrowly-spaced backhoe trenches along the long axis of the ridge (Elson and Swartz 1989a). Surface artifact densities were uniformly low, averaging 0.014 artifacts per square meter (n=4O). Seven subsurface features were identified in roughly 120 linear meters of backhoe trenches, including five pithouses (reduced to four during the data recovery phase), an homo, and a possible staging platform or structure that may be associated with the construction of the paved State Route 87 in the late 1950s. A 3D-em to 50-em thick layer of highly compacted sands and silts was found overlying all prehistoric features at the site; it, too, may be the product of late 1950s road construction activities. Data Recovery Phase Data recovery at the site began by removing the mechanically compacted sands and silts through a combination of hand- and machine-stripping techniques. A roughly 125 square meter area was hand stripped and an additional 105 square meter area was machine stripped (Figure 9.27). Once the pithouses within the right-of-way had been exposed in plan view, control units were excavated down to floor level. The remaining fill sediments were hand excavated without screening until 5 em above floor, The Rye Creek Drainage 299 AZ 0: 15: 90 (ASM) N220 N215 Fence line N210 ~ ry; N205 f TI N200 I w~ 1 ~I ~F ~I ~I NI95 «f: ~I fl e NI90 ,L __ NIB5 F.B ~ "'I '01 &1 t ----' E=- ~I ~ ~ NIBD ~ f NI75 10 0 0 w w (7) 10 0 w 0 10 w W -- 0 ~ LLI 1 F ----I N 0 5 I 1 meters ~ Backhoe trench F.2-9 Feotures II/II Unexcavated $ Roasting pit Horno ~ .. '. Elltrapolated ••••• pi thouse edge & Site datum 10 ~ W Figure 9.27. Site map of the Compact site (AZ 0:15:90 [ASM]). 300 Chapter 9 at which point all fill was screened through V4-inch mesh. Extramural features were exposed in plan view and then bisected and sampled in most instances. Summary The data recovery phase lasted from June 22 through July 17, for a total of 107 person days. Eight prehistoric features were identified within the project right-of-way. All of these were either sampled or excavated in their entirety. The historic staging platform was not investigated because of its apparent recent age. PITHOUSES Four pithouses were excavated as part of the data recovery phase. Basic summary information on them is presented in Table 9.9. More detailed descriptions are provided below. Table 9.9. Metric data from pithouses at AZ 0:15:90. Q !! .. B1.. _l :! a:lC;: ~:;; ;J '"e.. '"e ~ :! ~ 4.5 3.3 14.9 11.16-11.26 1.485 90 67 6.3 4.2 26.5 10.80-10.89 1.598 100 4.6 3.5 16.1 10.93-11.13 3.2 cr !: ~z ~! 3 100 4 5 ~ :! ..8 . ~ ~ -s -sCi Coo Coo ~~ . e ~ ~ i! .JI ~ CIl~ '"e ., .~ :s ~ ;:J~ .. . ..,cr i... ~ 'e .. !~ cr ~ 0 'D ~ . cr .. .e II cr 8 ~ ::z: e 8 C5 a:l 94 280 + . 142 58 80 + 94 10 5 + + ~ Ii: - MBD .. Meten Below Datum Feature 2 This feature was identified during the testing phase as possible burned pithouse. But because it straddled the project right-of-way, the feature was only sampled during the data recovery phase. A I-m by 2-m control unit was excavated adjacent to the east wall of Trench 2. The unit was hand-excavated in arbitrary 20-em levels until 5 em above the floor. The final 5 cm of fill and the unplastered floor were treated separately. No floor artifacts or features were observed, but the control unit did expose the southeastern comer of the wall. Based on this evidence in combination with the profiles exposed in the trench, it appears that the long axis of the house was aligned in a northeast-southwest direction. In the trench wall this profile extends a distance of 1.70 m. Feature 3 This feature was initially identified during the testing phase in both walls of Trench 3. At that time it was recorded as a burned pithouse. The results of the data recovery phase confirmed this assessment. The Rye Creek Drainage 301 Data recovery began by exposing the feature in plan view. This was done by hand-stripping the layer of compacted silts and overlying sheet trash deposits. Once the outline of the house had been determined, a 2-m by 2-m control unit was hand-excavated adjacent to the east wall of the trench. Fill sediments were excavated in 20-em arbitrary levels and screened through ¥.t-inch mesh. All artifacts except plainware sherds smaller than a quarter-dollar in size were collected and grouped by recovery level. Flotation and pollen samples were collected from each control recovery level. Following completion of the control unit, the remaining fill sediments were hand-excavated in 10 em or greater arbitrary levels. The fill was not screened until 5 em above floor; however, a grab sample of artifacts was made from each recovery level. The last 5 em of fill was treated as a single recovery unit and screened. Floor artifacts and features were mapped and recorded separately. Description. Feature 3 is a shallow pithouse located near the eastern edge of the terrace (Figure 9.28). The house pit was somewhat irregular in shape; the southern wall was oval whereas the northern wall was fairly straight. The maximum dimensions of the house were 4.5 m north-south by 3.3 east-west, with an estimated floor area of 14.9 square meters. The house pit was cut into native soil to an average depth of 25 em. The entryway was slightly bulbous and faced to the west. Two large upright rocks were wedged into the western wall of the entryway and another was located near the northern wall. The two western rocks may have served as steps, but the northern rock appears to have been used to buttress the comer support post. This inference is based on the fact that the two western rocks are situated at the edge of the house pit, whereas the northern rock is set in about 20 em from the wall, near a presumed posthole. The pithouse fill sediments consisted of a dark brown silty matrix with small gravel inclusions. The density of fill artifacts was moderate-to-high, averaging 184 artifacts per cubic meter. A distinct roof fall and wall melt layer was evident beginning at about 8 cm above floor and continuing down to floor. Fill deposits in this wall melt layer contained numerous flecks of charcoal and pieces of burned daub. Signs of rodent disturbance were evident throughout the fill. The floor of the house was un plastered. It was defined as a compacted surface directly overlying the sterile alluvium. Parts of it were heavily oxidized, in particular, the area inside the entryway and along the southern wall. The presence of a plastered hearth and several flat-lying artifacts further assisted in floor definition. The floor assemblage consisted of four manos, a possible pecking stone, a