<%BANNER%>

Evaluating Cave Use Through Spatial Analysis of Animal Remains from Maya Caves in Guatemala and Belize

Permanent Link: http://ufdc.ufl.edu/UFE0041312/00001

Material Information

Title: Evaluating Cave Use Through Spatial Analysis of Animal Remains from Maya Caves in Guatemala and Belize
Physical Description: 1 online resource (219 p.)
Language: english
Creator: Kavountzis, Erol
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: directionality, gis, maya, ritual, space, spatial, zooarchaeology
Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, M.A.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Scholars suggest caves in the Maya region were used for ritual activities and represented a connection to the underworld (Brady 1989; Stone 1995). Recent studies (Moyes 2001, 2002, 2004) use spatial and GIS analysis of artifact distributions to understand the role of spatial cognition in Maya ritual cave use based on directionality. Other studies suggest a variety of cave ritual functions and several cognized spatial divisions within caves such as north-south, east-west, right-left, light-dark, and restricted-open spaces. This zooarchaeological study uses Geographic Information Systems (GIS) to test the spatial distribution of animal remains within multiple cave sites in Belize and Guatemala. These cave sites include Caves Branch Rockshelter and Stela Cave in Belize and Cueva de El Duende, Cueva de Sangre, and Naj Tunich in Guatemala. Patterning is addressed using both spatial analysis tools from ArcGIS and a visual analysis of animal remains by taxonomic groups. While spatial patterning is present for key taxonomic groups of animal remains, overall patterning of skeletal remains is rare due to small sample sizes, limited distribution of remains, taphonomic processes, the use and placement of animal offerings, the shape of the caves, and the possible differential use of caves within the Maya region. If used properly, GIS coupled with with zooarchaeological analysis can provide a new method for investigating the use of animal remains in cave contexts.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Erol Kavountzis.
Thesis: Thesis (M.A.)--University of Florida, 2009.
Local: Adviser: Emery, Kitty F.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041312:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041312/00001

Material Information

Title: Evaluating Cave Use Through Spatial Analysis of Animal Remains from Maya Caves in Guatemala and Belize
Physical Description: 1 online resource (219 p.)
Language: english
Creator: Kavountzis, Erol
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: directionality, gis, maya, ritual, space, spatial, zooarchaeology
Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, M.A.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Scholars suggest caves in the Maya region were used for ritual activities and represented a connection to the underworld (Brady 1989; Stone 1995). Recent studies (Moyes 2001, 2002, 2004) use spatial and GIS analysis of artifact distributions to understand the role of spatial cognition in Maya ritual cave use based on directionality. Other studies suggest a variety of cave ritual functions and several cognized spatial divisions within caves such as north-south, east-west, right-left, light-dark, and restricted-open spaces. This zooarchaeological study uses Geographic Information Systems (GIS) to test the spatial distribution of animal remains within multiple cave sites in Belize and Guatemala. These cave sites include Caves Branch Rockshelter and Stela Cave in Belize and Cueva de El Duende, Cueva de Sangre, and Naj Tunich in Guatemala. Patterning is addressed using both spatial analysis tools from ArcGIS and a visual analysis of animal remains by taxonomic groups. While spatial patterning is present for key taxonomic groups of animal remains, overall patterning of skeletal remains is rare due to small sample sizes, limited distribution of remains, taphonomic processes, the use and placement of animal offerings, the shape of the caves, and the possible differential use of caves within the Maya region. If used properly, GIS coupled with with zooarchaeological analysis can provide a new method for investigating the use of animal remains in cave contexts.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Erol Kavountzis.
Thesis: Thesis (M.A.)--University of Florida, 2009.
Local: Adviser: Emery, Kitty F.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041312:00001


This item has the following downloads:


Full Text

PAGE 1

EVALUATING CAVE USE THROUGH SPATIAL ANALYSIS OF ANIMAL REMAINS FROM MAYA CAVES IN GUATEMALA AND BELIZE By EROL GEORGE KAVOUNTZIS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Erol George Kavountzis 2

PAGE 3

To my family and Jessica, without all of your support and help I could not have done this 3

PAGE 4

ACKNOWLEDGMENTS This thesis was only possible with the help and assistance of my advisor Dr. Kitty F. Emery, who helped to explore a nd understand the Maya and their use of caves. I would also like to thank Dr. Susan deFrance for her support and i nput for my research and Dr. Christian Russell whos GIS knowledge made most of this research and mapping possible. Without the work and help of all of those in the Zooarchaeology la b in the Florida Museum of Natural History (FLMNH), I not sure I could have done this. Big thanks to Elyse Anderson, Erin Thornton, Michael Kay, Melissa Ayvaz, and all the Ethnobota nical interns for making my analysis fun and productive at the same time. Also thanks to th e zooarchaeology intern s, Erin Ives and Jen Boekenoogen, for their work with organizing my faunal assemblages. This work was partially funded by a Tinker Fi eld Research Grant provided by the Center for Latin American Studies at the University of Florida. This funding allowed me to export my large collection of animal bones from Belize du ring the summer of 2007. I would like to thank the researchers who shared their faunal assemblages for this study including Dr. Jaime Awe, Dr. James Brady, Dr. Gabriel Wrobel, and graduate student Cameron Griffith. I thank my great neighbors and friends the Reut er family, including Ma son, Georgia, Ellie, and Wilson. I thank all of my wonderful Housing coworkers, GHDs and RAs, for making my experience at UF fun and enjoyable. I would also like to thank my family for all their support both mentally and financially. A big hug and thanks go to my great friends and support at UF, Elyse, Mary, and Jon, you are the three that made my experience wonderful. Finally, to my BFF Jess, thanks for helping me get through this lo ng, and sometime painful, journey, without your support I dont think I could have done it. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES .........................................................................................................................10LIST OF FIGURES .......................................................................................................................12ABSTRACT ...................................................................................................................... .............13CHAPTERS 1 INTRODUCTION ................................................................................................................ ..15Cognitive Archaeology ......................................................................................................... ..18Maya Cave Archaeology: .......................................................................................................1 9Maya Cave Zooarchaeology: ..................................................................................................20Geographic Information Systems and Maya Caves ................................................................22Maya Spatial Patterns, Zooarchaeology, and GIS ..................................................................22Summary ....................................................................................................................... ..........242 THEORETICAL FRAMEWORK: CAVES, LANDSCAPE, COGNITIVE ARCHAEOLOGY, AND THE USE OF SP ACE BY THE ANCIENT MAYA ...................26Overview ...................................................................................................................... ...........26Maya Cognitive and Landscape Archaeology ........................................................................27The Cognized Universe and Caves .........................................................................................29Separation of Space and the Use of Space by the Ancient Maya ...........................................33Cardinal Directions and Quadripartite: ...........................................................................33Left and Right Sides ........................................................................................................34Light and Dark .................................................................................................................36Open and Restricted Access ............................................................................................37Summary ....................................................................................................................... ..........383 PRIOR RESEARCH IN MAYA CAVE ARCHAEOLOGY, ZOOARCHAEOLOGY, AND GIS ....................................................................................................................... .........40Maya Cave Research: History and Progress over the Decades ..............................................40Brief History of Maya Zooarchaeology ..................................................................................42Ritual Use of Animals from the Maya Region .......................................................................43Previous Maya Cave Zooarchaeological Research .................................................................47GIS in Cave Research Projects and the Maya Region ............................................................49Current Research Goals ..........................................................................................................514 RESEARCH SETTING: FIVE CAVES FROM THE MAYA REGION UNDER ANALYSIS ...................................................................................................................... .......52 5

PAGE 6

Introduction .................................................................................................................. ...........52Primary Research Sites in Belize ............................................................................................53Caves Branch Rockshelter, Belize ..................................................................................53Stela Cave, Belize ............................................................................................................ 55Primary Research Sites in Guatemala .....................................................................................56Cueva de Sangre, Guatemala ...........................................................................................56Cueva de El Duende, Guatemala .....................................................................................57Published Source from Guatemal a: Naj Tunich, Guatemala ..................................................58Summary ....................................................................................................................... ..........595 ZOOARCHAEOLOGY, GIS, AND VISUAL ANALYSIS METHODS ..............................67Introduction .................................................................................................................. ...........67Zooarchaeological Methods ....................................................................................................6 8Geographic Information Systems (GIS) .................................................................................71Separation of Space within the caves .....................................................................................74Summary ....................................................................................................................... ..........766 RESULTS ..................................................................................................................... ..........88Introduction .................................................................................................................. ...........88Caves Branch Rockshelter ......................................................................................................90Crustaceans ................................................................................................................... ...91Actinopterygii ................................................................................................................ ..92Amphibia ...................................................................................................................... ...93Testudines .................................................................................................................... ....94Sauria ...............................................................................................................................94Serpentes ..................................................................................................................... .....95Aves .................................................................................................................................96Didelphidae ................................................................................................................... ...96Dasypus novemcintus ......................................................................................................97Chiroptera .................................................................................................................... ....98Canidae ....................................................................................................................... .....98Procyonidae ................................................................................................................... ..99Artiodactyla .................................................................................................................. ...99Tayassuidae ................................................................................................................... 100Cervidae .........................................................................................................................101Rodentia ...................................................................................................................... ...101Agoutidae and Dasyproctidae ........................................................................................102Sylvilagus sp. .................................................................................................................103Stela Cave .............................................................................................................................103Crustaceans ................................................................................................................... .104Actinopterygii ................................................................................................................ 104Amphibia ...................................................................................................................... .105Testudines .................................................................................................................... ..105Sauria .............................................................................................................................105Serpentes ..................................................................................................................... ...105 6

PAGE 7

Aves ...............................................................................................................................106Didelphidae ................................................................................................................... .106Dasypus novemcinctus ..................................................................................................106Chiroptera .................................................................................................................... ..107Canidae ....................................................................................................................... ...108Procyonidae ................................................................................................................... 108Artiodactyla .................................................................................................................. .108Tayassuidae ................................................................................................................... 109Cervidae .........................................................................................................................109Rodentia ...................................................................................................................... ...110Scuiridae ..................................................................................................................... ...110Agoutidae and Dasyproctidae ........................................................................................110Sylvilagus sp. .................................................................................................................111Cueva de El Duende .............................................................................................................111Crustacean .................................................................................................................... .112Actinopterygii ................................................................................................................ 112Amphibia ...................................................................................................................... .113Testudines .................................................................................................................... ..114Sauria .............................................................................................................................114Serpentes ..................................................................................................................... ...114Aves ...............................................................................................................................115Didelphidae ................................................................................................................... .115Dasypus novemcinctus ..................................................................................................115Chiroptera .................................................................................................................... ..116Canidae ....................................................................................................................... ...116Felidae ....................................................................................................................... ....117Artiodactyla .................................................................................................................. .117Cervidae .........................................................................................................................117Rodentia ...................................................................................................................... ...118Agoutidae and Dasyproctidae ........................................................................................118Cueva de Sangre ...................................................................................................................119Crustaceans ................................................................................................................... .119Actinopterygii ................................................................................................................ 120Testudines .................................................................................................................... ..120Serpentes ..................................................................................................................... ...121Aves ...............................................................................................................................121Didelphidae ................................................................................................................... .121Dasypus novemcinctus ..................................................................................................122Chiroptera .................................................................................................................... ..122Canidae ....................................................................................................................... ...123Artiodactyla .................................................................................................................. .123Tayassuidae ................................................................................................................... 124Cervidae .........................................................................................................................124Rodentia ...................................................................................................................... ...125Agoutidae and Dasyproctidae ........................................................................................126Naj Tunich ............................................................................................................................126 7

PAGE 8

Crustacean .................................................................................................................... .128Testudines .................................................................................................................... ..128Aves ...............................................................................................................................128Didelphidae ................................................................................................................... .129Dasypus novemcintus ....................................................................................................130Chiroptera .................................................................................................................... ..130Primates ...................................................................................................................... ...131Canidae ....................................................................................................................... ...131Felidae ....................................................................................................................... ....131Procyonidae ................................................................................................................... 132Tapirus bairdii ...............................................................................................................132Artiodactyla .................................................................................................................. .133Tayassuidae ................................................................................................................... 133Cervidae .........................................................................................................................134Rodentia ...................................................................................................................... ...135Agoutidae/Dasyproctidae ..............................................................................................135Sylvilagus sp. .................................................................................................................136Summary ....................................................................................................................... ........1367 INTERPRETATIONS ..........................................................................................................155Introduction .................................................................................................................. .........155Limitations to Spatia l Pattern Analysis ................................................................................156Taxonomic Trends .............................................................................................................. ..159Crustaceans ................................................................................................................... .160Ray-Finned Fishes .........................................................................................................162Amphibians .................................................................................................................... 163Turtles ....................................................................................................................... .....164Lizards ....................................................................................................................... ....166Snakes ........................................................................................................................ ....167Birds ......................................................................................................................... .....167Opossums ...................................................................................................................... 168Armadillos .................................................................................................................... .169Bats .......................................................................................................................... ......170Primates ...................................................................................................................... ...171Canids ........................................................................................................................ ....171Felids .............................................................................................................................172Raccoons ...................................................................................................................... ..173Tapirs .............................................................................................................................174Artiodactyls .................................................................................................................. .174Peccaries ..................................................................................................................... ...174Deer .......................................................................................................................... .....175Rodents ....................................................................................................................... ...176Squirrels ..................................................................................................................... ....177Agoutis and pacas ..........................................................................................................178Cottontail rabbit .............................................................................................................178Summary ....................................................................................................................... ........178 8

PAGE 9

8 CONCLUSIONS ................................................................................................................. .200Trends ........................................................................................................................ ...........200Limitations ................................................................................................................... .........202Recommendations ............................................................................................................... ..206Summary ....................................................................................................................... ........206LIST OF REFERENCES .............................................................................................................208BIOGRAPHICAL SKETCH .......................................................................................................219 9

PAGE 10

LIST OF TABLES Table page 4-1 Five Cave Site Information Includin g Proximity to Sites, Time Periods, and Excavation Methods...........................................................................................................665-1 Gentax Numbers for Taxonomic Cl assification of Faunal Remains .................................775-2 Element Numbers for Element Types ................................................................................865-3 Body Portion Numbers for Body Portions .........................................................................875-4 Sidedness Numbers for Element Sidedness .......................................................................875-5 Age Class Numbers for Age Classes .................................................................................875-6 Burning and Charring Numbers for Burning and Charring Descriptions ..........................876-1 Relative Frequency Values for Five Cave Sites. .............................................................1506-2 Separation of Space for Caves Branch Rockshelter, Belize ............................................1516-3 Separation of Space for Stela Cave, Belize .....................................................................1516-4 Separation of Space for Cueva de El Duende, Guatemala ...............................................1526-5 Separation of Space for Cueva de Sangre, Guatemala ....................................................1536-6 Separation of Space for Naj Tunich, Guatemala .............................................................1547-1 Separation of Space Summaries for Left versus Right Sides and North versus South Directions at Caves Branch Rockshelter, Belize. ............................................................1807-2 Separation of Space Summaries for Light ve rsus Dark and Open versus Restricted Regions at Caves Branch Rockshelter, Belize. ................................................................1817-3 Summary of NISP, % Taxa, Left Sided Elements, Body portions, % NISP Burned, Spatial Auto Correlation and Cokriging at Caves Branch Rockshelter, Belize. ..............1827-4 Separation of Space Summaries for Left versus Right Sides, North versus South, East versus West Directions Light versus Dark, and Open versus Restricted Regions at Stela Cave, Belize. .......................................................................................................1837-5 Summary of NISP, % Taxa, Left Sided Elements, Body portions, % NISP Burned, Spatial Auto Correlation and Cokr iging at Stela Cave, Belize. .......................................1847-6 Separation of Space Summaries for North ve rsus South Directions at Cueva de El Duende, Guatemala. .........................................................................................................185 10

PAGE 11

7-7 Separation of Space Summaries for East versus West Directions, Light versus Dark, and Open versus Restricted Regions at Cueva de El Duende, Guatemala. .....................1867-8 Summary of NISP, % Taxa, Left Sided Elements, Body portions, % NISP Burned, Spatial Auto Correlation a nd Cokriging at Cueva de El Duende, Guatemala. ................1877-9 Separation of Space Summaries for Left versus Right Sides at Cueva de Sangre, Guatemala. .................................................................................................................... ...1897-10 Separation of Space Summaries for North versus South Directions at Cueva de Sangre, Guatemala. ..........................................................................................................19 07-11 Separation of Space Summaries for East vers us West Directions at Cueva de Sangre, Guatemala. .................................................................................................................... ...1917-12 Summary of NISP, % Taxa, Left Sided Elements, Body portions, % NISP Burned, Spatial Auto Correlation a nd Cokriging at Cueva de Sangre, Guatemala. ......................1927-13 Separation of Space Summaries for Left versus Right Sides at Naj Tunich, Guatemala. .................................................................................................................... ...1947-14 Separation of Space Summaries for North versus South Directi ons at Naj Tunich, Guatemala. .................................................................................................................... ...1957-15 Separation of Space Summaries for East versus West Directions at Naj Tunich, Guatemala. .................................................................................................................... ...1967-16 Separation of Space Summaries for Light versus Dark Regions at Naj Tunich, Guatemala. .................................................................................................................... ...1977-17 Summary of NISP, % Taxa, Left Sided Elements, Body portions, % NISP Burned, Spatial Auto Correlation and Cokriging at Naj Tunich, Guatemala. ...............................198 11

PAGE 12

LIST OF FIGURES Figure page 4-1 General Location of Caves Sites in Belize and Guatemala. ..............................................604-2 Location of Operations and Units fo r Caves Branch Rockshelter, Belize .........................614-3 Location of Units for Stela Caves, Belize ..........................................................................624-4 Location of Surface Collections for Cueva de Sangre, Guatemala ...................................634-5 Location of Excavation Units for Cueva de El Duende, Guatemala .................................644-6 Location of Lots for Operat ion IV, Naj Tunich, Guatemala ..............................................656-1 Separation of Space at Caves Branch Rockshelter, Belize ..............................................1376-2 Separation of Space at Stela Cave, Belize .......................................................................1386-3 Separation of Space at Cuev a de El Duende, Guatemala ................................................1396-4 Separation of Space at Cueva de Sangre, Guatemala ......................................................1406-5 Separation of Space at Naj Tunich, Guatemala ...............................................................1416-6 Caves Branch Rockshelter, Crustacean Remains Ordinary Cokriging Light versus Dark Regions ...................................................................................................................1426-7 Caves Branch Rockshelter, Testudines Remains Ordinary Cokriging Light versus Dark Regions ...................................................................................................................1436-8 Caves Branch Rockshelter, Serpentes Remains Ordinary Cokriging Light versus Dark Regions ...................................................................................................................1446-9 Cueva de Sangre, Testudines Remains Ordi nary Cokriging Left versus Right Sides .....1456-10 Cueva de Sangre, Cervidae Remains Or dinary Cokriging North versus South Regions ....................................................................................................................... .....1466-11 Cueva de Sangre, Rodentia Remains Ordinary Cokriging Light versus Dark Regions ..1476-12 Naj Tunich, Aves Remains Ordinary Cokriging Light versus Dark Regions ..................148 12

PAGE 13

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Arts EVALUATING CAVE USE THROUGH SPATIAL ANALYSIS OF ANIMAL REMAINS FROM MAYA CAVES IN GUATEMALA AND BELIZE By Erol George Kavountzis December 2009 Chair: Kitty F. Emery Major: Anthropology Scholars suggest caves in the Maya region were used for ritual activities and represented a connection to the underworld (Brady 1989; Stone 1995). Recent studies (Moyes 2001, 2002, 2004) use spatial and GIS analysis of artifact di stributions to understa nd the role of spatial cognition in Maya ritual cave use based on directionality. Other st udies suggest a variety of cave ritual functions and several c ognized spatial divisions within caves such as north-south, eastwest, right-left, light-dark, and restricted-o pen spaces. This zooarchaeological study uses Geographic Information Systems (GIS) to test the spatial distribution of animal remains within multiple cave sites in Belize and Guatemala. Thes e cave sites include Caves Branch Rockshelter and Stela Cave in Belize and Cueva de El Du ende, Cueva de Sangre, and Naj Tunich in Guatemala. Patterning is addressed using both spa tial analysis tools from ArcGIS and a visual analysis of animal remains by taxonomic groups. While spatial patterning is present for key taxonomic groups of animal remains, overall patterning of skeletal remains is rare due to sm all sample sizes, limited distribution of remains, taphonomic processes, the use and placement of anim al offerings, the shape of the caves, and the possible differential use of caves within the Maya region. If used properly, GIS coupled with 13

PAGE 14

14 with zooarchaeological analysis can provide a new method for i nvestigating the use of animal remains in cave contexts.