piece of fire-cracked rock, and several flat-lying sherds. Three of the manos were in close association near the northwest comer of the house, about 80 em west of a small ash pit (Feature 3-3). Other floor features present include a plastered hearth (Feature 3-1) and 12 postholes along the outer edges of the house. No central postholes were found. Internal Features. Feature 3-1 is a plastered hearth located about 1 m from the entryway. The backhoe trench removed approximately half the hearth, and most of the remaining portion was destroyed when it collapsed into the trench following heavy rains. Nonetheless, several general comments about the feature can be made. First, it was basin-shaped in appearance, and had an estimated diameter of 30 em and an estimated depth of 8 em to 10 em. Second, a roughly 30-cm-wide plastered apron surrounded the hearth. This apron was raised 2 em to 3 em above the floor. Unfortunately, because of the disturbances noted, no artifacts were recovered and no fill sediments were available for flotation analysis. For similar reasons, no archaeomagnetic samples were collected. Feature 3-2 is a posthole located near the southeastern comer of the entryway. The feature was 15 em in diameter and had a maximum depth of 10 em. The sides of the pit were heavily oxidized, and pieces of a post that had burned in situ were recovered for wood species identification and possible radiocarbon dating. The only artifacts in association with the feature were two sherds and a mano that was partially imbedded into the southern wall of the pit, presumably to help support the post. Feature 3-3 is a small, oval-shaped pit located in the northeast comer of the house. The pit measured approximately 50 em north-south by 35 em east west, and extended below the floor to a depth of about 7 em. 302 Chapter 9 AZ 0: 15: 90 (ASM) Feature 3 N I 97 • NI96 • A ... NI95 a ,., '. N I 94 :I: U Z W a: l- I ,<~ NI93 fI;/ ~ 0 .. 0 0:": ~\' i '( Fea .I O' .0 • N a NI92 0 I 50 en O Pi t .: : ::. Hearth plaster a ® o I cm :': Possible '. posthole -r"''''' ~ Intrusive pit Sherd Mono Fire-crocked rock Rock o o w PROFILE -AI Rock Entryway \ Figure 9.28. Feature 3 at the Compact site. The Rye Creek Drainage 303 The fill sediments consisted of a fine-grained ashy matrix. The edges of the pit were oxidized. One piece of chert debitage was recovered from the fill deposits. Intrusive Features. A large basin-shaped pit (Feature 10) intruded into the southwest comer of the house. The pit was evident in profile in the west wall of the trench and in plan view on the house floor. Most of it was removed by the backhoe; the portion that extended below the house floor was not excavated. Abandonment and Postabandonment. Given the abundance of charcoal and burned daub in the fill sediments, the presence of oxidized patches on the house floor, and the burned post in situ near the entryway, a strong argument can be made that the entire house, or at least major portions of it, burned. The general lack of floor artifacts further suggests that the house had been cleaned out prior to the burning episode. Based on the results of the various contextual analyses, the house appears to have been trash-filled following its abandonment. Feature 4 Because this feature appeared in two separate trenches during the testing phase, initially it was assigned two separate feature numbers, 1 and 4. It turned out, though, that the backhoe had caught the entryway and back wall of the same structure. When the hand-stripping revealed this situation, the two features were combined into one and designated Feature 4. Data recovery began by exposing the feature in plan view. A control2-m by 2-m unit was then hand-excavated down to floor level. This control unit was placed adjoining Trench 2 in the center of the house. A partially plastered house floor was encountered about 5 em below the opening level. Consequently, a second 2-m by 2-m control unit was excavated adjacent to the first one. The remaining northern third of the house was excavated as a single recovery unit. The floor throughout the house had been badly disturbed by rodent activity and a series of postoccupational intrusive pits, probably connected with the use of Feature 6, a large homo that intruded into the southeastern comer of the house. The southern third of the house was not excavated, partly due to time constraints but also because of disturbances caused by the homo. Description. Feature 4 is a large, irregularly Shaped pithouse located near the western edge of the site (Figure 9.29). Although the backhoe removed the back wall of the house and the southern third was not excavated, the house is estimated to have measured 6.3 m north-south by 4.2 m east-west. The entryway was long (1.7 m), narrow (0.9 m), and slightly off-center; it faced to the east. The house pit was dug into native soil to a depth of at least 8 em to 10 cm. Given the disturbed nature of the site's surface, this figure probably does not accurately reflect the true depth of the house pit. The fill sediments consisted of a loosely compacted mixture of fine-grained sandy silts. Artifact densities were moderate-to-high, averaging 200 artifacts per cubic meter. Although a few pieces of charcoal were found, burned daub was conspicuously absent. Both the fill and floor levels had been badly disturbed by rodent and root activity. The floor of the house was poorly preserved in most places, largely due to the disturbances noted above. Nonetheless, a roughly 2.5-m-long, 30-cm-wide segment of floor plaster was preserved around the hearth in the center of the house. This plaster averaged 2 em in thickness. The floor in the rest of the house was defined as a compacted cultural surface overlying the sterile alluvium. There were no indications of oxidation or burning on this surface. The floor assemblage consisted of two manos, one fragment of ground stone, and several flat-lying sherds. The sherds tended to be located near the outer edges of the house; the manos and ground stone were located about 50 em from the hearth (Feature 4-1) and ash pit (Feature 4-2). These two features and the isolated posthole 304 Chapter 9 AZ 0: 15: 90 (ASM) 4 Feature NI99 o NI98 • NI91 NI96 6).0. J: ~ Fea. · '. '.: : " : '......... : . 4 - 2 ' ' ... . . Hearth •. , ' . ' 4 6 llil 0 . 0 I- ··:········ 0 . . . '.' . '. , A.h pI! : : .' p.' : : ::::·· N ..•.. ' . x: Fe? 4·1 u Z ILl II: I- NI95 NI94 Inlrullve Horno Fea. 6 NI93 1 N NI92 • 0 ~~.. . .. .( l'~ ... ' ' 0 ILl Probable edge o Shlrd Posthole M Mono GS Pit //1//1 Inlruslve pil Floor pla.'er N ~ ILl / / '"o iii / o 50 ~ cm Ground stone Rodent disturbance Unexcovaled area or o iii on CD iii ILl o o PROFILE Possible .,ep .t Figure 9.29. Feature 4 at the Compact site. __ ~_-ti The Rye Creek Drainage 305 near the northwest comer are the only floor features clearly associated with the use of the house. Five other floor pits were identified, but at least one of these is probably a large rodent burrow and most of the others appear to be intrusive (see below). Internal Features. Feature 4-1 is a well-plastered hearth located about 1.5 m southwest of the entryway. It was basin-shaped in appearance, with an average diameter of 50 ern and an average depth of 17 ern. A rodent burrow had disturbed the eastern collar and wall of the hearth, but otherwise it was in good condition. The fill sediments consisted of a medium brown, silty matrix that became increasingly darker in color near the bottom. Several sherds, including one that had been reworked into a spindle whorl or gaming piece, were the only artifacts recovered. Feature 4-2 is a small ash pit located 75 cm west of the hearth. The pit measured 25 ern east-west by 20 ern north-south with an average depth of 8 ern. The walls of the pit were oxidized a bright orange color. The fill sediments consisted of a fine-grained ashy matrix with no charcoal inclusions or artifacts. Feature 4-3 is a large irregularly shaped pit located about 1.0 m due west of the entryway and 25 ern northeast of the hearth. The pit measured 1.4 m north-south by 70 ern east-west, and extended roughly 30 ern below the house floor. The fill sediments consisted of a medium brown, silty matrix that was easily distinguishable from the orange colored native soil. Artifact densities in the fill levels were quite high, averaging almost 250 artifacts per cubic meter. Feature 4-4 is a large pit that cut through the floor of the house just south of the entryway. The roughly 50 percent of the pit that was excavated measured 1.35 m east-west by 75 ern north-south, with an average depth of 25 cm. The fill sediments consisted of a medium brown, silty matrix that contained a small number of charcoal flecks. The artifact density was generally low, averaging 76 artifacts per cubic meter. It bears mentioning that the sherd and fragment of ground stone shown on the house map (Figure 9.29) were at the top of Feature 4-4, not the bottom. Feature 4-5 is a medium-sized pit that cuts through the floor near the western edge of the entryway. The pit was irregular in shape, with a maximum length of 85 ern and a maximum width of 45 ern. The average depth at the bottom of the pit was 28 ern below the house floor. The fill sediments consisted of a medium brown, silty matrix that was easily differentiated from the native soil. Although there were occasional flecks of charcoal in the fill, the pit walls were not oxidized. It does not appear, therefore, that burning took place inside the pit. The density of artifacts in the fill was low-to-moderate, averaging 89 artifacts per cubic meter. Feature 4-6 is a medium-sized pit located roughly 35 cm west of the ash pit (Feature 4-2) and 50 ern inside the western wall. The pit was is oval in shape, with maximum dimensions of 55 ern north-south and 40 ern east-west, and an average depth of 30 ern below the house floor. The fill sediments consisted of a fine-grained ashy matrix in the upper 3 ern to 5 ern and a medium brown, Silty matrix in the lower 25 ern. Artifact densities were extremely high, averaging 470 artifacts per cubic meter. Several small pieces of charcoal were found near the bottom of the pit, but only the upper 5 cm of the pit wall showed evidence of burning. This suggests that the feature functioned secondarily as a firepit, after it had already been used for something else and partially filled in. Given that the upper ash lens originated at floor level, the use of the feature as a firepit probably coincided with the occupation of the house. By extension, the original pit probably predated the construction of the house. Intrusive Features. Feature 6, an homo, and an unexcavated oval-shaped pit (no feature number), clearly intruded into the house; the homo destroyed most of the southeastern quarter of the house and the ovalshaped pit destroyed a portion of the northeastern comer. In addition, three of the excavated floor pits (Features 4-3, 4-4, and 4-5), as a well as an unexcavated fourth pit, may also have been intrusive. Because the pithouse fill deposits were so shallow, the exact stratigraphic relationship of these floor pits to the house could not be determined. It would have been almost impossible, though, to move about inside the structure with the pits located where they were. It thus seems unlikely that the pits were associated with the use of the 306 Chapter 9 house. Rather, it seems more likely, especially given their proximity to Feature 6, that they were associated with the use of the homo. Abandonment and Postabandonment. Given the general lack of floor artifacts, it appears that the house was abandoned with no intention of returning. The structure appears to have decayed gradually over time; there is no evidence that it burned. At some point following the abandonment and collapse of the house, a large homo was dug into the southeastern comer. Other pits associated with the use of the homo were probably dug at that time as well. The results of the contextual analyses suggest that the fill of most of these pits and the house in general contain secondary trash deposits. Feature 5 This feature initially was recorded during the testing phase as a possible burned pithouse. This assessment was based on the presence of small flecks of charcoal in the fill sediments and an apparent oxidized floor. The results of the data recovery phase confirmed this assessment in part. A portion of the structure apparently burned, but what previously had been thought to be an oxidized floor turned out to be a plastered one instead. Data recovery began by exposing the feature in plan view. A 2-m by 2.35-m control unit was then handexcavated in 10 and 20 em arbitrary levels until 5 em above floor (Stratum 19). The last 5 em of fill were treated separately, as were floor artifacts and features. All fill sediments from the control unit were screened through v..