PAGE 15

CHAPTER 1 INTRODUCTION To the ancient and modern Maya, caves re present entrances into the underworld or Xibalba, and are places where various ritual events are carried out. Recent cave archaeology (Awe 1998; Brady 1989; Stone 1995) a ttests to the ritual nature of caves in the Maya world and demonstrates that these locations did not primaril y serve as dwelling places, refuges, or middens. Therefore, these features represent a known ritu al context and the arch aeological materials found within them can be assumed to represent, at le ast in large part, the activities associated with ancient ritual events. I use the zooarchaeological remains from five cave sites in Guatemala and Belize to improve our understanding of anci ent Maya rituals. We know from ethnographic, ethnohistoric, and iconographic records that one of the components of ritual activ ities was the use of animals as sacrifices and offerings (e.g., Brown 2005; Br own and Emery 2008; Pohl 1983; Tozzer 1941). The Maya codices present ritual events with offe rings of animal bread, or tamales, and animal parts (Bricker 1991; Taube 1988; Thompson 1972). Ma ny of these events ar e associated with caves (Bricker 1991; Stone 1995). Recent ethnogra phic and ethnoarchaeological studies show a close association between cav es/rockshelters and hunting r ituals conducted by contemporary Maya hunters living around Lake Atitlan in Guatemala which highlights the continued importance of animals and caves in Maya rituals (Anderson 2009; Brown 2005; Brown and Emery 2008). The writings of Friar Diego de Landa dated to 1566 contain accounts of seasonal and annual rituals which included the sacrificial offerings of animals associated with various landscape features including the cardinal directions and possibly caves (Tozzer 1941). Iconographic and epigraphic research has identified depictions and writing of caves and animal offerings in the codices and carvings of the an cient Maya. Some authors have suggested that 15

PAGE 16

these too can be linked to modern day even ts (Brady 1989; Pohl 1981, 1983; Taube 1988; Vail 1997). I have chosen to evaluate the possible connect ion between the spatial distribution of animal remains within Maya caves and the ancien t Maya worldview or cognized landscape. Ethnographic and ethnohistoric documents show that the contemporary Maya worldview is defined on the basis of various spatial patterns including the layering of the universe, cardinal directions, and sidedness (Brown 2004; Garcia-Zambrano 1994; Mathews and Garber 2004; Moyes 2001, 2002, 2004; Palka 2002; Pohl 1983). Im portantly, Maya ritual activities and behaviors are also based on these spatial divisions, so the remain s of these activities should be patterned in ways that duplicat e the Maya worldview. Many rese archers have argued that this same worldview was held by the ancient Maya as well and that analysis of archaeological spatial patterning, particularly in caves, can reveal ritual behaviors associated wi th the cognized spatial landscape of the ancient Maya (Brady 1989; Moyes 2001, 2002, 2005). In this thesis, I analyze the caves as ritual contexts and study the distribution of animal remains using GIS and visual analysis to identify spatial patterns related to the Maya worldview. I examine several dichotomized relationships th at have been shown by archaeologists to be replicated in artifact dist ributions in both cave and a boveground sites (Moyes 2001, 2002, 2004). These include sidedness or the left versus right sides of a space, cardinal directionality or the separation of space into northern versus southern and eastern versus western spaces, light versus dark spaces, and open versus restricted spaces. By analyzing the frequency of different taxonomic groups and skeletal propor tions in these specific spaces w ithin the cave, I reveal some preliminary patterns in animal use that reflect the ancient Maya worldvi ew and use of ritual space in caves. 16

PAGE 17

This research project was conduc ted in a total of five cave sites, two located in Belize and three located in Guatemala. In 1997, Dr. Jaime Awe (1998) star ted the Western Belize Regional Cave Project (WBRCP) which investigated multi ple cave sites around the modern town of San Ignacio. Two of these sites, Caves Branch Rockshelter and Stela Cave, are included in this study. Caves Branch Rockshelter (CBR) is located in the Cayo District of western Belize along the Caves Branch River Valley and the faunal assemblage is from the excava tion of the site during the summers of 2005 and 2006 by Dr. Gabriel Wrobel (Wrobel and Tyler 2006; Wrobel 2008). CBR is a known cemetery site that dates to th e Late Preclassic to Early Classic periods (Glassman and Bonor Villarejo 2005; Wrobel an d Tyler 2006; Wrobel 2008). Stela Cave (STC) is also located within the Cayo District along the Macal River Valley of western Belize and the faunal assemblage was from the 2004 season wh ich was excavated by Ph.D. student Cameron Griffith (Ishihara and Griffith 2004). STC dates from the Preclassi c to Late Classic periods and contains a large architectural feature in a back chamber that ma y have served as an altar or platform (Ishihara and Griffith 2004). Within Guatemala, two of the sites, Cueva de Sangre and Cueva de El Duende, were located at the site of Dos Pilas and were exca vated as part of the Petexbatun Regional Cave Survey by Dr. James Brady in the mid-to late-1990's under the supervision of the project director Dr. Arthur Demarest (Brady et al. 1997; Minjares 2003; Scott 1995). Cueva de Sangre is a long (over 3.5 km), multiple tunnel system that is lo cated under a hill about 3 km south of the El Duende Pyramid Complex. It dates to the Late Preclassic to the Late Classic periods (Minjares 2003). Cueva de El Duende is a collapsed cave w ith a single long tunnel system located just southwest of the El Duende Pyramid Complex and it dates to the Late Preclassic to Early Classic 17

PAGE 18

periods (Brady and Rodas 1992). Preliminary z ooarchaeological research was conducted by Kitty Emery in 1998. Emery's identificat ions are included in this study. There is one published source used for this an alysis, the site of Naj Tunich excavated by James E. Brady (1989) during the early to mid-1980s. Naj Tunich is the largest known cave site in the Maya region and may have served as a pilg rimage site. The ceramics date the site from the Late Preclassic to the Late Classic period (Brady 1989). The faunal remains from Naj Tunich were identified by Susan Colby at UCLA in 1984 and these identifications were included in this analysis. The description and place ment of excavation units and type s of analysis for these sites is presented in further detail in Chapter 4, Re search Setting: Five Caves from the Maya Region under Analysis. Cognitive Archaeology The idea that spatial patterning in archaeol ogical remains may refl ect ancient behaviors informed by a cognized landscape is a basic te net of cognitive archaeology and, in many ways, landscape archaeology. Landscape archaeology, simply identified as the study of the social and natural influences of a peoples landscape, has allowed researchers to move beyond economic studies of land use and into studies of the social and sacred influences of both the natural and built environment (Knapp and Ashmore 1999; Bra dy and Ashmore 1999). Cognitive archaeology is another field of archaeology th at provides an important theoreti cal overlay to these landscape studies. Cognitive archaeology can be defined as a basis for reconstructing the ancient mind by using the material cultural to understand the beha viors that resulted in the placement of these remains (Flannery and Marcus 1993, 1996, 1998; Re nfrew 1994; Renfrew and Zubrow 1994). Technological advances in spatial archaeology, such as the use of Geographic Information System mapping (GIS), has also aided in the de velopment of these studies of the ancient Maya landscape and mind. Maya landscape archaeologists have spent much time trying to understand 18

PAGE 19

the Maya ritual use of underground sites, caves, and cenotes, or water-filled sink holes, because these geological features are and were such an important element of the sacred Maya landscape (Ashmore 1989, 1991; Brady 1997; Brady and As hmore 1999). Maya cave archaeology, a more recent addition to archaeology in this region, has helped to solidify our understanding of the ancient Maya ritual us e of these openings to the underworl d. These theoretical underpinnings are further discussed in Chapter 2, Theoretical Framework: Caves, Landscape, Cognitive Archaeology, and the Use of Space by the Ancient Maya. Maya Cave Archaeology: Maya cave archaeology is a new field with a long history. Dating back to the 1840s archaeologists have recorded and documented cav e sites within the Maya region (Brady 1989). However, it was not until recen tly that the importance and ritual significance of these sites became apparent to Maya archaeologists. The Maya region is located within a limestone karstic environment that includes many ge ological features that are cut into the landscape through water degradation. In the Yucatn region or the nor thern Maya lowlands, the relatively flat dry environment contains open water-holes througho ut the landscape; these water-holes are the cenotes of the ancient Maya. Cenotes served as both water sources and sacred entrances into the underworld. The southern Maya lowlands and hi ghlands include areas of mountainous and flat tropical forest with a plethora of caves and ot her rock openings throughout the landscape. The ancient Maya viewed any entrance into the landscap e as a sacred opening into the earth and as a connection to the underworld (Anderson 2009; St one 1995). The Maya worldview is based on layering, directionality, and sidedness, and are f ound to occur in many ritual contexts at above ground sites in large scale events (Brown 2004; Mathews and Garber 2004; Palka 2002). Caves have been shown to represent smaller versions, or microcosms, of the larger Maya worldviews (Moyes 2001, 2002, 2004. For example, the quadripartit e placement of artifacts at Actun Tunich 19

PAGE 20

Muknals Main Chamber (Moyes 2001, 2002, 2005) may be a representation of the large scale foundation rituals for setting territori al borders (Garcia-Zambrano 1994). Current research in Maya cave archaeology has m oved from site-level analyses of cave use to a more regional approach to understanding ho w and why the Maya accessed and used these landscape features (Awe 1998). Some cave sites are hypothesized to have been ritual pilgrimage sites (e.g., Naj Tunich, Brady 1989), while others were likely elite controlled caves directly linked to a surface site or polity (e.g., the Dos Pilas caves including Cueva de Sangre and Cueva de El Duende) (Brady 1997; Brady et al. 1997). Finally, recent studies have begun to look at multiple cave systems (such as Caves Branch Rockshelter and Stela Cave) to see who was accessing and using these smaller cave groups and for what purposes (Awe 1998; Peterson 2006). All of these studies have help ed to form a new perspectiv e on our understanding of ancient Maya rituals because ritual activities which ar e often difficult to define at surface sites and within non-elite or commoner spaces are more readily defined in a cave context. There are some known difficulties in working with in cave sites, including the ta phonomic movement of artifacts, both natural, including animal and water movements within the cave, and cultural, in the form of looting. In addition, stratigra phy and chronology are often problematic because most cave archaeology projects concentrate on mapping and collecting surface artifacts rather than excavation. This is due to the la ck of clear stratigraphy in caves. Maya cave archaeology is further discussed in Chapter 3, Prior Research in Maya Cave Archaeol ogy, Zooarchaeology, and GIS. Maya Cave Zooarchaeology: Zooarchaeological research has also had an evolving ro le in Maya archaeology. As described by Emery (2004a), faunal remains from the excavations in the 1940s to 1980s were 20

PAGE 21

collected and analyzed by zoologists to produ ce a laundry list of animal s from archaeological sites. This practice of simply listing animals limited the role of information that could be collected from zooarchaeological materials to qu estions of subsistence a nd the types of animals used. Animals were therefore assigned the role of footuff in archaeological research during this period. However, there was a push for change duri ng the post-processual m ovement to revise our understanding of animals in societies (Emery 2004a; Pohl 1983). Zooarchaeology went from being defined as the use of animals for subsiste nce to the study of th e interactions between humans and animals (Emery 2004a). This included the role of animals in ritu al and also as sacred beings in many societies (e.g. Pohl 1983). Maya cave zooarchaeology has also gone through the same evolution. Early cave studies included a short laundry list of anim al species within the cave with little to no reference to the role of these animals in ritual or their pl acement within the caves (e.g. Pendergast 1969, 1971). However, after Mary Pohls ( 1983) ground breaking work on the r itual use of animals in the Maya region, specifically in caves and cenotes, research ers began to look into zooarchaeological materials within these settings. James Bradys (1 989) dissertation on Naj Tunich, was one of the first Maya cave archaeology projects to address both the role of caves in the ritual patterns of the Maya and the role of animal remains within this specific setting. Recent cave work included faunal analysis and interpretati ons in their research (Brady et al. 1991; Ishihara 2007; Peterson 2006). Few studies, however, have sp ecifically evaluated the ritual use of animals at cave sites and examined the patterning and meaning of th eir distribution (for an exception see Emery 2004c). This has left open the opportunity to study faunal remains in cave archaeological contexts in order to test ideas and patterns for th e ritual use of animals w ithin caves. The role of 21

PAGE 22

animals within ancient Maya culture and the changes in the study of faunal remains are discussed in detail in Chapter 3 Prior Research in Maya Cave Archaeology, Zooarchaeology, and GIS. Geographic Information Systems and Maya Caves Geographic Information Systems (GIS) has b ecome a very important tool in Maya archaeology and Maya cave archaeology in particular where artif act spatial dist ributions over cave surfaces are of great interest. GIS allows researchers to map at multiple scales, from the site to the region, and allows them to look for pa tterning within and between excavation units at these sites. Spatial distinctions have always been an important part of archaeological research, but with the addition of GIS, archaeologists are now able to store more information into georeferenced databases and test for statistically-significant patterning. One of the most important GIS papers in the Maya region identified spatial patterning in artifact distributions (ceramics, lithics, grounds tones, faunal remains, broken speleothems, and slate) that mirrored th e Maya worldview. Holley Moyes (2 001, 2002, 2005) was able to show a quadripartite distribution of artifacts within the Main Chambe r at Actun Tunich Muknal in Belize. A more recent GIS-based cave study at El Miron Cave in Spain included faunal remains and helped to answer questions about the s easonal use of caves (Marn Arroyo 2009). Although Marn Arroyos (2009) study took place outside of the Maya regi on, it shows the importance of analyzing the spatial distributi on of animal remains within a cave site. The use of GIS in archaeology and cave archaeology is explored in full detail Chapte r 3 Prior Research in Maya Cave Archaeology, Zooarchaeology, and GIS. Maya Spatial Patterns, Zooarchaeology, and GIS Zooarchaeological analysis included the iden tification of taxonomic groups, elements, sides of elements, age characteristics, and both cultural and natural effects on the faunal materials. For this study, the Number of Id entifiable specimens (NISP) was used for the 22

PAGE 23

quantitative analysis of remains. The GIS analysis included using two spa tial statisti cal methods, spatial autocorrelation and cokr iging, to identify pa tterns for the faunal remains by taxonomic groups. Due to the small sample sizes and lim ited distribution of so me taxonomic groups, a visual analysis was also perfor med. This visual analysis helped to identify some patterns that were not seen using GIS. In Chapter 5, Zooa rchaeology, GIS, and Visu al Analysis Methods, includes a further discussion in how these cav e sites were analyzed and their findings incorporated into this study. The results from the zooarchaeological, GIS, and visual analysis are presented for each individual cave by taxonomic groupings. The ability to present the data in this form allows the visualization of patterning for the placement of species, elements, and element siding to become apparent for some taxonomic groups. A further and more in-depth presentation of these findings can be found in Chapter 6. There were some limitations identified during the analysis of faunal remains and their distributions within the sites. These limitations include small sample sizes, limited distributions, differential collection me thodologies, natural and cultural taphonomic, differential cave shapes, and differential uses of these caves. The limitations are described in greater detail in Chap ter 7, Interp retations. The evaluation of zooarchaeologi cal spatial distributions using GIS revealed that there is some patterning of crustacean, turtle, squirrel, peccary, and deer remains. There is a lack of specific patterning for ray-finned fishes, amphi bians, snakes, birds, opossums, armadillos, primates, tapir, raccoons, rodents, agoutis and p acas, and cottontail rabbits. Finally, there is also a lack of patterning for some taxonomic groups that were expected to have close affiliations with either caves or who played a major role in ritu al offerings including liza rds or iguanas, bats, canids, and felids remains. Although there were no definitive patterns iden tified throughout all of 23

PAGE 24

the caves, the finding of some patterns indicates th at there were possibly some close connections between the placement of faunal remains with in the caves and the Maya worldview and delineation of space. This study provides a baseline for the analysis of spatial distributions of faunal remains in both cave and surface sites in the Maya region. The use of GIS along with the expansion of collection areas and consistent collection methods will produce detailed data pertaining to the spatial distribution of faunal remains within cave c ontexts. There is also a need for researchers to continue the excavation of many different cave si tes within a multitude of contexts. The more information and the storage of this informati on into geographically referenced databases may allow for a more ideal analysis of patterning of faunal remains with Maya cave sites. The collection and analysis will also allow for larger co llections of materials within caves that can be better analyzed using statistical tools availa ble in GIS programs. Chapter 8, Conclusions, provides a more through explanatio n of the patterns and future us es of GIS and zooarchaeology within Maya cave archaeology. Summary In this research project I analyze the fauna l assemblages from five Maya cave sites in Belize and Guatemala and reconstruct their spatia l distribution using GIS and visual assessment to answer questions about th e connection between the ancient Maya worldview, sacred landscape, ritual activit ies, and the use of animals in cav es. In this study faunal remains add another element to the recent analys es of spatial patterns in the material record of caves. Animals are closely associated with how the Maya viewed the world and these associations add another element to the understanding of cave use by the an cient Maya. I suggest that using GIS to map, analyze, and identify the pattern of an imal remains adds another method by which zooarchaeologists and archaeologists can conceptu alize and identify ritual and secular practices 24

PAGE 25

25 among an ancient people. Caves represent an impo rtant part of Maya archaeology and the use of them for rituals and animal offerings needs to be continually explored.