-inch mesh, and all artifacts except plainware sherds smaller than a quarter dollar in size were collected. A composite flotation was collected from the Stratum 19120 level, and a composite pollen sample was collected from the floor level. Once the control unit had been excavated, the remaining fill sediments were hand-excavated without screening until 5 em above floor. A grab sample of artifacts was made from each recovery unit. The last 5 em of fill was screened and treated as a single recovery unit. Floor artifacts and features were mapped and recorded separately. Description. Feature 5 is an irregularly shaped pithouse located near the site's western boundary (Figure 9.30). Although Feature 6, the homo, had removed the north wall of the house, its estimated area is 16.1 square meters, with maximum dimensions of 4.6 m east-west by 3.5 m north-south. The house pit was dug into native soil to an average depth of 25 em to 30 em. Although no entryway or hearth was identified, presumably because construction of Feature 6 destroyed them, the house appears to have faced to the north. The pithouse fill sediments consisted of a medium brown, fine-grained silty matrix. Charcoal flecking was evident in the middle fill levels, but not in the Stratum 19 level. The density of artifacts was moderate, averaging 104 artifacts per cubic meter. Moderate amounts of burned animal bone (rabbit or rodent size) also were recovered from the fill levels. The floor of the house was defined by a I-em to 2-em-thick caliche-based plaster. The only portions of the floor not plastered were a roughly I-m by I-m area near the presumed entryway, a small patch along the western wall, and a somewhat larger area along the eastern wall. Nine postholes were found at regular intervals along the outer walls. These postholes averaged 10 em in diameter and 5 em to 10 em in depth. No central postholes were found. Two manos were the only floor artifacts found; one was located near a posthole in the southwestern comer of the house, the other was located near a posthole in the northeastern comer. An apparent burned roof beam was recovered from floor contexts along the western wall. Internal Features. Other than the nine postholes, which were not assigned secondary feature numbers, the house contained no internal features. Intrusive Features. As discussed previously, Feature 6, the homo, destroyed most of the north wall of the house. No other obvious intrusive features were found. The Rye Creek Drainage 307 AZ 0: 15: 90 (ASM) Feature 5 Post hole 0 Sherd mef Metate frogment gs Ground stone 0 if' N I Figure 9.34. Feature 1 at the Arby's site. +N201 E 93 The Rye Creek Drainage 319 Figure 9.35. Photograph of Feature 1 at the Arby's site. of rearranging portions of the interior wall fall into a small, semicircular, structure opening to the south. This remodeling appears to account for the difference in the architecture of the southern wall. As a result, only the 5 em to 9 em beneath the floor of Feature 5 can be considered to be related to the occupation of Feature 1. The fill of Feature 1 below the floor of Feature 5 consisted of a gray-to-brown clayey silt with a fair amount of charcoal flecking and pieces of charcoal, as well as several pieces of burned daub. This, along with the fact that the internal posthole shows evidence of oxidation, strongly suggests that the structure burned. The fill contained a moderate-to-high artifact density of 149.3 sherds and 94.4 lithics per cubic meter. Artifacts recovered from the fill include 57 plainware and 37 redware ceramics, 63 pieces of lithic debitage, and two scrapers. An argillite polishing stone and a single mano were recovered from Stratum 19. The floor was not prepared or plastered, and consisted of a slightly compacted surface with flat-lying artifacts. Portions of the floor were slightly oxidized, although the burning was not uniform nor very intensive. The floor had been heavily disturbed through rodent and root activity. The floor assemblage consisted of a few scattered plainware sherds and a cluster of redware sherds situated near the entrance. Two internal features were present: Feature I-I, a small hearth, and Feature 1-2, an ash pit. Internal Features. Feature 1-1 is a small, well-fired hearth located approximately 1.5 m west of the entrance. The hearth was basin-shaped and roughly circular, measuring 30 em in diameter and 10 em deep. It appears 320 Chapter 9 that it was originally plastered with caliche, since small remnants are present along the edges and bottom; extensive root disturbance has removed the majority of the plaster, and archaeomagnetic samples were not recovered. The upper 3 em to 5 em of the fill were similar to the fill of the house, while the lower layers consisted of a fine white ash. No indications of remodeling were present and no artifacts were recovered from the fill. Feature 1-2 is a small ash pit situated approximately 15 em west of the hearth. The pit measured 24 em in diameter and was only 4 em deep. The pit was slightly oxidized and the fill contained dark gray ash mixed with pieces of charcoal and charcoal flecks. A single sherd was recovered from the fill. The pit may represent a small, secondary hearth, or it may represent an ash dump from the use of Feature 1-1. Intrusive Features. Feature 5, a small windbreak or brush shelter constructed from the walls and wall fall of Feature 1, is intrusive into the fill of Feature 1 (see below). Abandonment and Postabandonment. Feature 1 appears to have been purposely abandoned prior to the burning of the structure. This is based on the lack of a useable floor assemblage. The burning probably occurred a short time after abandonment, however, because the roof fall layer was found resting on the partially burned floor. A fair amount of cultural trash was then deposited within the structure, prior to the remodeling, indicating continued use of the site area. At some later point the structure was remodeled by the occupants of Feature 5, who constructed a relatively ephemeral rock-lined windbreak or brush structure from the southern wall and wall fall of Feature 1. Feature 2 Feature 2 is a linear cobble rock alignment, or series of alignments, situated approximately 30 m north of Feature 1 (Figure 9.33). The alignment is situated on the edge of the Rye Creek floodplain, and due to disturbance from Rye Creek, it is not clear whether there is a single alignment or several alignments; as many as three small separate alignments may be present. The entire alignment measures approximately 9.5 m long by 0.75 m to 1.0 m wide and is oriented northeast-by-southwest. Test excavations revealed alluvial silts and clays to a depth of 40 em behind the alignment, with quartzite cobbles stacked three courses high. A single piece of lithic debitage was recovered from the test unit. The cobble alignment appears to have been used for agricultural purposes, perhaps functioning as a kitchen garden area for Feature 1. This is inconclusive, however, because no Zea (corn) pollen was recovered. Furthermore, the location of the alignment does not appear to be suitable for trapping sediments because it is not situated upon even a slight slope. This may be due to the grading of the area east of the alignment for the construction of State Route 87, and it is possible that a slope existed prehistorically. This is suggested by the 4O-em accumulation of sediments behind a wall of cobbles three courses high. On the other hand, given the one-meter-wide width of the cobbles within the alignment, it is also possible that the alignment was used as a trap for moisture in a similar manner as rock-pile features. Perhaps both factors were in operation. Feature 3 Feature 3 is a very small masonry structure or slab-lined pit room located in the eastern component of the site, approximately 110 m southeast of Feature 1 (Figure 9.33). The western wall of the structure is situated less than 50 em east of the bladed State Route 87 right-of-way and the structure is visible from the road. Disturbance to the structure is clearly evident; as mentioned above, there are indications that the eastern component had been either plowed or bladed. As a result, the true extent of the disturbance and the number of additional features that were present are unknown. Description. Feature 3 measures 2.8 m north-south by 2.4 m east-west (6.7 square meters) (Figures 9.36 and 9.37). The structure is actually a slab-lined pit room rather than a surface masonry structure, because the floor feo,u re '3 Al 0 : \5:9 9 \ a \ D -- o .0 f 3-2 \"\ a m qs _J 50 ~ t N \ \Je!l i Col wall roc\<. wall P!ob fall e'l.lenl of leolU!e oble f'eoIU!e number rleo!lh sne!d Mono G!ound slOne FIgUre 9.36. Feature 3 at tlte AIby's site. A' 322 Chapter 9 Figure 9.37. Photograph of Feature 3 at the Arby's site. had been excavated to a depth of approximately 20 em beneath the prehistoric ground surface. The north and west walls are composed of a single row of upright tabular quartzite cobbles placed vertically against the sides of the excavated pit. Both walls measure 2.8 m in length, and are composed of cobbles ranging in size from 10 em to 15 em wide by 30 em to 40 cm high. The southern wall appears to have been similarly constructed, although only a small portion remains; depressions are present within the edges of the excavated pit from removed upright cobbles, however. A hard, compact, clay mortar was set between the upright cobbles, and evidence for chinking stones was noted at the northwest and southwest comers. The eastern wall is enigmatic. All that was found within this area was a somewhat amorphous north-south alignment of large, rounded, cobbles, lying 5 em to 10 em above the floor of the structure (Figure 9.36). The nature of this alignment is unclear, because the cobbles are not stacked and do not appear to represent an in situ wall; they are also rounder and more irregularly shaped than the cobbles found in the other walls. It is possible that these represent the disturbed remains of an eastern wall that was architecturally different than the other walls, or it is possible that the structure never contained an eastern wall. Extensive extramural stripping within this area failed to resolve this problem, since no additional cobbles or signs of disturbance noted. A nearly identical and well-defined three-walled structure is present within the project area (Feature 5 at AZ 0:15:53), and a relatively large number of three-walled structures have been recorded in the Tonto Basin (G. Rice 1985; Ciolek-Torrello 1987) and elsewhere (Ward 1978). It is assumed that the walls were covered with a brush-and-adobe superstructure, although no internal or external postholes were noted. Furthermore, given the nature of the disturbed southern and eastern walls, it is not possible to determine the The Rye Creek Drainage 323 entrance or orientation of the structure, although if it is in fact a three-walled structure, the entrance would have been to the east. The structure contained approximately 25 em of a grayish brown, clayey silt fill. Pieces of charcoal and charcoal flecking became more apparent in the lower levels, suggesting that a brush or wood superstructure burned. The fill contained a relatively high artifact density. The ceramic density was 244.7 sherds per cubic meter and the lithic density was 47.6 lithics per cubic meter. Artifacts recovered from the fill include 229 plainware and 179 redware ceramics, 72 pieces of lithic debitage, two lithic tools (a scraper and informal lithic tool), and two core tools (core and hammerstone). A polished argillite stone and two scrapers were recovered from Stratum 19. The floor consisted of a moderately compacted surface with several oxidized patches; no plastering or other preparation was noted. The floor assemblage contained several pieces of ground stone, a mano, and a few scattered sherds. Two internal floor features (Features 3-1 and 3-2), both hearths, were discovered. The location of the hearths, particularly Feature 3-1 which is the more substantial of the two, somewhat supports an eastern orientation (Figure 9.36). Internal Features. Feature 3-1 is a small, basin-shaped hearth with relatively deep, straight sides. The hearth is situated approximately 50 em west of the presumed location of the eastern wall. It measures 22 em northsouth by 20 em east-west and is 14 em deep. The upper 10 em of the hearth walls are caliche-plastered, although the bottom is not, and a slight lip extends from the hearth a few centimeters. The fill consisted of a light ashy brown silt with heavy charcoal flecking with 2 to 3 em of solid gray ash at the bottom. No artifacts were noted and archaeomagnetic samples were not recovered. Feature 3-2 is a small, circular, shallow hearth situated approximately 70 em southeast of Feature 3-1. The hearth measures 20 em in diameter and is 8 em deep. The top 3 cm of the fill contained an orange brown silt with little charcoal flecking while the bottom 5 em contained a gray brown, ashy silt. The floor area immediately surrounding the hearth was heavily oxidized. No artifacts were noted and archaeomagnetic samples were not recovered. Abandonment and Postabandonment. The oxidation of the floor and the presence of a burned roof fall layer indicates that the structure burned. The lack of a useable floor assemblage, however, suggests that the structure was purposely abandoned prior to the burning. The presence of two hearths suggests possible reuse