PAGE 26

CHAPTER 2 THEORETICAL FRAMEWORK: CAVES, LANDSCAPE, COGNITIVE ARCHAEOLOGY, AND THE USE OF SPACE BY THE ANCIENT MAYA Overview This research project relies on the theo retical framework of cognitive archaeology. Cognitive archaeology is defined as the study of mate rial cultural and its placement within a site as a vector to understanding ancient behaviors of humans in an attempt to reconstruct the ancient mind (Flannery and Marcus 1993, 1996, 1998; Renfrew 1994; Renfrew and Zubrow 1994). I also draw on elements of landscape ar chaeology theory in unders tanding the relationship between cognized worldviews and the use of space across a landscape. Landscape archaeology is defined as the study of the soci al and natural influences of humans on the landscape and it has allowed researchers to move beyond the economics of land use to include the social and sacred influences of the natural and built environm ent (Knapp and Ashmore 1999; Brady and Ashmore 1999). In this study, I compare spat ial patterns in archaeological animal remains across ritual landscapes (caves) with the ethnographical ly and ethnohistorically understood Maya worldview or cognitive map. In this chapter, I review the approach es of cognitive and landscape archaeology and discuss the evidence for an ancient Maya cognized landscape. Over the last few decades there has been a sh ift in how archaeologist s interpret artifacts and archaeological remains. After the 1960s, th e focus of archaeology was on science and the use of the scientific method in answering questions involving economic, subsistence, and other measurable aspects of society (Binford 1962; Trigger 2006). During this time, New Archaeology identified three forms of mate rial culture: the technomic or economic and subsistence technology, the socio-technic or soci al aspects including ki nship and status, and the ideo-technic which included religion a nd ideology (Binford 1962:217). The ideo-technic was identified as the unanswerable parts of religion, ideology, symbol ogy, and cosmology which 26

PAGE 27

were considered unavailable to archaeologists using the material culture we analyze (Binford 1962; Moyes 2001). The human mind was considered inaccessible in the archaeological record. More recently, archaeologists ha ve begun to approach these unanswerable issues. Using the theoretical frameworks of cogn itive and landscape archaeology, re searchers have attempted to link archaeological spatial patterning to ancient religion, ideology, cosmology, and symbology. In Maya archaeology, links between the cognized landscape and cosmology are an important part of our understanding of the use of space in architectural placement across communities (Ashmore 1989, 1991), within natural landscapes (Brady 1997; Mathews and Garber 2004) and even in representations of hu man interactions (Palka 2002). In the sections of this chapter, I define and review cognitive and landscape archaeology and discuss their roles in unders tanding links between the cognized space of the ancient Maya, their spatial use of la ndscapes, and their ancien t religious beliefs, id eology, and cosmology. I then define the Maya worldview and the specific spaces demarcated by the Maya in architecture, daily life, and the Maya codices. This chapter will form the ba sis for understanding the possible use of cognized space and its reflection in the di stribution of animal remains from Maya cave sites. Maya Cognitive and Landscape Archaeology Cognitive archaeology was redefined by multip le authors over the last two decades (Flannery and Marcus 1996, 1998; Preucel 2006 ; Renfrew 1994). Flannery and Marcus (1996:351, 1998:36) define it as the study of all those aspects of ancient culture that are the products of the human mind which they separa ted into groups that in clude cosmology, religion, ideology, iconography, and other symbolic beha viors. Renfrew (1994:3) defines cognitive archaeology as the the study of pa st ways of thought as inferred from the material remains. All agree, however, that the main focus is the an cient mind and the links between symbolism and 27

PAGE 28

material artifacts, placing empha sis on the intentionality of artif act placement and its reflection of an ancient worldview (F lannery and Marcus 1993, 1996, 1998; Renfrew 1994; Renfrew and Zubrow 1994). Rituals are more easily identifiable in the ar chaeological record than concepts of religion or ideology because they leave material signatu res within the archaeological record (Fogelin 2007). Cognitive archaeology allows for researchers to move beyond the habitual patterns formed by ritual events to understand the specific religious and ideologica l reasons for behaviors. The Maya, in particular, use space and modify space in ways that reflect their worldview and, therefore, their religion and ideology. As peopl e use and modify space, they leave marks and remains that we can interpret as archaeologists. Therefore, the material remains found in the archaeological record can be used to approach the ancient mind. The Maya were very intentional in their placement of artifacts w ithin caves and this can be seen in the placement of large ceramics, both whole and broken, on both high and low inaccessible parts of the caves (Stone 2005). The ability to understand the Maya worldv iew has allowed for a cognitive archaeological reconstruction of the ancient Maya mind. An important approach to cognitive studies is through landscape archaeology, or the study of the human landscape. Landscapes have always been a part of what archaeologists study (Anschuetz et al. 2001; Knapp and Ashmor e 1999); however, our perspectives on how landscapes are formed, used, and identified by ancient people have ch anged. The concept of landscape as neutral and passiv e (Bender 2002:323) has been rej ected as archaeologists began to think of landscapes as active agents in ancient worldviews (Brady and Ashmore 1999). Humans cultivate, build, and change their lands cape over time and these changes do not always reflect simply the requirements of basic human subsistence and shelter. Instead their use and 28

PAGE 29

modification can reflect such epiphenomenal c oncepts as religion and sacredness, ethnicity, ownership, and even connections to the past and ancestry. Landscape archaeology is a movement away from a simplistic understa nding of the ancient human-land relationship in terms of landuse, subsistence, economics, and politics. Instead landscape archaeology considers the more complex and dynamic relationship between people and landscapes by also including the social, symbolic, and sacred values of landscapes (Knapp and Ashmore 1999). The Maya are known to have perceived cosmol ogical links between their natural and built landscapes that were manifest in domestic, civi c, and wider spatial scales (Brady and Ashmore 1999:126). Wendy Ashmore (1989, 1991, 2002)is one of the main proponents of the close connection between Maya land-us e and worldview. She has shown that the directionality and layering of ancient Maya built landscapes reflec t a pan-Mesoamerican cosmology. For example, temple-pyramid complexes are considered to re present the duality of cave and mountain (Brady 1997; Stone 1995; Vogt and Stuart 2005), while the quadripartite plaza group represent the four quarters of the universe (Ashmore 1989, 1991; As hmore and Sabloff 2002; Coggins 1980; Stone 2005). Similarly, on a smaller scale, offerings in Maya households to the four corners and center represent the division of the wo rld into four cardinal quarters plus the center or axis mundi (Brady 1991, 1997; Brady et al. 1997 ; Garcia-Zambrano 1994; Pugh 2005). The Cognized Universe and Caves In Maya cosmology, caves repres ent a particularly important landscape feature that defines them as intrinsically ritual, power ful, and symbolically potent. Here I discuss some aspects of the Maya worldview that relate directly to caves and associated landscape features in both the ethnographic and archaeological reco rd. Directionality and layering are considered the two most important frameworks within the worldview, of the ancient Maya. An emphasis was placed on the separation of space within the horizontal a nd vertical planes, respectively (Coggins 1980; 29

PAGE 30

Kunen et al. 2002; Mathews and Garber 2004; St one 2005). The Maya universe is partitioned into three main layers, including the upper-world that is made up of thirteen layers, the earth, and the underworld consisting of nine layers (Mathe ws and Garber 2004). The directional division of the Maya worldview is defined as a quadripart ite separation of space related to the cardinal directions. All of the dir ections meet at a center poi nt, termed the axis mundi which is associated with both caves and the ceiba tree (Brady 1991, 1997; Coggins 1980; Mathews and Garber 2004; Pugh 2005). The ceiba tree is believed to be the connection between the three world layers, by holding up the sky, or upper-world, and having its roots reach deep into the ground, or underworld. Caves are identified as another connection point be tween both the horizontal and vertical planes of the Maya worldview. Highlighting the work of Eliade, Brady (1991:6) defines the axis mundi as the universe the most sacred of places, a place of prestige a place of inexhaustible abundance the spot where three levels of the universe meet the crea tion of man took place. This idea of center is incorporated in the patterning of the settlement s and within the landscape of the ancient Maya (Brady and Ashmore 1999). Caves are described as the center of the universe because they are located at the inte rsection of the thre e layers of the world (Brady 1991). Caves are located within the mountains which are associated with the upp er-world. The mouth of the cave is the middle layer or the earth. It represents the condu it between the earth a nd the underworld (or Xibalba ). Besides representing the center of the Maya uni verse, caves are also associated with the underworld (Stone 1989), water (Prufer and Kindon 2005), the home of clouds and rain (Ishihara 2008; Vogt and Stuart 2005), human and agricultur al fertility (Stone 1995 ), ancestry and place of origin (Pugh 2001; Vogt and Stuart 2005), politic al and social power (Halperin 2005), and as places for ritual events (Brady and Prufer 2005b; Prufer and Brady 2005b). 30

PAGE 31

Ethnographic research can also help in solidifying our unders tanding of the role of caves and cenotes as centers in the Maya worldview. In early Colonial records, natural features including mountains with caves and water-holes play a large role in the foundation rituals for setting territorial bor ders (Garcia-Zambrano 1994). To fulfill the spatial requirements, settlements were placed in relation to five mount ains, one for each cardinal direction and a center point (Garcia-Zambrano 1994). Ethnographic re search on modern Maya communities found similar associations between caves and settle ment locations. Tzeltal Maya communities are situated in relation to sacred caves from which they derive their community name (Brady 1991:1, 1997; Vogt and Stuart 2005). Integrating the sacred landscape into identification of place and person, the Tzotzil Maya communities are located around waterholes, along with caves, and individuals take their surnames from th ese water hole or cave features (Brady 1991:2, 1997). Ethnographic studies in the highlands of Guatemala also f ound that caves and rockshelters are still used as hunting shrines and places of communication between the Maya and the Guardian of the Animals (Brown 2005:137-138; Emery and Brown 2008). In recent years, archaeologists have begun to research caves within the Maya region and many studies found that the caves and cenotes play an important role in the placement of sites (Pugh 2005; Brady 1997; Brady et al. 1997). The cav e and cenote as the axis mundi of the archaeological site can be found in both the sout hern and northern lowlands. Within the northern lowlands of the Yucatan peninsula, many archaeol ogical sites wereassociated with and connected to a cenote (Pugh 2005) including Mayapan (Brown 2005) and Chichen Itza (Brady 1997). In the southern lowlands, at the site of Dos Pilas large public architect ure wasconstructed on top of an extensive cave system (Brady 1997; Brady et al. 1997). Archaeologists argue that the placement 31

PAGE 32

of elite architecture in relation to cave features is directely related to the ideological foundations of elite sociopolitical rule and control. Outside and within the Maya region associations are found between large architectural monuments and man-made caves. In the Azt ec region, under the Pyramid of the Sun at Teotihuacan, Heyden (1975:131) identified a manmade cave that was found to be associated with the mythical Chicomoztoc, Seven Caves, [or] place of creati on in ancient Mexican mythology. Other man-made caves have been iden tified at other sites (Brady 1991) within both the Mexican and the Maya regions. This Pan-Mesoamerican occurrence makes the use of caves as important in the association of civic and el ite architecture (Brady 1991). Nevertheless, not all caves in the Maya region are f ound in relation to arch aeological site and site construction. These unassociated caves may have represented pilg rimage sites, for example Brady (1989, 1991) suggests that Naj Tunich and Cueva de las Pinturas were used by the ancient Maya but do not have settlements asso ciated with them. Caves in the Maya region have also been associ ated with calendrical rites, specifically the New Year and Ueyab (Wayeb) ceremonies, and also as places for the katun ending celebrations (Pugh 2001; Stone 1989, 1995, 2005). These calendrical asso ciations may be due to the fact that caves are viewed as the place of primordial em ergence of ancestors and places of origin (Pugh 2001). The ethnographic record also finds that within some modern Tzotzil and Tzeltal Maya groups in the highlands of Chiapas annual ceremonies are still performed at caves within the region (Vogt and Stuart 2005). Th ese rituals have been influenced by Christian ideals and images, but they do show a historical connec tion between the ancient and modern-day Maya cave rituals (Vogt and Stuart 2005). 32

PAGE 33

Separation of Space and the Use of Space by the Ancient Maya The use of space, particularly ritual space, is directly linked to worldview. In his early work at Naj Tunich, Brady (1989) acknowledged the fact that caves were not physically uniform and that these differences in cave formations were important to the ancient Maya. The ancient Maya undoubtedly considered each cave a separa te landscape within which the worldview could also be symbolically expressed. Here I review the spatial separations th at are of particular interest in this study of the cave landscape. Cardinal Directions and Quadripartite: Along with other Mesoamerican groups, the Maya associate the world, or their community, or themselves with the center of the cognized un iverse and from this center extend points in all four cardinal directions creating the quadripart ite divisions of the world (Garcia-Zambrano 1994; Kunen et al. 2002; Mathews a nd Garber 2004). The Maya direct ions begin in the east and move counterclockwise to the north, west, a nd south. The glyphical interpretations and representations of the di rections have been debated by epig raphers. Some have connected them with the movement of the sun during the day, sugg esting that east and west are associated with the rising and setting of the sun, while the use of north and south are more variable including the zenith and nadir of the sun (Bricker 1988), up and down (Coggins 1988), or that the horizon represented north (Closs 1988). Each cardinal direction has a se ries of symbolic meanings a nd connotations. The east, with the rising of the sun, was considered positive and it represented fertility and the birth of the sun (Coggins 1980) The west or the setting sun, on the other hand, was closely associated with death and the underworld (Coggins 1980) North and south are not very well defined or understood and may in fact represent both up and down respect ively (Bricker 1988; Cl oss 1988; Coggins 1988). Other associations for the quadrip artite separation of the ancient Maya cosmos have been closely 33

PAGE 34

identified with the order and chaos associated with the creation of the Maya world (Pugh 2001). The center of this square or intersection of lines can be used to define both ritual use of space (Hanks 1990; Moyes 2001) and the Rituals of Foundations (Garcia-Zambrano 1994) used by the Maya to define the borders and placement of communities which usually contain a center defined by a mountain with a cave or water hol e. Archaeologists have identified the five directional associations within the formation of archit ecture (Coggins 1980; Pugh 2001), in the codices and other inscriptions (Bricker 1983, 1988; Closs 1988; C oggins 1980), the placement of communities (Garcia-Zambrano 1994) and also in the placement of artifacts (Griffith and Helmke 2000; Ishihara a nd Griffith 2004; Moyes 2001, 2002, 2005; Moyes and Awe 1998). Epigraphic accounts from the Maya codices sh ow an association be tween directionality and artifactual offerings. The accounts in the Maya codices include animal offerings and sacrifices which work well for this research pr oject. For example, in the Dresden Codex animal offerings are linked to Chac in relation to the four world directions in the almanac on pages 29b30b. The directional offerings identified in this section include to the east an offering of a tortoise, to the north a fish, to the west an iguana and to the south an offering of a wild turkey is made at 13 day intervals in this 4 event cycle. The almanac on pages 29b-30b has been described as a summer rain-making ceremony by Victoria Bricker (1991). This is a known epigraphic example of directional o fferings of specific taxonomic animal groups and it also may have some connections with caves and rain. Left and Right Sides The human body, in both the ethnographic and ar chaeological record, has been shown to be the most fundamental spatial domain of the Maya (Brown 2004:42). William F. Hanks (1984) uses the body as a source of reference when he identifies t hree paired oppositions: front/back, left/right, and up/down (Brown 2004:42) The most significant of these dichotomies 34

PAGE 35

to the ancient Maya is the relationship between the left and right sides, which is idenfied in the left and right sides of the human body, left/ri ght spatial orientation, and handedness have had important and symbolic meanings (Palka 2002:419). In the ethnographic and iconographic record, the sacredness of sidedness is an important part of the an cient Maya worldview that is also linked to the cardinal directions. The right side is identified with the east di rection, the rising sun, and birth (Coggins 1980), and also with men, power and purity (Palka 2002). While on the opposit e side, or hand, the left is a representation of the west, the setting sun, death, and the underworld (Coggins 1980), and is considered the domain of women and the weak. The primacy for the right side can be identified in the pictorial representation of Maya kings on st ela, as rulers are usua lly depicted facing the right. While their subordinates are pictured facing th e left, or in an inferior position. In relation to architectural and archaeological ev idence, little research has been conducted on the right and left dichotomy. In the zooarchaeological record, Mary Pohl (1983) showed that in zooarchaeological assemblages from ritual contexts there is an a ssociation with the ritually charged use of left skeletal elements over right skel etal elements. Other researchers working with ritual contexts, specifically in caves, also attempted to look fo r this type of right over left sided element patterning but none of these stud ies have been able to confirm this occurance (i.e., Anderson 2009; Brady 1989; Emery 2004c). Sidedness is an intrinsic feature of ritual spac e to the Maya. The Maya households are used as representations and also for specific ritual events. Stone (1995) makes reference to the identification of cave as being c onsidered a manifestation of the house. Therefore, caves could also used in the analysis and understanding of the left and right sidedness of the home and rituals that may have occurred within them. Caves, houses, and other such features are considered 35