of the structure, perhaps over several seasons. The cobbles located above the floor in the approximate location of the eastern wall are enigmatic. They may be unrelated to the feature and indicate additional reuse of the site area, perhaps serving as a temporary windbreak. The moderate-to-high trash fill within the structure further indicates that this portion of the site area continued to be used after the structure was abandoned. Feature 4 Feature 4 is a small rock-lined hearth situated in the western component approximately 1.25 m east of Feature 1 (Figure 9.33). The hearth measures 34 cm in diameter and has a maximum depth of 14 em. The upper portion of the hearth pit is lined with small rocks, S em to 15 em in diameter, some of which show signs of thermal alteration. The lower portions of the hearth walls, and the hearth bottom, are highly oxidized. Artifact density within the fill of the pit was low, only four pieces of lithic debitage, a redware sherd and a plainware sherd were recovered. No animal bone was noted. An extramural surface containing a relatively large number of pieces of lithic debitage was located in the vicinity of the hearth, and it is possible that limited lithic reduction was occurring here. A single Zea cupule fragment was recovered from the flotation sample. 324 Chapter 9 Feature 5 Feature 5 is a small cobble-lined brush structure or windbreak situated within the fill of Feature 1 (Figure 9.33), approximately 5 em to 9 em above the floor. The walls of the structure were uncovered initially in the northern control unit of the Feature 1 excavation. Although the cobbles were found to be aligned in a roughly semicircular pattern, due to the similarity of the rocks to the internal wall fall (the walls were in fact constructed out of Feature 1 wall fall), they were removed. The location of the alignment was noted and sketch-mapped as patterned wall fall prior to removal. When the southern half of the structure was excavated a continuation of the alignment was noted, and several flat-lying artifacts and an ephemeral hearth (Feature 5-1) were found. At this point a new feature number was assigned and bags were changed from the previously designated Feature 1 Stratum 10 (fill) to Feature 5 Stratum 10 (fill). Description. Feature 5 consists of a semicircular cobble alignment measuring 3.2 m north-south by 2.8 m eastwest (9.0 square meters). The alignment was constructed from the wall fall of the north, west, and east walls of Feature 1. These cobbles were laid flat and consisted solely of a single course. The south wall appears to have been constructed through the partial removal and remodeling of the south wall of Feature 1. The south wall consists of upright tabular cobbles extending approximately 1.0 in from either end, leaving a 6O-em wide entrance to the south. The 18 cm of fill consisted of a gray brown, clayey silt with a high density of charcoal flecking and a moderate number of pieces of charcoal. Artifact density was low in comparison to the other structural features, consisting of 52.5 sherds and 51.3 lithics per cubic meter. Artifacts recovered from the fill include 58 plainware ceramics, 26 redware ceramics, a single indeterminate Cibola whiteware, 81 pieces of lithic debitage, a scraper, and a Conus shell tinkler. A second Conus shell tinkler was recovered from just outside the eastern wall of the structure during the testing phase. No ground stone or other artifacts were recovered. The floor was extremely ephemeral, and could only be defined through the location of the hearth (Feature 5-1) and the base of the cobble walls. No floor artifacts were noted. The relatively high charcoal density suggests that the structure burned, although the floor was not oxidized. This is probably due to the nature of the superstructure, which was most likely insubstantial and composed primarily of brush, and therefore could not support a prolonged, hot, bum. Internal Features. Feature 5-1 is a small ephemeral hearth situated in the west-central portion of the structure. The hearth was initially noted as an ash stain corresponding with the depth of the base of the walls. Upon excavation, a poorly defined, slightly oxidized, and very shallow pit was found, measuring approximately 40 em to 50 cm north-south (the northern edge was not definable) by 40 em east-west with a depth of only 3 em. The pit fill contained charcoal flecks, ash, and three plainware sherds, within a silty matrix. The overall lack of oxidation strongly suggests that the hearth was not used for any great duration. No archaeomagnetic samples were recovered. Abandonment and Postabandonment. Feature 5 appears to be an ephemeral, and temporarily used, cobblelined brush structure. The structure evidently burned soon after abandonment, as suggested by the presence of burned fill resting on the floor. The structure was then filled with a low density of trash, indicating continued, although probably low-intensity, use of the site area after abandonment. This is further supported by a sherd match between the fill of Feature 5 and the fill of Feature 3, suggesting that both portions of the site were in use at this time. The high density of the fill within Feature 3, however, suggests that the east side of the site was used more intensively. Feature 6 Feature 6 is a small, extramural pit or posthole situated approximately 2.3 m northeast of the northeast comer of Feature 1 (see Figure 9.34). The feature measures 20 cm north-south by 16 em east-west and is 16 em deep. A large rock abutted the hole to the east. The pit was filled with gray ashy sediments, suggesting possible The Rye Creek Drainage 325 burning, although the soil was not oxidized. A single redware sherd was recovered. The exact nature of this feature is unclear. Although its small size and shape suggests that it may have been a posthole, perhaps for some sort of extramural shade or ramada, no other extramural postholes were noted. ARTIFACT SUMMARIES Artifact totals are summarized by feature and stratum in Table 9.11. Note that these totals reflect only the major artifact classes. Also note that only artifacts in clear association with a particular feature are included here; thus, artifacts from surface contexts (Stratum 0), mixed contexts (Stratum X), or disturbed contexts (Stratum 80) are not included in this table. Table 9.11. AZ 0:15:99 artifact totals by feature and stratum. Debit. Flk Tool Core Tool Mano 0 0 0 0 11 51 1 0 0 2 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 29 131 13 1 1 0 0 1 0 0 39 33 0 2 3 0 2 1 0 0 0 1 0 0 0 0 1 0 0 0 0 276 215 7 o o o 4 o o o o o o 6 0 0 1 0 0 0 81 0 1 0 0 0 0 0 0 0 1 0 1 0 169 3 0 0 0 0 0 0 0 0 0 0 Buff BlW oth Dec 7 30 10 1 0 0 0 0 0 0 0 0 100 0 0 0 Pln Red 11 19 20 30 11 46 2 0 3 3 3 10 19 20 131 98 1 4 50 5 5 10 30 58 3 6 50 0 Fea. Str. Met. Oth GS Shell Total ----------------------.-.-------------------------------------------------------------------------------------------0 0 0 0 0 0 09 0 0 0 0 0 2 -----------------------------------------------------------------------------.--------.----------------------------.79 5 26 0 --------------------------------------------------------------------------------------------------------------------Total 351 260 o 2 221 8 3 2 o 3 The site produced 852 artifacts from the subsurface excavations of the six primary features. Ceramics account for 72.1 percent of this total, chipped stone for 27.2 percent, ground stone for 0.6 percent, and shell for 0.1 percent. Within the ceramic assemblage, plainwares and redwares together account for 99.5 percent of the total, with plainwares slightly outnumbering redwares by a ratio of 1.35-to-1. The remaining 0.5 percent of the ceramic assemblage consists of two indeterminate-but-conjoining Cibola whiteware sherds from the fills of Features 3 and 5, and one Showlow Black-on-red corrugated sherd from the lower fill of Feature 3. Debitage accounts for 95.3 percent of the chipped stone assemblage, core tools account for 3.4 percent, and flake tools account for 1.3 percent. The ground stone assemblage consists of two manos, two polishing stones, and a tabular knife fragment. Additional whitewares (n=5) were recovered from the surface of the eastern locus, and consist of a single indeterminate Tusayan whiteware, three indeterminate Little Colorado whitewares, and the above-mentioned Flagstaff-style Little Colorado Whiteware. In addition, two pieces of shell, Conus tinklers from disturbed areas within Feature 1/5 were recovered, as was a single polishing stone and a single polished stone, both of red argillite. 852 326 Chapter 9 POLLEN AND FLOTATION DATA Pollen Data Three pollen samples were analyzed from this site: one from each cobble room (Features 1 and 3) and one from the cobble terrace (Feature 2). Nineteen different taxa were identified, nine of which were found in all three samples: Ambrosia-type (ragweed and related species), High Spine Compositae (sunflower family), Cheno-am, Gramineae (grasses), Boerhaavia-type (spiderling), Sphaeralcea (globe mallow), Pinus (pine), Quercus (oak), and Juniperus (juniper). There were few economic taxa identified. Zea pollen was found in samples from both the cobble rooms, while Cereus (saguaro and related taxa) pollen was additionally found in the sample from Feature 1. Salix (willow) was found in Feature 2, the terrace, although no Zea pollen was noted. Flotation Data Four samples, totaling 12.5 liters of sediment, were analyzed from this site. Two of these samples were from the lower filVfloor levels of structures (Features 1 and 3), one was from an extramural hearth (Feature 4), and the other was from a hearth associated with the remodeled cobble-lined brush structure (Feature 5). All of these features date to the early Classic period. Zea (com) was the most common taxa in the botanical sample, accounting for just over 50 percent of the 3.75 relative plant parts recovered. This is the only site in the project area where com was found in such relative abundance; com was recovered from all four analyzed samples. Cheno-ams were the next most common taxa at the site, accounting for 32 percent of the sample, followed by agave (10.7 percent) and Echinocereus (hedgehog cactus, 6.7 percent). Agave was only recovered from Feature 5, while hedgehog cactus was only found in Feature 3. SITE CHRONOLOGY The ceramic data, which includes a single Showlow Black-on-red sherd (AD. 1050-1200) and a single F1agstaffstyle Little Colorado Whiteware (AD. 1100-1250), strongly suggest that the site dates to the early Classic period (AD. 1150-1300). This is supported by the presence of Salado Red Corrugated (approximately AD. 1200-1350+), the high frequency of redwares (42.1 percent of the ceramic assemblage), the relative lack of Tusayan whitewares, and the corresponding higher frequencies of Little Colorado and Cibola whitewares. No absolute dates were recovered. SITE INTERPRETATION The Arby's site appears to represent a series of seasonally occupied field houses. Due to the disturbance caused by the construction of State Route 87, the true size and nature of the site are unknown; additional structures may have been removed by the construction. Furthermore, it is difficult to conclusively determine whether the masonry features on either side of the road are contemporaneous and related to each other, or indicative of separate occupations. The sherd match between the fills of Features 3 and 5 suggest that the entire site area was in use after the abandonment of the structures, although this cannot be extrapolated to suggest contemporaneous use of the site area when the structures were occupied. The site contains a much greater diversity of artifacts and a higher artifact density than found at most of the other fieldhouse sites in the project area, and it is possible that the occupation was of a more substantial nature; the analysis of seasonality presented in Chapter 26 still indicates a seasonal occupation. The site probably was reused over a number of years, accounting for the density and diversity of the artifact assemblage. The Rye Creek Drainage 327 The close proximity of the large and arable Rye Creek floodplain suggests that agriculture was a primary site function. This is supported by the presence of a linear cobble terrace (Feature 2) along Rye Creek and by the recovery of corn remains in both the pollen and flotation samples. Given the sherd match between the fills of Features 3 and 5, however, it is plausible to suggest that Feature 1 was occupied prior to either feature. The much higher redware-to-plainware ratio of Feature 5 (2.3-to-1) as compared to Feature 1 (0.8 to 1) and Feature 3 (0.8 to 1) also tentatively suggests that Feature 5 was the latest occupied of the three structures. This is further supported by the relative lack of trash filling within Feature 5, indicating that at least this portion of the site was not intensively reoccupied after the abandonment of the structure. The greatest number of artifacts at the site were recovered from the fill of Feature 3, suggesting that the eastern portion of the site was occupied on a more intensive basis than the western portion of the site. Given, however, the disturbance caused by the construction of State Route 87 and the blading or plowing of the eastern site area, the nature and intensity of the occupation are unknown. 