PAGE 36

active agents with a personal perspective as though seated looking out through the entrance (Brown 2004). Several authors have suggested that sidedness is an intrinsic feat ure of ritual space (Palka 2002). Research at the site of Aguateca (Palka 2002) ha s shown that the left side of the house was used for food and textile production (ethnographically asso ciated with women), while the east side was associated with fine crafting and scribing (associated with men). Therefore it is possible that cave landscapes were also used with attention to their left and right sides. Light and Dark Bradys (1989:402) research found a relationship between the types of artifacts associated with the tunnels of the interior (or dark regions ) and those associated with the entrances to the exterior (or light regions) of caves. The symbolic association of light and dark (upper-world and underworld) is fairly intuitive in this separa tion (Brady 1989). It is likely that the Maya considered the darker regions more closely relate d to the underworld, and therefore more sacred and elite. While the light parts of the caves us ually include lit or dimly lit regions of the cave were used for public rituals that could be atte nded by all class groups. Brady (1989) used artifact differences to suggest that the ceremonies held in light regions were public rituals whilethose held in dark regions were privat e ritual events. He also noted th at the size of these enclosures played a role in the size of the groups atte nding these ceremonies (Brady 1989; Stone 2005). Caves and some architectural features have both a light, including well lit and twilight areas, and dark regions within them. The temple on top of a pyramid has been identified as human-made representations of the cave-mountai n relationship (Vogt and Stuart 2005).To the Maya the opening of the temple can be seen by all those in the plazas below; however, the indoor recesses or darker regions of these temples were not accessible of seen by those in the plaza. This access and inaccessibility to rituals may have been a part of the way elite distinguished their 36

PAGE 37

rituals from the commoners in the plaza below. This separation of space intensified the divide between the elites and non-el ites in the Maya worldview. Open and Restricted Access The construction of archaeological sites is no t haphazardly planned. There is intentionality in the placement and construction of architectu re based on the Maya worldview. Cosmologically speaking, the placement of large monuments and structures has strong connection ancestor veneration. The most prominent or longest lived ar eas of the site are usua lly the parts inhabited by the elite. The commoners are usually found outs ide of the ceremonial centers and the areas lived in by the elites. The accesses to the elite re sidences are also rest ricted with only single openings into their large courtyar ds. The most apparent aspect of Maya elite center construction is the division between areas of restriction, elite areas, and the open areas, acces sible to all status groups (Ashmore and Sabloff 2002). The use of space within a cave is an important part of understanding the idea of access to these naturally sacred places with in the Maya worldview. Accessibili ty to elite structures is an important part of understanding elite versus nonelite access to many of the Maya ceremonial centers. It has been shown to define how some caves were utilized by who had or had not had access to them. At the site of Dos Pilas, large monumental buildings have been found to have been built on top of caves (Brady 1997; Br ady et al. 1997). The access to these caves was restricted to only the elite residents. Many of these caves also had architectural construction mainly in the form of walls to restrict or close off different areas of the cave. These walls could be used to block off multiple entrances to certain chambers, t hus restricting what once might ha ve been a more open area. At multiple cave sites within western Belize, there are examples of stela erections within caves with restricted access (Awe et al. 2005; Healy 2007). The restriction and open access is also part of 37

PAGE 38

the elite and non-elite access to pa rts of the site and they are re lated to both private and public rituals, respectively. Monolithic erection and the restricted access to these areas within caves has been identified as an important part of the ritual behavior of the ancient Maya (Stone 199 5). With relation to the restricted and open access within caves, Dr. Andr ea Stone (2005) identified the prevalence of whole ceramics that were hidden in high (almost inaccessible) ledges in caves and also in low crevices. This suggested that these less accessible areas may ha ve defined more sacred and important ritual sites or events for th e Maya because of their accessibility. Stone (2005) defined a different model that she called the High-Low System of ritual offerings, which is similar to the hot-cold sy stem from present-day Mesoamerican people (Ishihara and Griffith 2004). The high within high-l ow system is related to the sun, or the hot, while the low is related to the dark, or the cold, both of which show the close association between these two systems. Stone (2005) found that more whole ceramics are found to be hidden in these high (almost inaccessibl e) ledges in caves and also in low crevices, and suggested that these less accessible areas may have defined more sacred and important ritual sites or events for the Maya. North and south have also been asso ciated with high and low, respectively. These associations may have played a role in the ut ilization of these inaccessi ble areas within caves. Summary Cognitive archaeology and landscape archaeolog y provided the theoretical framework for this research study. Caves are a sacred and ritually significant part of the Maya world and the use of these naturally and culturally influenced landscapes are vital to Maya cosmology. Cognitive archaeology includes both the intentional placement of these remains and the repetitive ritual distribution that have also o ccurred in certain parts of the caves. Using these theoretical frameworks, I attempted to reconstruct the anci ent Maya mind in relation to faunal offerings 38

PAGE 39

from cave. The five uses of space, directionality sideness, dark and light, and open and restricted access each play a role in the modern and an cient Maya cognition of their landscape. By studying these five spaces, I was able to better understand and analyze how the ancient Maya used caves as microcosms for their universes. 39

PAGE 40

CHAPTER 3 PRIOR RESEARCH IN MAYA CAVE ARC HAEOLOGY, ZOOARCHAEOLOGY, AND GIS Maya Cave Research: History and Progress over the Decades Only over the last few decades, have archaeo logists begun to look at the importance of caves to the ancient Maya (Bra dy 1989; Brady and Prufer 2005b). Using the European concepts of caves as habitation sites, ea rly archaeologists did not recognize other possible uses of caves in the Maya region (Brady 1989:3). Ja mes Brady (1989) was one of th e first archaeologists to demonstrate that caves in the Maya region were used in Prehispani c times, as some are still used today, for ceremonial or ritual events (Brown 2005; Brown and Emery 2008). The concept of the cave and its importance to the Maya is seen in both the use of caves around archaeological sites and in the architecture and artwork of the ancient Maya. Fo r example, archaeologists theorize that the temple-pyramid complex imitates th e cave-mountain relationship because the single opening of these windowless shrines mimicked th e deep, darkness of caves (Vogt and Stuart 2005). Many archaeologists have given in depth hist orical accounts of the investigation of caves in the Maya region (Brady 1989; Brady and Pruf er 2005; Ishihara 2007; Peterson 20060). I draw on these works to create a brief hist ory of Maya cave archaeology below. Brady (1989:10) splits the history of Maya cave archaeology into three periods: the Early Period (1840-1918), the Middle Period (1914-1950), and the Recent Period (1950-present). The early period included research by some of the forefathers of Ma ya archaeology and their findings within multiple cave sites in the Maya region sparked some interest into cave studies. Few of the sites, however, were excavated and archaeologists were only in terested in the findings of treasures and exotic artifacts. The drawings of John Lloyd Ste phens and Frederick Catherwood, the findings from the dredging of the Cenote of Sacrifice by Edward H. Thompson, and the 40

PAGE 41

excavations for early human occupation in the Yu catan by Henry C. Mercer, provide some of the important backdrops to the great discoveries from this time period During the Middle period (1914-1950), there was a hiatus in major archaeological research throughout the world. Due to multiple world wars and economic downspins, archaeology was little studied or done. Despite th e rarity of archaeological wo rk at the time, Thomas Gann conducted research within the Maya region and sp ecifically within caves. Ganns work identified all caves in the Maya region as habitation site s even though the artifact ual remains suggested otherwise (Brady 1989). In the recent period, Brady (1989) identifies all of the research that had been completed up until his dissertation. But, this period has been extended another 20 years since its publication (Brady and Prufer 2005b). The first research paper to include the analysis of caves as places for rituals and habitation was wr itten by J. Eric Thompson in 1959. In Thompsons (1959) The Role of Cave in Maya Culture building upon ethnographic and historical sources, caves are identified as places of ritual instead of places of hab itation. Thompson lists eight possible uses for caves including as sources of drinking wate r, sources of sacred water, plac es of ritual rites, for burials or cremations, as art galleries, ceremonial depositi on of artifacts or remains, areas for refuse, and other non-specific uses (Brady 1989: 32). Most of these uses revolve around water, rituals, human burials, offerings of goods both artifacts and ecofacts, and as art galleries This list of ceremonial uses for caves in the Maya region is still an important pa rt of how researchers identify and explore the use of caves in this culture area. Archaeologists and iconographers alike are now studying the Maya cave systems as an intrinsic feature of th e ancient Maya ritual landscape and ceremonial universe (Brady 1989; Healey 2007). The material rema ins are ritually linked and shoul d form a spatial pattern of 41

PAGE 42

distribution when they are analyzed. Recent et hnographic and ethnoarchaeological studies have shown a close association between rockshelters and hunting rituals in within the modern day Maya located around Lake Atitlan in Guatemala which show relative importance of caves in Maya rituals (Brown 2005; Brown and Emery 2008). Iconographic and epigra phic research has found depictions and writing of caves in codices and carvings of the ancient Maya. These depictions of the caves have been associated with the cave in many diff erent ways including its depiction as a house, a hole, a place of fertili ty, and the link to the underworld (Stone 1995). Zooarchaeology has been included in some recent studies; however it has not yet been fully integrated into these studies (Ish ihara 2007; Peterson 2006; Spenard 2006). Brief History of Maya Zooarchaeology In the past, Maya archaeologists viewed anim al remains and their associations within archaeological contexts as evidence of past subs istence practices or as a means to reconstruct ancient environments (Emery 2004a, 2004b). However, to the ancient Maya, animals also served an important function in ritual behavior (Emery 2003, 2004a, 2004b, 2004c; Pohl 1983). Pohl (1983) provided the first review of the ritual importance of an imal remains in caves, cenotes, burials, and caches. Some of Pohls (1983) ideas became a framework for the analysis of ritual faunal deposits, most notably her theory on the ritu ally charged use of left elements over right elements (Beaubien 2004; Moholy-Nagy 2004; Teeter 2004). Early zooarchaeological studies attempted to identify the possible ritual associations of animal bones from caves by looking for the patter ning and associations presented by Pohl (1983) (e.g., Brady 1989). There are only a few zooarch aeological studies that analyze cave fauna sufficiently (Brady 1989; Emery 2004; Pendergas t 1969, 1971) and only a few have looked at multiple cave sites within a region. Usually only small faunal assemblages or parts of an assemblage are identified (Ishihara 2007; Peterson 2006). 42

PAGE 43

Beyond even the simple analysis of animal a ssemblages recovered from cave sites, the archaeological and ethnographic evidence for pattern ed associations between artifacts related to specific ritual beliefs or be haviors suggests that zooarch aeological remains can provide important additional data. Animals are vital actors in most Maya rituals today and in the past (Pohl 1981), therefore their remains ar e also likely to have been discarded or intentionally placed in patterned spatial distributions and/or in pa tterned associations with other artifacts. These remains are vital to a full unde rstanding of the archaeological patterning being revealed through other archaeological studies of ancient cave-related rituals. Ritual Use of Animals from the Maya Region In Mary Pohls 1983 book chapter Maya Ritual Faunas: Vertebrate Remains from Burials, Caches, Caves, and Cenotes in the Maya Lowlands she describes in detail each of the major vertebrate animal species and families that ar e represented and most co mmonly used in rituals from the archaeology, ethnography, and ethnohistoric records. This list includes deer, monkey, peccary, dog, felines, fish, snakes, opossums, armadillo s, crocodiles, turtles, turkeys, birds, and small mammals (i.e., rats, bats, etc.). This list is a direct reflection of the types of animals most commonly depicted in the Maya codices and t hose most commonly found in the archaeological record. Some of the major animal groups are highlighted and reviewed from this paper and newly identified associati ons are also discussed. The most commonly identified faunal remain at most archaeologica l sites and in the epigraphies in the Maya region is the deer. The white-tailed deer ( Odocoileus virginianus) is identified more often but the brocket deer ( Mazama sp.) is also an important deer species. The deer is associated with the sun and the renewal rites of the Maya New Year ceremonies known as the cuch rites which have been identified as being related to the modern day cargo rites (Pohl 1981, 1983). Within the Maya codices the deer, sp ecifically the white-tailed deer (von Nagy 43

PAGE 44

1997) is an important part of hunting and tr apping scenes (Taack 1973; Vail 1997; von Nagy 1997), and as ritual offerings (Bricker 1991). Peccar ies (Tayassuidae) are w ithin the same order, Artiodactlya, as the deer, and they are also co mmonly identified at arch aeological sites. There are no identified associations but peccary skulls are often found in the archaeological record and may have been intentionally treated differen tly from other parts of the body (Pohl 1983). There are two major groups of monkeys located in the Maya region, the howler monkeys ( Aloutta palliate and Alouatta pigra ) and the spider monkey ( Ateles geoffroyi). Monkeys are rarely identified in the archaeological record, bu t they are associated with the arts and creation myths of the ancient Maya (Bak er 1992). Scribes are often depicted as monkeys and their body parts are sometime incorporated in the depictions of deer which may be a link between these two animals (Pohl 1983). There have been a few monk ey remains found within caves in the Maya region. Dogs, or the family Canidae, are also an im portant part of the ar chaeological record, the domestic dog ( Canis familiaris ) represents one of the very few domesticated species in the Maya world. Dogs were at time both an important food source, but also an import ant part of rituals, including the New Year ceremonies. Mythological literatures (Pohl 1983) identified dogs as being associated with assisti ng those across the river of the underworld and were commonly offered in burials for this purpose. The felines or cats, the family Felidae are usually associated with the largest repres entative species in the Maya region, the jaguar ( Panthera onca). Jaguars are closely associated with lineage, rulership an d power, and also as an underworld deity (Pohl 1983). Jaguar paws and pelts are f ound in the epigraphic and archaeological record and may have played a role in the accessi on of rulers (Pohl 1983). 44

PAGE 45

Fish have been associated with some rituals like the cuch rite and marine fish, in particular, can be found in burials and elite residences. Mari n fish are considered the most representative species in the trade between co astal sites and inland sites, and these marine goods represent status differentiation at inland sites. Fish remains have been identifi ed as part of the cult of the sea which shares some associations with renewa l ritual events (Pohl 1983). Marine shells are also important symbols to the ancient Maya, they have been found to be associated with death and birth, and also with fertility, rain, and wate r (Pohl 1983). Fish and shells are both found to be associated with water and may also have a close association with caves since these land formations are also associated with being places of pure water and fertility. There are some important reptilian families iden tified as being ritually significant to the Maya, including snakes (Serpentes), crocodiles ( Crocodylus sp.), turtles (Test udines), and lizards (Sauria) represented mainly by Iguanas (Iguanidae). Snakes are found in depictions of ritual bloodletting and may be associated with this act of sacrifice to the gods. Crocodiles are associated with the earth and the ancient Maya de pict and describe the world as the back of a crocodile (Pohl 1983). Turtles are associated with water and may have also been used as drums. These turtle drums may have provided the music during ritual events, especially those that occurred within caves (Zender 2006). The association of water and turtles may be an important connection between the depositions of their remains within caves that are also associated with watery worlds. The iguana is an important part of the offerings by the ancient Maya to the gods. Iguana offerings in the form of iguana breads (Thompson 1972) or iguana tamales (Taube 1988:238) are found throughout the Maya codices and they have been also identified to offerings in to the western dire ction (Bricker 1991). 45

PAGE 46

There are numerous species of birds located within the Maya region. The most notable and only other domesticated animal, besides dogs, bees, and possibly the Muscovy duck, from the Maya region is the turkey ( Meleagris gallopavo ). The turkey was identified as an important part of ritual offerings during the Po stclassic and is found depicted mu ltiple times in the codices (Bill et al. 2000; Bricker 1991, 1997; Pohl 1983; Ta ube 1988; Thompson 1972) and the ethnographic accounts of Diego de Landa (To zzer 1941). Other birds have speci fically important functions including the quetzal whose bright green feathers are shown in the ornamentation of kings, the macaw which was traded long distances and is conn ected with the sun because of its bright red wings, and finally, the owl, which is identified with omens (Pohl 1983). Both medium-sized mammals, including the op ossums (Didelphidae) and the armadillos ( Dasypus novemcinctus), and small mammals, including rodents (Rodentia), and bats (Chiroptera), are identified as be ing significant animals to the Ma ya. Opossums are depicted in the Maya codices as Bacabs and may be part of the New Year and Uayeb rites (Thompson 1972; Love 1986; Taube 1988). The armadillo is ra rely represented pictorially, but within Mesoamerica it has been associated with fertility (Pohl 1983). Bats have symbolic links to the underworld and caves, and they may be an important symbolic representation of the underworld (Brady 1989; Pohl 1983). Rodents are also found w ithin caves and they too may have some associations with the underworld. Pohl (1983), besides individuall y identifying important animal species to the Maya, also discusses some of the symbolic representations and use of animal body elements, element sides, and animal age classes. Deer haunches are offered throughout the Maya codices and are theorized to have a close connection with the offering of animals by e lites (Bricker 1991, 1997; Thompson 1972). The haunches are considered a prime cut of m eat to the ancient Maya. In 46

PAGE 47

Pohls (1983) work at Seibal and Copan and the Maya codices depictions, she found that the left haunches were the most important part of the deer offered in rituals and consumed by the Maya elite. Pohl (1983) identifies directional symbolism in the ritual deposition or offering of faunal remains, most notably her theory on the ritually charged use of le ft elements over right elements (Beaubien 2004; Moholy-Nagy 2004; Teeter 2004). The age of animals is also considered an important part of how and when animals are offere d for specific sacrifices. When the age classes of animals are identified in ritual contexts and the epigraphic lit erature, there are examples of younger animals as being necessary in the sacrificial offerings of animals for certain rites. For example, immature deer are needed for some rituals to the rain god and fertility goddesses and young dogs have been identified in the New Year ritu als (Pohl 1983). The offering of animal remains may also have a close association with the three main elements of the Maya including air, earth, and water (Kunen et al. 2002; Math ews and Garber 2004). At the site of Caracol, Diane Z. Chase and Arlen F. Chase (1998) identifi ed a cache as having three layers that contain a top layer of bird remains symbolizing air a mi ddle layer of mammalian re mains symbolizing land or earth, and a lower level containing fish remains that symbolizes the watery underworld. Previous Maya Cave Zooa rchaeological Research Zooarchaeology offers an interesting tool of analysis for understanding ritual. During the excavation of surface sites it is difficult to di stinguish between the animal remains from consumption or ceremonial practices. At surface sites most midden, occupation, and fill contexts contain the remains of animals used for bot h food and domestic and public ritual. Caves, however, do present a known ritua lly utilized archaeological cont ext, which zooarchaeologists can use for our advantage. There are a few resear ch studies that have attempted to incorporate the faunal materials into the overall analysis of ar tifacts from these sites. These sites are located in both Guatemala, including Naj Tunich (B rady 1989), Cueva de los Quetzales (Emery 2002), 47