328 Chap" 9 CHAPTER 10 TESTING PHASE SITES Deborah L. Swartz and Mark D. Elson Nineteen sites were recorded by Archaeological Research Services, Inc. during the survey for the realignment and widening of State Route 87 (Stone 1986). The testing of these sites was undertaken by the Institute for American Research (now Desert Archaeology, Inc.). The testing phase identified 13 sites that contained evidence of surface or subsurface cultural features within the right-of-way and were therefore considered to be potentially eligible for inclusion in the National Register of Historic Places and deserving of further work. These have been described in the preceding chapters. The testing of the remaining six sites indicated that no significant surface or subsurface features were present within the right-of-way. Three of the six sites contained masonry structures but these were outside of the right-of-way and would not be impacted by the road construction. All six sites were fully recorded during the testing phase and no further work was undertaken at these sites during the data recovery phase. The following summaries of the sites that were tested but not mitigated are taken from the testing report (Elson and Swartz 1989a). Site AZ 0:15:101 was not part of the Rye Creek Project but is included here because it is located just north of the project area and was tested at the same time as the Rye Creek Project sites. It was tested by the Institute for American Research for Mr. Ken Haught who wanted to purchase the property from the Tonto National Forest. The description of the site in this chapter is summarized from the Haught testing report (Elson and Swartz 1989b). 330 Chapler 10 AZ 0:15:51 (ASM) AZ 0:15:51 is a two-room masonry structure situated on the west end of an east-west trending ridge that is part of the Mazatzal pediment (Figures 1.3 and 6.1). A tributary of Hardt Creek originates on this ridge approximately 60 feet lower and 70 m southeast of the site. The site measures 52 m east-west by 40 m northsouth (2,080 square meters) with all but the western 7 m outside of the proposed State Route 87 right-of-way (Figure 10.1). There is a very light vegetation cover consisting of a few juniper, acacia, prickly pear cactus, and grasses. This site was originally recorded by the Arizona State Museum and the Museum of Northern Arizona and reevaluated by Stone (1986:9). It is described as two contiguous cobble masonry rooms within an associated low-density artifact scatter. The testing assessment of this site agrees with the previous interpretations. Testing Methods The portion of the site that lies within the proposed right-of-way was totally collected as a single unit due to its small size and low artifact density. A grid was placed at 10-m intervals across the two structures and the structures were mapped and photographed. No test excavations were undertaken because the structures are outside of the right-of-way. Feature Descriptions Feature 1. This feature is a rectangular, dry-laid cobble masonry room with a contiguous room to the east, Feature 2 (Figure 10.1). The walls are constructed of unshaped river cobbles, which are found along the side of the ridge to the south. The cobbles range in size from 15 em to 60 cm in diameter. The interior dimensions of the room are approximately 5 m north-south by 3 m east-west (15 square meters). These dimensions may, however, underrepresent the true size of the structure due to the amount of internal wall fall. The walls and associated wall fall measure 3 m wide, except the wall shared with Feature 2, is only 2 m wide, and there are some indications that the walls were two courses wide. It is very difficult to determine which rocks represent the walls and which are wall fall. The interior of the structure, which is clear of rocks, is extremely narrow east-west, suggesting that the west wall fell at least partially inward. Without excavation data it is difficult to estimate the original height of the walls from the amount of rubble because both features are slightly mounded above the original ground surface. Feature 2. This feature also is a rectangular dry-laid cobble masonry structure, which is east of Feature 1 and contiguous to it. The walls are constructed of the same unshaped cobbles as Feature 1. The interior dimensions of this room are slightly larger, measuring apprOximately 5 m north-south by 4.75 m east-west (23.75 square meters). There is much less wall fall associated with the walls of this feature than Feature 1. The walls and associated fall measure 2 m wide on the north and west walls but only about 1 m wide on the east and south sides. The rubble to the south is dispersed and may have been rearranged after the wall fell or may represent an alcove. From surface data alone it is difficult to determine whether these two rooms were contemporaneous. It is possible that they were contemporaneous and the walls of Feature 1 were simply higher or wider than those of Feature 2. Alternatively, Feature 1 may have been occupied after Feature 2 and the cobbles from Feature 2 were used in the construction of Feature 1. Artifact Assemblage Artifacts were only collected from within the proposed right-of-way, thereby eliminating the majority of the site area. The total collection of this area consists of 48 artifacts. This calculates to approximately 0.4 artifacts per square meter. The artifact assemblage includes seven plainware ceramics, six redware ceramics, and 35 pieces of chipped stone. The surface collection recovered only a single lithic tool, a rhyolite/andesite "plano" AZ 0: 15: 51 " "" NI30 \)OQClO()~ ---------- - - ---- "..------ -- ~ M S ~~ ~~~b .... .... Site BOundary--+/ Within Ri ght -of - Way / I o I '" .... .... " \ \ \ \ \ / NI20 I / NIIO I I ,, I I I I I I \ \ NIOO o m w I °00 o.o~~o OO~ a:: .(>l~ ~ () g " 0 O~2~0t)'I>~ &8~ ~'o~o 'be ~ {oo uot .... N90 ,, \ Q 0 0. 0 00 00 ()O .... .... .... ........ GO 00 ;' ;' .... ( ,. .... " 00 0 lS ;' ;I' Site Boundar Outside Right -of-Way '" / 00 " / ..... .... " ;' ;' "" "" ----o o o w w N o o ,..., o W w W Figure 10.1. Site map of AZ 0:15:51 (ASM). ':'~: .. ~ \ • •• \ \ \ NIOO Meters 0 • -... T2 \ \ N 110 10 L '_ _ _ _- ' , \ : .. .' S \ \ o (":':':.::. ~/ ..... :::'.: :, / \ \ Construction Disturbance \ \ "\.. .' \ \ \ J\ \ \ \ \ N90 \ \ Site Boundary Within Right-of-Way \ \ \ \ \ \ \ \ N80 o ,... o o o o lLJ lLJ lLJ UJ