PAGE 48

the Grieta Principle at Aguateca (Ishihara 2007), cave sites surrounding Cancuen (Spenard 2006), and also in Belize, including the caves sites Eduardo Quiroz Cave (Pendergast 1971), Actun Balam (Pendergast 1969), Actun Polbilch e (Pendergast 1974), and the Sibun Valley Caves (Peterson 2006). In the early to mid-1970s, Pendergast (1969, 1971, 1974) incorporated faunal analyses into the study of Actun Balam, Eduardo Quiroz Cave and Actun Polbilche. These analyses provide information about the types of animal remains from the site; however, they are not fully incorporated into the overall analysis of the s ite and this keeps the z ooarchaeological and other artifactual remain analyses separated. At the si te of Naj Tunich, Brady (1989) identified possible uses for the cave in elite ritual events and accession ceremonies, and he also incorporated the analysis of faunal material conducted by another res earcher into the overall analysis of the site. This analysis helped in solidifying and identifyi ng some of the patterns that were hypothesized by Pohl (1983). Recent studies, the Sibun Valley Caves (Pet erson 2006) and the Grieta Principle at Aguateca (Ishihara 2007), have also incorporated the analysis of faunal remains into the overall identification of patterns within the site. In 2002, Emery analyzed a sample of the faunal remains from the site of Cueva de los Quetzales and was the first paper to specifically identify and discuss in detail the overall patterning of the remains from this elite-controlled faunal assemblage. Over time, the analysis of faunal remains in Maya cave archaeology has increased and these remains are being incorporated and iden tified as important vita l parts for the overall analysis at these cave sites. However, there ar e some recent studies at the cave site of the Cancuen area that lack the necessary labor atory identifications necessary in using zooarchaeological materials in the analysis of faunal remains (Spenard 2006). Animal remains 48

PAGE 49

were identified very sparsely from photogra phs, and although they were incorporated into discussion of these sites, I feel they were a st ep back in the analysis of faunal remains from archaeological contexts. In a recent study from this year, Elyse Anders on (2009) examined the faunal remains from multiple cave sites in Belize and Guatemala to identify possible ancient hunting caches. The ethnographic record has shown that multiple rockshelters around the Lake Atitln in Highland Guatemala contain large number of animal remains being cached and offered as part of preand post-hunt ceremonies (Brown and Emery 2008).A nderson (2009) identifies whether or not the ancient Maya also cached or interred the rema ins of animals within cav es during hunting rituals as proposed by Emery and Brown (2008). These st udies were able to identify the possible relationships between the modern and ancient use of rockshelters and caves for hunting caches. The specific placement of animal remains more accurately accessed using spatial computer analysis programs to identify the patterns that may be lost during visual analysis. GIS in Cave Research Projects and the Maya Region Geographic Information Systems (GIS) has b ecome a prominent part of archaeological research because it offers arch aeologists a tool to better unders tand site-level and artifact-level spatial distributions. It also pr ovides a database management syst em to organize archaeological data. To some archaeologists GIS provides a tool for analysis, but it also can be seen as a methodological and possibly theoretical addition to study of archaeological sites. Initially used to look at the regional level of sites, GIS can also serve as a great m ode of analysis at the siteand even within the site-level (Ebert 2004; Kvamme 1999). Therefore, the scale of GIS can be an important addition to understanding ancient cultures. GIS is a relative new science to the field of archaeology and anthropology. It has only recent ly become an important addition to most archaeological research project. From a zooarchaeo logical perspective, GIS can serve as a tool to 49

PAGE 50

identify the spatial distribution of faunal remain s at an archaeological site (Marn Arroyo 2009; Nardini and Salvadori 2003).Within Maya cave res earch, GIS has also been shown to be a great tool in the identifica tion of patterns associated with the ancient Maya wo rldview (Moyes 2001, 2002, 2005). GIS was initially used to study large-scale regional studies in archaeology (e.g., Belli 1999). But more and more studies have begun to us e it to look at the distribution of artifacts within a single archaeological site (e.g., Arata 20 08). As a zooarchaeologist, I am interested in using GIS as a tool for understa nding the spatial distribution of faunal remains within a site to answer questions of subsistence, social classes, trade, and ma ny other aspects of the human and animal relationship. On a site le vel, the mapping of the distribution of faunal remains can help to answer questions of site usage and seasonality. In a recent study by Marn Arroyo (2009), the spatial distribution of faunal remains at El Miron Cave, Spain were analyzed to find if there were patterns in both the types of animal remains id entified in the cave and their overall placement within the cave. Marn Arroyo (200 9) showed that the remains iden tified indicated that the site was temporarily occupied during the summer and that differential butchering techniques were used for the main two species identified at the site. The red deer, a large-sized mammal from the region, were butchered at the kill site while medi um-sized mammals, Ibex, were brought back to the cave whole. The research at El Miron cav e presents an interesting model for future zooarchaeological studies, including my thesis, because it shows that the patterning of faunal remains can be used in unders tanding past human behaviors. Within the Maya region, GIS has also helped in the analysis of patterning within cave sites. Archaeologists working in the Western Belize Regi onal Cave Project incorporated GIS analysis into their cave research (Moyes 2001, 2002, 2005; Moyes and Awe 1998; Griffith and Helmke 50

PAGE 51

2000). These studies produced a spatial understanding of the patte rns of remains within these cave sites. They also produced a large scale da tabase that includes all of the necessary and important artifactual information in relation to artifact placement both vertically and horizontally. GIS offers resear chers the opportunity to give la ter researchers the chance to reconstruct and reanalyze the remains from an arch aeological site. For this thesis, the use of old maps and the addition of databases about the faun al assemblages from these sites has helped to build a basis for the study of zooarch aeological remains from these sites. Current Research Goals Maya cave zooarchaeology is a new and budding field in Maya archaeology. Until recently, Maya zooarchaeological research provide d archaeologists primarily with a laundry list of species, with little to no interpretation of the animal remains being identified within these projects. New research is looki ng at the ritual significance of animal remains in these known ritual contexts of caves. The Ma ya have a strong connection with rituals and the use of animals within these rituals. My research project is pus hing it further by adding a spatial component of using GIS to understand the distri bution and intentional placement of these remains within the caves by the ancient Maya. The next chapter descri bes in full detail the cave sites that were analyzed in this study. Each site represents an example of how caves we re used by the ancient Maya for ritual events and the influe nce of animals during these rituals. 51

PAGE 52

CHAPTER 4 RESEARCH SETTING: FIVE CAVES FROM THE MAYA REGION UNDER ANALYSIS Introduction The Maya region is located in Mesoameri ca and includes the Yucatan Peninsula in Mexico, Guatemala, Belize, and parts of Honduras and El Salvador. This region lies on a karstic limestone foundation that is characterized by si nkholes, cenotes, caves, ro ckshelters, and other such natural formations. My research concentr ates on faunal assemblages from two cave sites from Belize and three from Guatemala. One of these faunal assemblages was previously identified in 1984 by Susan Colby and published in James Brady's dissertation on Naj Tunich. The remaining four cave faunal assemblages from Caves Branch Rockshelter, Stela Cave, Cueva de Sangre and Cueva de El Duende were identifie d by me for the purpose of this thesis. Here I present the site and excavation information for each assemblage. The location, use patterns, and excavation methods, including placement of units or surface collections and cave formations, for each site are highlighted and used to explain the selection of these sites for my analysis. In 1997, Jaime Awe (1998) began a research project called the West ern Belize Regional Cave Project (WBRCP) that was interested in understanding the use of multiple cave sites around the modern town of San Ignacio. Their main goals were to identify patterns and difference for the use of caves within the regi on, over time, and by different social classes. WBRCP started off by mapping and excavating know n cave sites around San Ignacio, but the project also continued to examin e and look for new cave sites with in the region that were later excavated. The overall theme of the WCRCP res earch questions were to understand how and by whom were these cave were being used th rough the mapping and excavation of multiple cave sites in this region (Awe 1998). Awe (1998) describes each cave as being a different and unique natural feature, since no two caves are ever the same. Some sites are simple rockshelter 52

PAGE 53

overhangs, like Caves Branch Rockshelter; how ever, others are more complex and include multiple chambers that show a large amount of human construction and movement of objects into these chambers, such as the large limestone monument located in the back chamber of Stela Cave. The idea that all caves are morphologically uni que is hypothesized to have played a role in the type of rituals that may have been performe d at these sites (Awe 1998). Stela Cave and Caves Branch Rockshelter were both part of the WB RCP and their mapping and excavations were done under supervision of Dr. Jaime Awe (1998). Primary Research Sites in Belize Caves Branch Rockshelter, Belize Caves Branch Rockshelter (CBR) is located in the Cayo District of Western Belize along the Caves Branch River Valley, approximately 20 k ilometers southeast of the modern capital of Belize, Belmopan (Figures 4-1, 4-2) (Wrobel 2008). A rockshelter is a cave formed by a ledge of overhanging stone, the most important part of the rockshelter is the dripline which is a line on the ground in front of a cave formed from the dri pping of water down from the overhanging rocks above. CBR is 35 meters long north to south, 15.2 meters high, and the dripline goes to a maximum depth of 10 meters (Glassman and Bo nor Villarejo 2005). The Caves Branch River runs to the south of the site and may have been an important source for procuring animals whose remains were possible deposited within the site. CBR was excavated in 1994 and 1995 by Juan Luis Bonor Villarejo (Bonor 1998, 2003; Glassman and Bonor Villarejo 2005) and recently from 2005 to 2007 by Gabriel Wrobel (Wrobel 2008; Wrobel and Tyler 2006). The archaeological site of CBR is well known for containing a large number of human burials The ceramics assemblages cover a long time span and include numerous ceramics dating to the Floral Park / Hermitage / Spanish L ookout and New Town Comp lexes or the Late Preclassic to the Early Classic (Wrobel 2008:3) Accelerator Mass Spect rometry (AMS) of the 53

PAGE 54

human remains from the site include a range of 800 years that work within this time frame (Table 4-1). There were 15 to 20 mounds located in front of the rockshelter which appears to represent the local popu lation who utilized CBR as a cemetery (Bonor 1998, 2003). The use of the rockshelter as a cemetery and its lack of grave goods appear to give the site a more domestic rather than ritual function for the loca l population (Bonor 1998, 2003). Wrobel (2008) used Petersons (2006:13) hypothesis fo r the Sibun River Valley cave si te. That is that the elites controlled larger and more impressive caves for their rituals, while leaving the smaller and less impressive caves to be used by commoners. Thus, CBR and other smaller cave sites may represent differences in the use of caves between elite and commoners and not the separation of different ritual events (Healy 2007; Wrobel 2008). The CBR faunal material for this analysis comes from the recent excavations by Dr. Wrobel in 2005 to 2007. I analyzed the remains from the 2005 and 2006 seasons which included excavations of the northern (O peration 1A), central (Operatio n 1B and 1D), and southern (Operation 1C) portions of the rock shelter. These excavations were interested in looking at the overall distribution of human remains at the si te (Figure 4-2) (Wrobel and Tyler 2006; Wrobel 2008). During the 2006 season, excavations were con tinued in Operations 1A and 1B, with a new operation, Operation 1D, being excavated within the small cave located in the eastern-most section of CBR. During the first field season in 2005, an attempt was made to define stratigraphic layers; however, dis tinct layers were not present. Therefore, all levels were excavated in 20 cm arbitrary levels (Wrobel a nd Tyler 2006; Wrobel 2008). The soil is typically a gray silty limestone soil that is homogenously mixed chronologically and is defined as grave fill (Wrobel and Tyler 2006). Ceramic sherds dating from the Late Preclassic to the Early Classic 54

PAGE 55

period were found within the same level. Multiple disarticulated burials were found within the site caused by the excavation of gr aves during different time periods. The excavation units of CBR were made on a grid system of 1 meter by 1 meter units which aided my analysis of the spatial distribu tion of faunal materials (Figure 4-2). For the GIS analysis of this site, the units were used as the specific areas of an alysis. This site provided some interesting information because it is a rockshelter with known huma n interments on a large scale. The soils are not stratified and represent grave fill which might show some different patterning than the other caves in this study that do not have multiple burials associated with them. There is also the possibility for high natural disturbances because of both run-off of water and also animal intrusions into the soils within the rockshelter (Wrobel and Tyler 2006). Stela Cave, Belize Stela Cave is located in the Macal River Valley of western Belize and was also a part of the Western Belize Regional Cave Project (Figures 4-1, 4-3) (Ishihara an d Griffith 2004). This small cave is easily accessible and was looted by modern people. The ceramics within the cave date from the Preclassic to th e Late Classic periods (Table 4-1). The cave was mapped and excavated in 2003 (Ishihara and Griffith 2004). The excavation and mapping of the cave was done to examine the extent and nature of the vari ous architectural features and their role in the use of space in Stela Cave (Ishihara and Gr iffith 2004:57). The cave consists of multiple chambers which contain architecture modificatio ns (Figure 4-3). These modifications may have served multiple functions; the most notable is the large square stone slab in Chamber 3 that is theorized to have been used as a platform for pe rformances or as an alta r (Ishihara and Griffith 2004). No site has been identified near Stela Cave. In Chamber 1, six units were excavated. Units 1 to 5 were 1 meter (east/west) by 4 meters (north/south) and unit 7 was a 1.5 meter extension north off of un it 2 (Figure 4-3). In the back 55

PAGE 56

chamber, Chamber 3, there were four excavati on units (Units 8 to 11). North of the large limestone platform or altar there was a 2 me ter by 2 meter excavation unit. Along the western wall of the chamber, there was a very small a nd shallow surface (Unit 10) collection taken along the wall. Unit 9 was a 1 meter by one meter unit ex cavated within a small alcove in the northern part of Chamber 3. Finally, next to a large st one formation, unit 11 was excavated; it followed the contour of the rocks with an outline of a unit that was about 2 meter by 2 meters in size. Faunal materials were identified for most of the units, except for units 9 and 10. Primary Research Sites in Guatemala The next two caves, Cueva de Sangre and Cueva de El Duende, are locat ed at the site of Dos Pilas in the Petexbatun regi on, in the Pasion River Valley a nd were part of the Petexbatun Regional Caves Survey (Figures 4-1, 4-4, 4-5). Th ere have been about 30 caves identified in the Dos Pilas region and seven of these caves have been mapped and excavated and/or surface collected. Cueva de Sangre, Guatemala Cueva de Sangre, Guatemala is located above a settlement at the site of Dos Pilas about 3 kilometers southeast of the El Duende Pyramid which is located 1 km we st of the site core (Figures 4-1, 4-4) (Brady 1997; Brady et al. 1997; Brady and Scott 1997 ; Minjares 2003). Cueva de Sangre is a long cave measuring 3,314 meters in total length that runs underneath a small hill with 4 entrances (Brady 1997). The main area of ex cavation included a 400 me ter long section of the cave that had a detailed ma p of each artifact's location. Cuev a de Sangre was incorporated into the architecture and mounds on top of this site and to limit the access to this tunnel system many of these openings being deliberately bloc ked by stones (Minjares 2003). The artifacts from the site were both from surface collections and excavations. Cueva de Sangre is an interesting site because of its waterlogged, clay-rich environment. Special collection techniques were 56

PAGE 57

incorporated including the use of deflocculant and water screening in order to excavate through clay-rich soils (Brady and Scott 1997). Being the most complex cave system at Dos P ilas, there are multiple cave operations at the site. A total of eleven operations were either ex cavated or surface collected (Figure 4-4). Most of the surface collections came from Operation 1 within the main entrance of this complex tunnel system. Operation 1 contained three architectur al constructions including a wall built at the entrance of this cave and a 6 meter long limestone pavement area (Minjares 2003). This part of the cave floods during the rainy season and ha d a large amount of clay and muddy build up. Faunal materials were also located in the other op erations located in the cave, however, there are few descriptions of the other parts of the caves available (Minjares 2003). Ceramics from this cave date from the Late Preclassic to the Late Cl assic with most of the materials dating to the Early and Late Classic periods (Table 4-1). Cueva de El Duende, Guatemala Cueva de El Duende, Guatemala is a cave site that was located just southwest of the El Duende pyramid at the site of Dos Pilas (Figur es 4-1, 4-5) (Brady 1997; Brady et al. 1997). The site was identified from the collapse around the entrance of the cave and the site includes a chamber that then opens up to a long tunnel, which runs for about 300 meters (Minjares 2003). The cave has a circular opening that is about seven meters in diameter. The main chamber is 17 meters long and about 7 meters wide and the heig ht varies from 4 to 5 meters (Brady and Rodas 1992). The cave was split into three operations that was mapped and excavated (Figure 4-5). Operation 1 is located at the cave entrance and runs into the first main chamber of the cave (Minjares 2003). This area contained the remains of stone constructions that may have been used to block the entrance of the cave during ancient times. Operation 2, west of operation 1, has an entrance to the south that consists of a large, dry room and contains the largest artifact scatter in 57

PAGE 58

the cave. Operations 1 and 2 are both located within regions of the cave that are light by natural light (Minjares 2003). Operation 3 consists of two large tunnels with low ceilings and has a floor that maybe flooded during the rainy season b ecause of the high amount of clay and mud in this part of the cave (Minjares 2003:42). Operati on 3 has no natural lighting and is the dark-zone of the cave. The ceramics from Cueva de El Duende date to the Late Preclassic to Early Classic periods (Table 4-1) (Brady and Rodas 1992). There is a preliminary study of the faunal remains completed for Cueva de El Duende by Dr. Kitty Emery which were incorporated into my analysis from the site (Brady et al. 1991: 726-730). Published Source from Guatemala: Naj Tunich, Guatemala Naj Tunich is one of the largest known caves in the Maya region (Figur e 4-1). It is located in the Maya Mountain of Guatemala close to the Belizean border (Brady 1989). The cave was discovered in 1980's and multiple field seasons to ok place at the site during the early to mid1980s. The site is most notably known for having a large number of cave drawings with glyphs that are important for understanding the ritual use of caves by the Ma ya elite during accession ceremonies. A total of 9 operations were excavated in the cave, but the faunal remains analyzed in this thesis are from Operation IV. Operation IV was used for this study because it has a map that depicts all of the surface collection locations (Figure 4-6) The faunal remains from the surface collections and some excavations in Naj T unich were analyzed by Susan Colby at UCLA in 1984. Faunal material was analyzed and the id entifications include the taxonomic information, NISP, element, side, age, and presence/absence of burning. The cer amics in the cave date from the Late Preclassic to the Late Classic periods (Brady 1989). There are no sites located near Naj Tunich and the site is considered a pilgrimage site for the Maya, with the elite accessing the inside portions of the cave for private rituals and the commo ners having access to and utilized the entrance of the cave for la rge public rituals (Brady 1989). 58

PAGE 59

Summary The five caves identified in this research pr oject present a varied representation of the types of caves found within the Maya region. Th e variation in cave types, placement in or adjacent to sites, time periods, and the possible ritual events which occurred in these caves are important aspects to consider during this analysis (Awe 1998). Caves Branch Rockshelter is a known ancient Maya cemetery with a possible settlem ent site located adjacen t to it. Stela Cave contains a large altar or stela that was intenti onally placed in the cave and may have served as a place for ritual events. Cueva de Sangre and Cuev a de El Duende are associated with elite architecture at Dos Pilas and may have been only utilized by elites. Finally, the largest known cave site, Naj Tunich, may have served as an ancient pilgrimage site for the Maya. Each of these sites provides different conditions (or settings) for inve stigating the use of animal remains during ritual events. 59

PAGE 60

Figure 4-1. General Location of Cave s Sites in Belize and Guatemala. 60

PAGE 61

Figure 4-2. Location of Opera tions and Units for Caves Branch Rockshelter, Belize 61

PAGE 62

Figure 4-3. Location of Units for Stela Caves, Belize 62

PAGE 63

Figure 4-4. Location of Surface Collectio ns for Cueva de Sangre, Guatemala 63

PAGE 64

Figure 4-5. Location of Excavation Units for Cueva de El Duende, Guatemala 64

PAGE 65

Figure 4-6. Location of Lots for Op eration IV, Naj Tunich, Guatemala 65

PAGE 66

66 Table 4-1. Five Cave Site Information Incl uding Proximity to Sites, Time Periods, and Excavation Methods. Cave Site Proximity to Indentified Sites Time Periods Excavation Methods Caves Branch Rockshelter Small site nearby Late Preclassic to the Early Classic Excavation Stela Cave No identified sites nearby Preclassic to the Late Classic Excavation Cueva de Sangre Within Large Ceremonial Center Late Preclassic to the Late Classic Excavation / Surface Collection Cueva de El Duende Within Large Ceremonial Center Late Preclassic to Early Classic Excavation Naj Tunich No identified sites nearby Late Preclassic to the Late Classic Surface Collection

PAGE 67

CHAPTER 5 ZOOARCHAEOLOGY, GIS, AND VISUAL ANALYSIS METHODS Introduction This thesis examines the ancient Maya use of cognized space in caves by describing the distribution of animal remains across cave use surf aces. To test the spatial distribution of animal remains I used Geographic Information Systems (o r GIS) to compare the distribution of animal remains from five caves to various cognized spatial patterns sugge sted by ethnographic, ethnohistoric, and archae ological research. In this chapter I review the zooarchaeologica l methods I used in the primary identification and analysis of animal remains from four cave sites, Caves Branch Rockshelter, Stela Cave, Cueva de El Duende, and Cueva de Sangre and th e methods that I have used in incorporating data from published zooarchaeological inform ation from the cave of Naj Tunich. In this discussion I also note the biases that must be recognized in the assemblages as a result of different excavation, identification, and quantification/presentation methods. In the second section I review the methods used in incorporating maps from each site using GIS. The two spatial analysis tools, Spatial Au tocorrelation and Cokriging, are discussed. Also, my need to use a visual analysis to help with the interpretations of the sites is addressed. The standards I used in defining the cognized spaces ar e identified and the spat ial analyses that are used to test them. In this study I concentrated on the following categories of the separation of space within each cave: the cardinal directions (n orth versus south and east versus west), the dichotomies between the right/left, dark/light, and open/restricted spaces in the cave. These are not the only cognized spaces in the probable Maya universe, but they are the ones best suited to the analysis of these cave a ssemblages. To test thes e separations of space, I compared the differential placement of specific speci es. If larger assemblages of identified faunal 67

PAGE 68

remains were available, I would have also incl uded the analysis of ages, elements, body portions, body sides and specimen treatment such as burni ng and alterations by butchering or artifact manufacture. The separation of space within the caves was based on excavation unit placement and therefore the spatial distin ctions are somewhat limited. As well, some of the patterns overlap, for example, when the ri ght and left sides of the cave ar e also east and west. Therefore, in this section I discuss the biases inherent in my spatial analysis. Zooarchaeological Methods In 2007, the faunal remains from Cave Branch Rockshelter and Stela Cave were exported to Dr. Kitty Emery of the Environmental Ar chaeology Program of the Florida Museum of Natural History (EP-FLMNH) under a permit from the Belize Institute of Archaeology to Dr. Jaime Awe, project director. The animal rema ins from Caves Branch Rockshelter were excavated from 2005 to 2007 by Dr. Gabriel Wrobel, and those from Stela Cave were excavated in 2003 by a PhD student Cameron Griffith. Both cav e sites were excavated as part of the Belize Valley Archaeological Reconnaissance Project dir ected by Dr. Jaime Awe. As the first stage of my research project, I worked w ith the project excavators in Belize to organize, label, and choose appropriate samples for export, and I hand-carried the remains to the FLMNH laboratories. The faunal remains from Cueva de Sangre and Cueva de El Duende are on loan to Dr. Kitty Emery from Dr. James Brady by permit from the Guatemalan Institute of Archaeology and History and are housed in the Environmental Ar chaeology collections of the FLMNH. These two cave sites were excavated in the 1990s by Dr. Ja mes Brady and his colleagues as part of the Petexbatun Regional Archaeology Project directed by Dr. Arthur Demarest. The remains were imported to the USA in 1998 following initial an alysis and sorting by Dr. Emery, and were transferred to the EAP-FLMNH in 2001. Preliminar y reports on their faunal remains have been published by Dr. Kitty Emery and her identifications are included with my own in this study with 68

PAGE 69

her permission. I also used published data from the sites of Naj Tunich (Brady 1989) from Guatemala. This cave site was excavated by Brady in 1982 and 1983 and the animal remains were identified by Susan Colby at UCLA in 1984. Th e result from this analysis was presented as an appendix in Bradys dissertation. The four sites from which I analyzed animal remains are Caves Branch Rockshelter (20052006 seasons only), Stela Cave (2003 season), Cueva de Sangre (all seasons), and Cueva de El Duende (all seasons). In this an alysis, animal remains excavated from cave sites were identified using standard zooarchaeological methods such as those described in Reitz and Wing (2008) and elsewhere. The identifications relied on the comparative collections of the Environmental Archaeology Program of the Florida Museum of Na tural History. In general, all materials were identified to the lowest taxono mic grouping. However, due to th e high diversity and lack of comparative specimens at the species level, speci fically rodents (Rodentia), bats (Chiroptera), and birds (Aves) these remains were only id entified to these broa der taxonomic groupings. However, these specimens were othe rwise treated identically to all others in that I recorded the same element, age, and treatm ents as with other groups. For each specimen, I recorded where possible th e lowest taxonomic identification, type of element, body portion of the element, the region of the element from which the specimen originated (distal, proximal, mid-region), the completeness of the specimen (by percent), the sidedness of the element, the age and sex charact eristics of the specimen, and the condition of the bone including burning (blackened or calcined), ar tifactual modification (butchering and worked materials), and natural modificat ions (water damage, weatheri ng, animal gnawing, root action). In calculating age of the specime ns, I used three broad categorie s: (1) Juvenile, which includes unfused bones with juvenile epiphys is and other clear indications of very young ages (2) Juvenile 69

PAGE 70

+ Immature, which are defined by characteristic s such as unfused epiphyses and unerupted dentition, (3) Subadult + Adult, id entified by characteristics such as bone showing fusion lines or that are completely fused, and erupted and worn adult dentition. Burning, natural modifications, and artifactual modifications were categorized as present or absent from the bones. For materials that showed burning, these remains were classi fied as blackened, calci ned gray, and calcined white because these categories can be linked to specific burning temperatures (Reitz and Wing 2008). For the GIS analysis, all taxonomic group s, body portions, elements, degrees of burning, element sides, and age were assigned a numeric c ode (refer to Tables 5-1, 5-2, 5-3, 5-4, 5-5, 5-6). These are discussed in more detail in the next section. In zooarchaeology there are two main measur ements for quantifying faunal remains. The Number of Identifiable Specimens (or NISP) is the number of remains identified in each taxonomic unit (a single species or other ta xon). The NISP can over-e stimate the number of actual animals in the assemblage because some animals have multiple small bones and this measurement does not take into account fragme ntation of elements (Emery 2004 Chap 2; Grayson 1984; Reitz and Wing 2008). For example, the NISP of armadillos can be overestimated because each armadillo has over 200 scutes a nd when calculating the NISP, each scute is identified as a single armadillo specimen. The Minimum Number of Individuals (MNI) calculates the smallest number of individual animals that might be represented by each taxonomic unit (Emery 2004b; Grayson 1984; Reitz a nd Wing 2008). It is based on the fact that the largest number of discrete elements must be the minimum number of animals represented. For example, if a set of fifteen deer femurs incl ude nine complete left femurs and six complete right femurs, there is only a possibility of th e assemblage having nine complete deers. 70

PAGE 71

There is a bias of small sizes when usi ng MNI because the use of specific elements decreases the amount of remains th at can be used in the analysis Therefore, sample sizes are smaller, yet more accurate, estimates of the living faunal assemblage; but, they are a misrepresentation of the actual num ber of possible animals at the si te. In this study I have used the NISP as the basis for all calculations becau se sample sizes are small enough to affect the MNI accuracy. However, the MNI is presented fo r comparison in all tables. The NISP for this study included the larger and general taxonomic units, even including some of the classes and specific number of individual bones identified in these groups. Geographic Information Systems (GIS) The second part involved mapping the sites using ArcGIS, a GIS program that is used as a tool to define and quantify the spatial distribution of the faunal materials within the cave sites. I created the ArcGIS maps by scanning and transf erring older published and excavation field maps to computer images that were then incorporat ed into ArcGIS based on their known coordinate systems (as provided by the archaeologists). The only site to have a georeferenced map with correct scale and spatial position is Stela Cave and this was provided by Cameron Griffith the main excavator at the site. Scale was kept from the mapped images for all caves and those without specific coordinates were used for an alysis and reconstructe d from these published maps. There are two forms of images that are used in GIS, including vector and raster images. Vector images are represented by points, lines, and polygons that are used to mark specific locations and outlines of areas. While raster images consists of a series of pixels to represent values that are continuous or discrete. For example, elevation is be st represented by raster images because there are not well-defined areas for these da ta. Maps are made into vector images so that each unit can have a single area represented by it in the cave. Once these maps are made the 71

PAGE 72

faunal data can be incorporated and linked to the parts of the map so that animal remains distribution can be analyzed and quantified on a unit to un it, and possibly level to level, basis. The caves that I am using for this analysis were not fully excavated, nor were animal remains point-plotted (mapped with specific coordi nates per find); therefore, this analysis is based on relative proportions of animal remains be tween excavation units a nd surface collections at each site. The faunal remains were from both ex cavated units and surface collections and were either represented using points or polygons becau se the size for each collection was not recorded at every site. I used these spatial analyses, Sp atial Autocorrelation and Cokriging, to test for possible patterning of faunal remains in the caves. Spatial autocorrelation identifies the spatial distribution of species within the cave as being either disper sed, random, or clustered throughout the cave. Cokriging was used to identify the spa tial patterning of the fa unal specimens in relation to the dichotomy between the five separations of space, including left versus right sides of the cave, north versus south, east versus west, light versus dark, and open versus restricted space. The spatial analyses used in this study have certain sample size criteria in orderto produce viable data outputs. Spatial autocorrelation need s a sample size of at least ten but works more efficiently with a sample of thirty or more. For cokriging to analyze, a sample of ten units is needed. Therefore, some of the caves, includi ng Stela Cave (eight units) and Cueva de El Duende (eight units) did not run a cokriging spatial analysis. Due to the small sample size, I also conducted a visual analysis of the faunal remains to test for patterns related to the cardinal directions (east/west and north/south), right/left si des of the caves, restricted/open spaces that are and are not blocked off, and dark/ light areas within the caves, in relation to the type of elements, side of the elements, body portions, age, and bone conditions. I defined these separations of 72

PAGE 73

space by identifying them manually and then inco rporating these findings into the analysis. There are areas of overlap between these sepa rations which I addre ss in this chapter. GIS can better recognize and sort by numbers than by alphanumeri cal representations because ArcGIS is very specific in how text is identified and incidental spaces in sections are considered separate entries. To prepare the spr eadsheets with all of the faunal identifications from the cave sites, I am using specific number s cales in each of the fields recorded. The number number scheme is being used to represent ta xonomic groupings (Table 5-1), elements (Table 52), body portions (Table 5-3), sides of elements (T able 5-4), age characteris tics (Table 5-5), and the conditions of burning and char ring (Table 5-6). For example, I am using a numbering system for the taxonomic identifications of the faunal re mains; this list worked with both the general taxonomic groupings of class, order, family, genus and species identification and common names. The numbering for the taxonomic groups is provided in Table 51 and consisted of a much more indepth numbering system because of the amount of taxonomic groups identified. For the other parameters, I used a simple num bering system for the identification of the elements, left/right sides, porti ons, age identification, and the bur ning/charring of the bones. Due to the small sample size of the faunal remains fr om each cave, the GIS spatial analysis of the sites only includes th e taxonomic groupings and identifications. I decided on using single excavation units or su rface collections as the divisions of analysis because caves in the Maya region are known to ha ve mixed contexts. Also, most of the units were not excavations, but surface collections. I di d not take the size of the units into account during my spatial analysis, and most of the units were treated as single points. The location of recovery (provenience) of each of the bones has been recorded by the excavators. For each site, the detail of provenience informa tion is slightly different, but at CBR and Stela Cave the faunal 73

PAGE 74

remains have a specific catalogue nu mber that is related to the Op eration, Unit, Level, and Lot. At CBR in particular, each of these units were taken from an arbitrary 1 meter by 1 meter grid that was placed over the site before excavation. The numbers increase from north to south with letters that increase from the west to east. Many of the sites include su rface collections and each of these surface collections were recorded and identified as individual units. Spatial pattern recognition was facilitated by usi ng the maps with labeled excavat ion units. These maps are the important part of the next step in the analysis of these faunal remains in the caves. Separation of Space within the caves I am working under the idea that each cave pres ents a completely different space since no two caves are the same. These differences might al so have an important role in the type of ceremonies that may have occurred within the caves (Awe 1998; Brady 1989). Therefore, I first compared the animal remains between caves to look for patterning. I also looked at many different types of delineation of space within each of the caves. I have defined four different measurements including cardinal direc tions (north versus south and ea st versus west), left versus right sides, light versus dark regions, and open versus restricted spaces. During the analysis of space within the caves some of these delineations may overlap or not exist within certain cave contexts. Separations of space were both for the w hole cave and also within different sections of the cave. The extent to which I compared thes e divisions was limited by the type and placement of excavations within the caves. There is also ov erlap in some of these patterns in most of the caves. This led me to discuss each of the spatia l patterns of space individually when trying to identify patterns. The four delineations of space that I analyzed for each cave and the use of the separations of space depended on the size and formation of ea ch cave. Cardinal directions are and were important to the Maya, and simple separations of space were used for this analysis as each cave 74

PAGE 75

is split into the north versus south and the eas t versus west. The use of quadrants would be difficult to do because of the small sample sizes from the caves that are being analyzed. There is a strong ritual component in the le ft and right sides of buildings a nd this might also be important in caves as well. The cave was split into the left and right side from the caves perspective as an individual or agent by facing. The light versus dark regions of the ca ve was defined by the archaeologists. Light is defined as naturally li t and dimly lit parts of the cave, usually located near the opening, while dark is lo cated in the recesses of the cave that are not lit by natural light.. Finally, the open versus restricted spaces are de fined by the construction of walls or natural formations that limited access to parts or chambers of the cave. Cokriging involves adding different variables; I analyzed the total NISP per species within each unit or surface collection in relation to each of the above mentioned separations of space. The faunal remains were quantified using the sum NISP for each unit or surface collection. To start off my analysis I looked simply at the overall distributi on and pattern of distribution of burnt or charred bones at all cave si tes. I am interested in seeing if there are any specific patterns that form between all of the caves. All faunal remains, including those that have not been identified to specific species, were used for the an alysis of burnt remains. For the next step in the spatial distribution of remains, I identified possi ble patterning in the distribution of identified species, which were grouped in these specific taxonomic groupings: bo th large and small taxonomic groups including crustaceans (Decapoda, infraorder Brachyura), Actinopterygii (rayfinned fishes), Lepisosteidae (gar family), Siluri formes (catfish family), Amphibia (amphibians), Testudines (turtles), Sauria (lizards), Serpentes (snakes), Aves (birds), Meleagris gallopavo (wild turkey), Didelphidae (opossum faimy), Dasypus novemcinctus (nine-banded armadillo), Chiroptera (bats), Primates, Canidae (canid family), Felidae (cat family), Procyonidae (raccoon 75

PAGE 76

76 family), Tapirus bairdii (Bairds tapir), Artiodactyla (e ven-toed ungulates), Tayassuidae (peccary family), Cervidae (deer family), Odocoileus virginianus (white-tailed deer), Mazama sp. (brocket deer), Rodentia (r odents), Scuiridae (squirrel family), Agoutidae (pacas family) and Dasyproctidae (agouti family), and Sylvilagus sp. (cottontail rabbits). Identifying possible patterns in terms of species and family distribu tions of remains is an important step in the process. Species patterns were analyzed to see if species are placed into cer tain parts of the cave in relation to my spatial separa tions. Due to the small sample size from each of the caves, I was unable to work within the other parameters of interest, including elements, body portions, left/right elements, and age of identified specim ens. If given a larger sample I would have performed both of the spatial anal yses on these specific parameters. Summary My goal of this chapter is to provide the met hods for the analysis for identifying the spatial patterning of the faunal materials at ancient Maya cave sites using GIS. Due to the small sample sizes from these cave sites, I was only able to test taxonomic groupings of animals in this analysis using GIS. I include the visual analys is of these small samples to help with the identification of spatial patterning of animal rema ins from these cave sites. I have provided an outline of a possible method of analysis using GI S with a larger sample that is from a well excavated context. The use of a numerical system to represent the identifiable aspect of these remains is a necessary part of preparing the information for its use in larger zooarchaeological collections. The inclusio n of zooarchaeology and GIS at ar chaeological sites allow for more complex identifications of patterned use of anim al remains from archaeological sites and as a possible proxy for understanding the ritual use of animals by the ancient Maya. The use of GIS also allows for the creation of a database of zooarc haeological materials which have a spatial relationship within archaeological sites.

PAGE 77

Table 5-1. Gentax Numbers for Taxonom ic Classification of Faunal Remains GEN TAX Phylum/Class Order Family Taxa Common name 1 Class unknown 2 taxa identifiable 3 Gastropoda univalves or snails 4 Gastropoda, freshwater freshwater snails 5 Gastropoda: Prosobranchia 6 Neotaenioglossa Thiaridae Melania sp. river snail 7 Mesogastropoda Pleuroceridae Pachychilus sp jute 8 Mesogastropoda Pleuroceridae Pachychilus indiorum 9 Mesogastropoda Pleuroceridae Pachychilus glaphyrus 10 Mesogastropoda Ampullariidae Pomacea flagellata apple snail 11 Gastropoda, terrestrial/arboreal land and tree snails 12 Gastropoda, terrestrial/arboreal Mesogastropoda Cyclophoroidea Neocyclotus dysoni 13 Gastropoda, terrestrial/arboreal Stylommatophora Oleacinidae Euglandina cf. cumingii tree snails 14 Gastropoda, terrestrial/arboreal Stylommatophora Bulimulidae Orthalicus sp. 15 Gastropoda, terrestrial/arboreal Stylommatophora Bulimulidae Orthalicus undatus 16 Gastropoda, terrestrial/arboreal Stylommatophora Bulimulidae Orthalicus vivans 17 Gastropoda, terrestrial/arboreal Archaeogastropoda, srf Neritacea Helicinidae Helicina amoena 18 Gastropoda, marine marine snails 19 Mesogastropoda, srf Strombacea Strombidae 20 Mesogastropoda, srf Strombacea Strombidae Strombulus sp. conch 21 Mesogastropoda, srf Strombacea Strombidae Melongena melongena brown conch [crown conch?] 22 Mesogastropoda, srf Strombacea Strombidae Strombus gigas queen conch 23 Mesogastropoda, srf Strombacea Strombidae Strombus alatus Florida stromb 24 Mesogastropoda, srf Strombacea Strombidae Strombus pervianus 25 Mesogastropoda, srf Strombacea Strombidae/Cassidae 77

PAGE 78

Table 5-1. Continued GEN TAX Phylum/Class Order Family Taxa Common name 26 Mesogastropoda, srf Doliacea Cassidae Cassis madagascariensis Clench's helmet 27 Neogastropoda, srf Volutacea Xancidae Xancus angulatus West Indian chank [lamp shell] 28 Neogastropoda, srf Vo lutacea Columbellidae Columbella mercatoria dove shells 29 Neogastropoda, srf Volutacea Olividae olive shells 30 Neogastropoda, srf Volutacea Olividae Oliva sp. 31 Neogastropoda, srf Volutacea Olividae Oliva caribaeensis/sayana 32 Neogastropoda, srf Volutacea Olividae Oliva reticularis 33 Neogastropoda, srf Volutacea Olividae Oliva sayana 34 Neogastropoda, srf Volutacea Olividae Olivella dealbata 35 Neogastropoda, srf Volutacea Olividae Olivella nivea 36 Neogastropoda, srf Volutacea Fasciolariidae Pleuroploca gigantea horse conch 37 Neogastropoda, srf Vo lutacea Marginellidae Prunum apicinum 38 Pelecypoda bivalves 39 Pelecypoda, marine marine bivalves 40 Pelecypoda, marine Arca imbricata 41 Pelecypoda, marine Spondylidae Spondylus sp. 42 Pelecypoda, marine Spondylidae Spondylus americanus 43 Pelecypoda, marine Isognomon alatus flat tree oyster 44 Pelecypoda, freshwat er freshwater bivalves 45 Pelecypoda, freshwater Unionidae unionid clams 46 Pelecypoda, freshwater Unionidae Lampsilis sp. freshwater mussel/clam 47 Pelecypoda, freshwater Unionidae Nephronaias sp. freshwater mussel/clam 48 Pelecypoda, freshwater Unionidae Psoronaias sp. 49 Decapoda, infraorder Brachyura unidentifiable crab 50 Decapoda, infraorder Brachyura Gecarcinidae crab 51 Decapoda, infraorder Brachyura Gecarcinidae Cardisoma sp. land crab 52 Vertebrata animals with backbones 53 fishes 78

PAGE 79

Table 5-1. Continued GEN TAX Phylum/Class Order Family Taxa Common name 54 Chondrichthyes Myliobatiformes Dasyatidae/ Myliobatidae eagle and stingrays 55 Actinopterygii ray-finned (bony) fishes 56 Actinopterygii Lepisosteiformes Lepisosteidae gars 57 Actinopterygii Lepisosteiformes Lepisosteidae Atractosteus tropicus broad-head gar 58 Actinopterygii Lepisosteiformes Lepisosteidae Lepisosteus sp. slender gars 59 Actinopterygii Cypriniformes Catostomidae Ictiobus sp. suckers 60 Actinopterygii Cypriniformes Catostomidae Ictiobus meridionalis Usumacinta buffalo sucker 61 Actinopterygii Siluriformes catfishes 62 Actinopterygii Siluriformes Ariidae Arius sp. sea catfish 63 Actinopterygii Siluriformes Ictaluridae/ Pimelodidae freshwater catfishes 64 Actinopterygii Siluriformes Ictaluridae ictalurid catfishes 65 Actinopterygii Perciformes Cichlidae freshwater cichlids 66 Actinopterygii Perciformes Cichlidae Cichlasoma sp. freshwater mojarra 67 Actinopterygii Perciformes Scaridae Sparisoma sp. parrotfish 68 Tetrapods limbed animals 69 Amphibia amphibians 70 Amphibia Anura frogs and toads 71 Amphibia Anura, intermediate 72 Amphibia Anura, small 73 Amphibia Anura Ranidae frogs 74 Amphibia Anura Bufonidae Bufo sp. bufonid toads 75 Amphibia Anura Bufonidae Bufo marinus marine/cane toad 76 Reptilia reptiles 77 Reptilia Crocodilia Crocodylidae Crocodylus sp. 78 Reptilia Testudines turtles 79 Reptilia Testudines Testudines, large Dermatemys mawii/Staurotypus triporcatus 80 Reptilia Testudines Testudines, small 79

PAGE 80

Table 5-1. Continued GEN TAX Phylum/Class Order Family Taxa Common name 81 Reptilia Testudines Kinosternidae mud and musk turtles 82 Reptilia Testudines Kinosternidae Kinosternon sp. mud turtle 83 Reptilia Testudines Kinosternidae Kinosternon scorpiodes scorpion mud turtle 84 Reptilia Testudines Kinosternidae Kinosternon cruentatum red-spotted mud turtle 85 Reptilia Testudines Kinosternidae Staurotypus triporcatus Mexican giant musk turtle 86 Reptilia Testudines Dermatemyidae river turtles 87 Reptilia Testudines Dermatemyidae Dermatemys mawii Central American river turtle 88 Reptilia Testudines Emydidae box and freshwater (pond) turtles 89 Reptilia Testudines Emydidae Trachemys scripta slider 90 Reptilia Testudines Emydidae Terrapene carolina Yucatan box turtle 91 Reptilia Testudines Emydidae Rhinoclemys areolata black-bellied (furrowed) turtle 92 Reptilia Squamata: Suborder Sauria lizards 93 Reptilia Squamata: Suborder Sauria Iguanidae iguanas 94 Reptilia Squamata: Suborder Sauria Iguanidae Iguana iguana green/common iguana 95 Reptilia Squamata: Suborder Serpentes snakes 96 Reptilia Squamata: Suborder Serpentes Viperidae pitvipers 97 Reptilia Squamata: Suborder Serpentes Viperidae Crotalus durissus Neotropical rattlesnake 98 Aves/Mammalia mammal or bird 99 Aves birds 100 Aves, large 101 Aves, large/intermediate 102 Aves, intermediate 103 Aves, small 104 Aves Ciconiiformes Ardeidae herons and egrets 105 Aves Gruiformes Rallidae Laterallus ruber/ jamaicensis Black Rail or Ruddy Crake 106 Aves Gruiformes Rallidae Gallinula chloropus common gallinule/moorhen 107 Aves Psittaciformes Psittacidae parrots and macaws 108 Aves Psittaciformes Psittacidae Ara macao macaw 80

PAGE 81

Table 5-1. Continued GEN TAX Phylum/Class Order Family Taxa Common name 109 Aves Piciformes Ramphastidae toucans 110 Aves Strigiformes Strigidae Tyto alba barn owl 111 Aves Falconiformes Accipitridae Buteo platypterus broad winged hawk 112 Aves Falconiformes Falconidae Herpetotheres cachinnans laughing falcon 113 Aves Galliformes curassow, bobwhite, quail, turkey 114 Aves Galliformes Phasianidae Colinus sp. bobwhites 115 Aves Galliformes Phasianidae Colinus nigrogularis black-throated bobwhite 116 Aves Galliformes Phasianidae: Meleagridinae Meleagris sp. turkeys 117 Aves Galliformes Phasianidae: Meleagridinae Meleagris gallopavo northern/domestic turkey 118 Aves Galliformes Cracidae Penelope purpurascens crested guan 119 Aves Passeriformes passerine or perching birds 120 Aves Passeriformes Hirundinidae Tachycineta albilinea tree swallows 121 Aves Passeriformes Hirundinidae Petrochelidon sp. cave/cliff swallows 122 Aves Passeriformes Emberizidae Ammodramus savannarum Aimophila sp. Grasshopper, Rusty, or Botteri's Sparrow 123 Aves Passeriformes Emberizidae Cardinalis sp. cardinals 124 Aves Columbiformes Columbidae doves and pigeons 125 Mammalia 126 Mammalia, very large 127 Mammalia, large 128 Mammalia, large/intermediate 129 Mammalia, intermediate 130 Mammalia, intermediate/small 131 Mammalia, small 132 Mammalia Didelphimorphia Didelphidae 81

PAGE 82

Table 5-1. Continued GEN TAX Phylum/Class Order Family Taxa Common name 133 Mammalia Didelphimorphia Didelphidae Didelphis sp. opossums 134 Mammalia Didelphimorphia Didelphidae Didelphis marsupialis common opossum 135 Mammalia Didelphimorphia Didelphidae Chironectes minimus water opossum 136 Mammalia Didelphimorphia Didelphidae Philander opossum four eyed opossum 137 Mammalia Didelphimorphia Didelphidae Marmosa mexicana Mexican mouse-opossum 138 Mammalia Insectivora Soricidae Cryptotis micrura small-eared shrew 139 Mammalia Xenarthra (or Edentata) Dasypodidae 140 Mammalia Xenarthra (or Edentata) Dasypodidae Dasypus novemcinctus nine-banded armadillo 141 Mammalia Chiroptera bats 142 Mammalia Chiroptera Phyllostomidae: Stenodermatinae leaf nosed bats 143 Mammalia Chiroptera Phyllostomidae: Stenodermatinae Artibeus sp. 144 Mammalia Chiroptera Phyllostomidae: Stenodermatinae Artibeus sp., small 145 Mammalia Chiroptera Phyllostomidae: Stenodermatinae Artibeus jamaicensis Jamaican fruit-eating bat 146 Mammalia Chiroptera Phyllostomidae: Stenodermatinae Artibeus literatus great fruit-eating bat 147 Mammalia Chiroptera Phyllostomidae: Desmodontinae Desmodus rotundus 148 Mammalia Microchiroptera small bats 149 Mammalia Primata Cebidae Alouatta / Ateles howler/spider monkey 150 Mammalia Primata Cebidae Alouatta sp. howler monkeys 151 Mammalia Primata Cebidae Ateles sp. spider monkeys 152 Mammalia Primata Cebidae Ateles geoffroyi Central American spider monkey 153 Mammalia Primata Hominidae Homo sapiens sapiens Homo sapiens 154 Mammalia Carnivora 155 Mammalia Carnivora, large 156 Mammalia Carnivora, intermediate 82

PAGE 83

Table 5-1. Continued GEN TAX Phylum/Class Order Family Taxa Common name 157 Mammalia Carnivora, small 158 Mammalia Carnivora Canidae/Felidae cats and dogs 159 Mammalia Carnivora Canidae dogs, foxes, coyotes, wolves 160 Mammalia Carnivora Canidae, small foxes, small dogs 161 Mammalia Carnivora Canidae Canis sp. dogs, coyotes 162 Mammalia Carnivora Canidae Canis lupus familiaris domestic dog 163 Mammalia Carnivora Canidae Urocyon cinereoargentus gray fox 164 Mammalia Carnivora Procyonidae racoons, coatis 165 Mammalia Carnivora Procyonidae Procyon lotor northern raccoon 166 Mammalia Carnivora Procyonidae Nasua narica white nosed coati 167 Mammalia Carnivora Procyonidae Poto flavus kinkajou 168 Mammalia Carnivora Mustelidae 169 Mammalia Carnivora Mustelidae Spilogale / Conepatus sp spotted/hog-nosed skunk 170 Mammalia Carnivora Felidae 171 Mammalia Carnivora Felidae, large Puma concolor and Panthera onca 172 Mammalia Carnivora Felidae, intermediate ocelot, jaguarundi 173 Mammalia Carnivora Felid ae, small ocelot, margay 174 Mammalia Carnivora Felidae Leopardus pardalis ocelot 175 Mammalia Carnivora Felidae Leopardus weidii margay 176 Mammalia Carnivora Felidae Herpailurus yagouroundi jaguarundi 177 Mammalia Carnivora Felidae Panthera onca jaguar 178 Mammalia Perissodactyla Tapiridae Tapirus bairdii Baird's tapir 179 Mammalia Artiodactyla even-toed ungulates 180 Mammalia Artiodactyla, la rge white-tailed deer, peccaries 181 Mammalia Artiodactyla, intermediate brocket deer 182 Mammalia Artiodactyl a Tayassuidae peccaries 183 Mammalia Artiodactyla Tayassuidae Tayassu sp. peccaries 184 Mammalia Artiodactyla Tayassuidae Tayassu tajacu collared peccary 83

PAGE 84

Table 5-1. Continued GEN TAX Phylum/Class Order Family Taxa Common name 185 Mammalia Artiodact yla Cervidae deers 186 Mammalia Artiodactyla Cervidae Odocoileus virginianus white tailed deer 187 Mammalia Artiodactyla Cervidae Odocoileus hemionus mule deer 188 Mammalia Artiodactyla Cervidae Mazama sp. brocket deer 189 Mammalia Rodentia 190 Mammalia Rodentia, large 191 Mammalia Rodentia, intermediate 192 Mammalia Rodentia, small 193 Mammalia Rodentia Sciuridae squirrels 194 Mammalia Rodentia Sciuridae Sciurus sp. tree squirrels 195 Mammalia Rodentia Geomyidae pocket gophers 196 Mammalia Rodentia Geomyidae Orthogeomys sp. 197 Mammalia Rodentia Geomyidae Orthogeomys hispidus Hispid's pocket gopher 198 Mammalia Rodentia Heteromyidae pocket mice 199 Mammalia Rodentia: Myomorpha 200 Mammalia Rodentia: Myomorpha Muridae (Cricetidae) New World rats and mice, voles, hamsters, etc. 201 Mammalia Rodentia: Myomorpha Muridae (Cricetidae) Sigmodontinae Oryzomys sp. rice rats 202 Mammalia Rodentia: Myomorpha Muridae (Cricetidae) Sigmodontinae Ototylomys phyllotis big-eared climbing rat 203 Mammalia Rodentia: Myomorpha Muridae (Cricetidae) Sigmodontinae Sigmodon hispidus Hispid's cotton rat 204 Mammalia Rodentia: Myomorpha Muridae (Cricetidae) Sigmodontinae Peromyscus sp. deer mice 205 Mammalia Rodentia: Caviomorpha Erethizontidae Coendu sp. porcupines 206 Mammalia Rodentia: Caviomorpha Agoutidae/ Dasyproctidae 84

PAGE 85

85 Table 5-1. Continued GEN TAX Phylum/Class Order Family Taxa Common name 207 Mammalia Rodentia: Caviomorpha Dasyproctidae Dasyprocta puncata 208 Mammalia Rodentia: Caviomorpha Agoutidae Agouti paca 209 Mammalia Lagomorpha Leporidae Sylvilagus sp. rabbits 210 Mammalia Lagomorpha Leporidae Sylvigalus floridanus cottontail rabbit 211 Mammalia Lagomorpha Leporidae Sylvilagus brasiliensis forest rabbit

PAGE 86

Table 5-2. Element Numbers for Element Types El# Element El# Element El# El ement El# Element El# Element 1 carapace 32 quadrate 63 maxillary molar 2 94 humerus 125 metatarsal, vestigal 2 carapace/plastron 33 parasphenoid 64 maxillary molar 3 95 radius 126 tarsal centrali 4 3 plastron 34 pharyngeal, lower 65 maxillary molar 4 96 ulna 127 tarsal centrali + 4 4 scale 35 preoperculum 66 mandible 97 wing element 128 tarsometatarsus 5 scute 36 operculum 67 mandibular incisor 98 wing element + spur 129 phalanx 6 vertebra 37 cleithrum 68 mandibular incisor 1 99 radioulna 130 phalanx, proximal 7 vertebra, atlas 38 supracleithrum 69 mandibular in cisor 2 100 carpometacarpus 131 phalanx, intermediate 8 vertebra, axis 39 postcleithrum 70 mandibular incisor 3 101 carpal/tarsal 132 phalanx, distal 9 vertebra, cervical 40 branchiostegal ray 71 mandibular canine 102 metapodial 133 phalanx, terminal 10 vertebra, lumbar 41 pterygio phore 72 mandibular carnassial 103 metacarpal 134 phalanx/metapodial 11 vertebra, thoracic 42 basipt erygiophore 73 mandibular premolar 104 metacarpal 2 135 long bone 12 vertebra, saccral 43 pterygoid 74 mandibular pr emolar 1 105 metacarpal 3 136 postcranial element 13 coccyx 44 spine 75 mandibular premolar 2 106 metacarpal 4 137 element identifiable 14 vertebra, precaudal 45 spine, dorsal 76 ma ndibular premolar 3 107 metacarpal 5 138 bacculum 15 vertebra, caudal 46 spine, pectoral 77 mandibular premolar 4 108 innominate 139 worked bone 16 vertebra, ultimate 47 antler 78 mandibular molar 109 femur 140 shell 17 rib 48 cranium 79 mandibular molar 1 110 patella 141 valve 18 sternum 49 maxilla 80 mandibular molar 2 111 tibia 142 dactyl 19 keel 50 maxillary canine 81 mandibular molar 3 112 fibula 143 UID 20 neurocranium 51 maxillary carnassial 82 mandibular molar 4 113 synsacrum 144 tibiofibula 21 basioccipital 52 maxillary incisor 83 tooth 114 tibiotarsus 145 supraethnoid 22 vomer 53 maxillary incisor 1 84 incisor 115 astragalus 146 radius/ulna 23 angular 54 maxillary incisor 2 85 canine 116 astragalus/calcaneus 24 articular 55 maxillary incisor 3 86 premolar/molar 117 calcaneum 25 dentary 56 maxillary premolar 87 premolar 118 metatarsal 26 ceratohyal + epihyal 57 maxillary premolar 1 88 molar 119 metatarsal 3 27 ceratohyal 58 maxillary premolar 2 89 pre-coracoid 120 metatarsal 3 or 4 28 epihyal 59 maxillary premolar 3 90 coracoid 121 metatarsal 4 29 hyoid 60 maxillary premolar 4 91 scapula 122 metatarsal 5 30 hyomandibular 61 maxillary molar 92 scapula/coracoid 123 metatarsal 5, cf 31 premaxilla 62 maxillary molar 1 93 clavicle 124 metatarsal, cf 86

PAGE 87

Table 5-3. Body Portion Nu mbers for Body Portions BP # Body Portion 0 UID 1 cranial 2 axial 3 axial, pectoral 4 axial, pelvic 5 front limb, upper 6 front limb, lower 7 front limb, distal 8 hind limb, upper 9 hind limb, lower 10 hind limb, distal 11 postcranial 12 distal Table 5-4. Sidedness Numbers for Element Sidedness L1/R2 Element Sides 0 N/A 1 Left 2 Right 3 Both Table 5-5. Age Class Numbers for Age Classes Age # Age Classes 0 N/A 1 Juvenile 2 Juvenile + Immature 3 Subadult + Adult Table 5-6. Burning and Charring Numbers for Burning and Charring Descriptions Bu/Ch # Burning/Charring 0 None 1 Blackened 2 Calcined Gray 3 Calcined White 87

PAGE 88

CHAPTER 6 RESULTS Introduction Two methods of analysis were used in this study, a visual or descriptive comparison, and a statistical analysis using ArcGIS The comparative use of these methods is appropriate for this study because there were a limited number of excav ations done at most sites, and the sample sizes within any of these excavations were sma ll, therefore the interpre tations available through GIS analyses were limited. The spatial statis tical methods used in this study, Spatial Autocorrelation (Morans I) and Cokriging, both n eed specific sample sizes for accuracy as identified by ArcGIS. Spatial autocorrelation requ ires a sample size of more than ten and ideally above thirty specimens and cokriging needs a sa mple size of ten units or surface collections. Larger taxonomic groupings that included multiple species were used to increase sample sizes for the GIS analysis. Therefore, the patterns revealed by the GIS study are best interpreted with reference to the specific species and specime n distributions which are assessed using the descriptive visual comparison. In this chapter, the distribution of taxonomi c groups within each cav e site as revealed by both visual assessment and GIS analyses. When la rge enough sample sizes are available I also include information about the elements, sides, an d regions of the faunal remains in the visual analysis. The GIS spatial statis tical methods used in this st udy, spatial autocorrelation and cokriging, both provide informa tion on specific patterns. The speci fic information about both of these spatial statistical tools is taken from th e ArcGIS computer program Spatial autocorrelation can be defined as the evenness of distribution of variable, in this case taxonomic groups, throughout a defined space or area, in this case th e entire cave site. For the descriptions of each cave site below, distribution is defined as clus tered, dispersed, or random. Clustered remains are 88

PAGE 89

identified as having a positive autocorrelation and falling into defined spatial groupings. Dispersed remains are identified as having a negative spatial autocorrelation and a spatial pattern that varies across the site. Random patterns are defined as those that show no spatial autocorrelation and are neither clustered nor disp ersed. Small sample sizes can skew the results and the lowest number of samples for an accurate measurement for spatia l autocorrelation is thirty. In those cases, spatial autoco rrelation is defined as failed. Kriging is defined as a geostatistical analysis that identifies the spatial relationship of a specific single dataset (in this case the taxonomic groups). It can be used to identify areas of possible high concentration and to quantify the evenness of spatial distribution of remains across a site. Therefore, while spatial autocorrelation defines a distribution as being random, clustered, or dispersed, it is cokriging that identifies the area of clusteri ng and quantifies the evenness of the dispersion of the animal remains. Cokr iging analyzes the same dataset and spatial relationships but allows for simultaneous analysis of multiple variables or, in this case, different spaces within the cave site. Therefore, cokr iging was used to analyze the taxonomic group distributions in relation to the separations of sp ace within each cave. This test compared NISP values in caves with more than ten excavated unit s. The interpretive value of this test is limited by the patchy nature of excavation, so it can only reveal clustering between operations/units, not over the entire surf ace of the site. The surface view (presented in the chapter figures) is an estimate or assumption of continuous patterning based on the di stributions revealed in the units /operations for which data is available. Cokriging maps use a graded scale of color to represent areas of low and high concentrations. Therefore, not al l areas fit into a specific and easily identifiable grouping and a brief description of the area and possible distri butions noted by cokrig ing must be provided. 89

PAGE 90

Therefore in this chapter, areas of higher con centration and possible di stribution are described for each taxonomic group that was processed using cokriging analysis. Distributions defined by cokriging are defined as being a ssociated with a specific locale, including an individual unit or group of units, or as being within or between certain areas within the site. Cokriging analysis is described by the specific areas of highest concentrations in rela tion to actual units or by their relationship between or adjacent to specific areas within the site. For both spatial autocorrelation and cokriging, a small sample size or limited dist ribution can lead to exaggerated or failed attempts for these to programs to analyze th e spatial distribution of remains at the site. I present measures of ubiquity, taxonomic dist ributions as described above for GIS spatial analyses in relation to each of the cave spaces in this study, and relative proportions of the different taxa, elements, and sides, and burned specimens. For this analysis, ubiquity is the measure of the number of operations, units or surface collections that contained specific taxonomic groups. All other values are based on relative proportion of NISP or number of identified specimens in the group. Sample size issues are discussed for specific taxonomic groups to aid in interpreting the GIS and visu al assessment results. I present some limited information on species habitats and habits wher e it pertains to the specific cave findings, although generalized details of species habitats, uses, and meanings are presented in Chapter 7, Interpretation. Caves Branch Rockshelter Cave Branch Rockshelter (CBR) is made up of six operations, and th e materials excavated during the 2005 and 2006 seasons are found in four of these areas, Opera tions 1A, 1B, 1C, and 1D (Figure 4-2). These four opera tions are located in the northern (Op. 1A), central (Op. 1B and Op.1D) and southern (Op. 1C) regions of the rock shelter. Within these four operations were 24 units that contained faunal rema ins and were included in this analysis. Op 1A contained eight 90

PAGE 91

units with faunal remains, Op 1B had nine units Op 1C had four units and Op 1D had three units. These 24 units are on a grid system, each measuring as 1 mete r by 1 meter square. The spatial divisions, or sepa rations of cave space, defined for CBR included the left (1A) versus the right (1B, 1C, 1D) sides of the cave, northern (1A) versus southern (1B, 1C, 1D) parts of the cave, and the dark, cave, (1D) and light (1A, 1B, 1C) regions of the cave (Figure 6-1, Table 6-2). Location and placement of the excavati on units did not allow for any distinctions to be made between the eastern and western parts of the site. Due to the open nature of the rockshelter and no architectural impasses, no di stinctions were made for areas of open and restricted parts of the site. A total of 381 remains were id entified to specific taxonomic gr oups out of the total sample of 1276 specimens identified at CBR. These, listed in ranked order from high to low relatively frequency, include Dasypus novemcinctus (7.99%), Crustaceans (5.09%), Testudines (3.29%), Serpentes (2.90%), Rodentia (2.35%), Agou tidae/Dasyproctidae (1.88%), Aves (1.65%), Artiodactyla (1.41%), Actinoptery gii (1.02%), Cervidae (0.86%), Odocoileus virginianus (0.71%), Didelphidae (0.63%), Sauria (0.47 %), Amphibia (0.39%), Procyonidae (0.31%), Canidae (0.24%), Tayassuidae (0.24%), Chiroptera (0.08%), Mazama sp. (0.08%), and Sylvilagus sp. (0.08%) (Table 6-1). In total the spa tial distribution of twenty taxonomic groups are described in detail below, along with details of body portion, body side, and specimen treatment. Crustaceans A total of 65 remains were id entified as Crustaceans (Decapod a, infraorder Brachyura) at the site of CBR. Most of the units (20 out of th e 24 units and four of four operations) contained crustacean remains (Figure 6-1). Spatial autocorrelation identified the cr ustacean remains as being clustered and having less than 1% likelihood that this cluste red pattern could be the result 91

PAGE 92

of random chance. Cokriging of the site showed that a higher concentration of crustacean remains were located in Operation 1A than in other site areas (Figure 6-6). This can be interpreted as a higher co ncentration of crustaceans (crabs) in relation to th e left (n=33, 50.8%) versus right (n=32, 49.2%) sides of the cave, th e northern (n=33, 50.8%) versus southern (n=32, 49.2%) part of the cave, and the light (n=62, 96.9%) versus dark (n =2, 3.1%) regions of the cave. These higher concentrations were also identified in the visual analysis. Operation 1A (n=33, 50.8%) included the largest concentration of crustacean remains at the si te. Although there are drastic differences between the light and dark regions of the cave, there are no real distinctions in the concentration of remains in the left over th e right side of the cave or the northern over the southern parts of the cave. Visual inspection sh owed that the crustacean remains were almost ubiquitous throughout the site and were located in all excavated areas of the cave. Almost all of the remains (n=64, 98.5%) were dactyl or claw fragments and were the distal part of the front limbs. There are no differences in the proporti on of left (n= 28, 50 %) and right (n=28, 50%) elements of the crustaceans remains within the site. None of these remains were burned. Actinopterygii A total of thirteen Actinopteryg ii (fish) remains were identif ied throughout the site. These were found in each operation although only in se ven of the 24 units excavated. Cokriging was not possible because of the limited number of samp les. Spatial autocorrel ation testing indicated that the materials were slightly clustered, but probably randomly distributed at the site. There are no areas that GIS identified as having a specific concentration of fish remains. Only one bone (n=1, 7.7%) was specifically identified to a genus, Sparisoma sp., or the parrotfish family. The rest were unidentifiable (n=12, 92.3%) and could be the remains of either local freshwater fish or other marine fish. CBR is located adjacent to a ri verine area and fish remains may be from this local source. The parrotfish, which is a saltwate r species, could only have gotten to the site by 92

PAGE 93

human activity since the site is located many hundreds of mile s from the coast. This single bone was located in Unit 30G, Operation 1C, located on the right side of the cave, in the southern most part of the excavated region, and in the light area of the site (Figure 6-1). The lack of any other identified taxa limits identification of any specific patterning in fish remains at the site. The fish remains were distributed with more rema ins on the right (n=10, 76.9%) over the left (n=3, 23.1%) sides of the cave, the southern (n=10, 76.9%) over northern (n=3, 23.1%) part of the cave, and the light (n=7, 53.8%) over the dark (n=6, 46.2%) regions of the cave. Amphibia A total five specimens were identified to the class of Amphibia (amphibians) at CBR and all of these remains were identified as bei ng from the Order Anura (frogs and toads). The remains were only located in Operations 1A and 1B and within five out of the 24 units (Figure 61). Small sample size and the limited distribution of amphibian remains did not allow use of the spatial analyses programs (spatial autocorrelati on and cokriging). There was a slightly higher number of remains in the right and south (n=3, 60%) than the left and north (n=2, 40%) sides of the cave, and all of the remains were found within the light (n=5, 100%) rather than in the dark (n=0, 0%) regions. These five bones do not demo nstrate any specific pattern and there are no higher concentrations of these remains in different parts of the site. Amph ibia, particularly the large Bufo marinus, are known residents of caves (Emery, pers. comm. 2009). There is a slow moving river located near th e site (Bonor 1998, 2003; Wrobel 2008; Wrobel and Tyler 2006) and may be where these amphibian remains came from. The majority of these remains were from the appendicular region of the body including humer us (n=2, 40%), femur (n=1, 20%), and a tibiofibula (n=1, 20%). One bone (20%) was iden tified from the axial region of the body, and was identified as an innominate fragment. This distribution suggests a fairly complete skeletal representation. Three remains were sided, tw o to the left and none to the right. 93

PAGE 94

Testudines A total of 42 Testudines (Turtl e) specimens were identified at the site of CBR and these remains were located in all of the operations a nd within 16 of the 24 units Most of the remains (n=22, 52.4%) were identified within the order Testudines; however there were three specific taxonomic groups identified, including Kinostern idae (mud and musk turtles) (n=17, 40.5%), Trachemys scripta (common slider) (n=2, 4.8%), and Dermatemys mawii (Mesoamerican river turtle) (n=1, 2.4%). Spatial Autocorrelation id entified a random pattern of distribution of Testudines remains at CBR while cokriging identifie d some patterning that was not centered on a single operation. The highest area of cokriging occurred between Operations 1A, 1B, and 1D in an area not specifically identifiabl e by an operation or unit (Figure 6-7). Visual analysis of the distribution of turtle remains s hows that the left and north side s of the cave contained 14 (33.3%) turtle specimens while the right and south si des contained 28 (66.7%) specimens. The light region had 36 (85.7%) specimens and the dark re gion had only six (14.3%) remains. Most of the turtle specimens were from the carapace/plastron (n=39, 92.8%) or axial region of the skeleton with a few of the remains from the pelvic re gion (n=2, 4.8%) or long bone s (n=1, 2.4%). Eleven of the remains could be sided and these incl uded nine right elements (81.8%) and two left elements (18.2%). CBR is located adjacent to a river and many of these turtles may have been easily captured and utilized from this area. Sauria A total of six remains were identified as being Sauria (Lizard) remains and within each operation at least one unit contained these remain s. Five out of the 24 units contained lizard remains including units 10G, 12G, 23J, 30H, and 21O (Figure 6-1). These remains were evenly distributed in the left (n=3, 50 %) and right (n=3, 50%) sides of the cave, and the northern (n=3, 50%) and southern (n=3, 50%) parts of the cave. There was a slightly higher concentration of 94

PAGE 95

lizard remains in the light (n= 5, 83.3%) over the dark (n=1, 17.7%) regions of the cave. Four out of the six lizard remains were specifically identi fied as being from the Iguanidae (iguana) family, located in 10G, 21O, and 23J. These iguana rema ins were located in equal amounts in both the northern and left sides (n=2, 50%) and right and southern sides (n =2, 50%) of the cave and fewer remains were identified the dark region (n=1, 25%) than in the light region (n=3, 75%) of the cave. Spatial Autocorrelation identified random patterning. The sample size was too small and there were too few remains for cokriging. Visual inspection showed no sp ecific patterns in the overall distribution of either the general group of lizards or the igua na family. The iguana remains were identified as humerus (n=1, 25%), mandible (n=1, 25%), maxilla (n=1, 25%), and innominate (n=1, 25%) fragments. The remains iden tified to Sauria were all vertebrae. There were four elements, all identifie d as iguana, that were sided and three were right-sided and one was left-sided. Serpentes A total of 37 Serpentes (snakes) remains were identified, distributed in all four operations and in 13 of the 24 units at CBR. Spatial Autoco rrelation identified a ra ndom pattern that was recognized as being neither cl ustered nor disperse d throughout the site. Cokriging identified Operation 1D as the region with the highest interpolation, or the highest c oncentration, of snake remains (Figure 6-8). The disparity between the random distribution identified by spatial autocorrelation and the region of high interpolat ion with cokriging is due to the small sample size. Having such small samples does not allow for these spatial analysis tools to identify true patterns within the site. These findings identified a higher concentration of remains within the right and southern (n= 31, 83.8%) section over the left and northe rn (n=6, 16.2%) section of the cave, and the light (n=22, 59.5%) over the dark (n=15, 40.5%) region. A large number of snake remains (n=15, 40.5%) were located within the da rk region of the site in a single unit, 21O, 95

PAGE 96

suggesting that in fact the dark region has a much higher concentration of snake remains than the light region despite the GIS findings All of the remains were vert ebrae, no cranial segments or ribs were recovered. Some of the remains had signs of burning (n=13) in units 24J, 23H, and 21O, located in the central portion of the site. Te n of these burnt remains were located in the small cave-like opening of 21O and were lo cated in the dark region of the site. Aves A total of 21 remains were identified to the cl ass Aves (birds), with these remains being on the smaller size scale and probably from smalle r passerines. These remains were found in Operations 1A, 1B, and 1C and within 11 out of the 24 units. Spatial autocorrelation identified the pattern as being random but showing a slig ht clustering within th e site (z-score = 1.22). Cokriging identified a higher concen tration of remains in Operation 1A, which is located in the left and northern sides of the cave. Visually ther e were similar distributions of bird remains on the left (n=11, 52.4%) and the right (n=10, 47.6%) sides of the cave and the northern (n=11, 52.4%) and the southern (n=10, 47.6%) regions of th e cave. One of the patterns formed by the distribution of bird remains is the lack of rema ins within Operation 1D or the dark region of CBR. All other remains are found within the other three operations located under th e rockshelter dripline in the light region. The majority of the elements were from the appendicular (n=14, 66.7%) regions of the body, with a few axial (n =4, 19%) and some unide ntifiable (n= 3, 14.3%) remains. Only four of the 21 elements could be sided, and there was a higher proportion of right (n=3, 75%) than left (n=1, 25%) sides. Didelphidae A total of eight remains were identified to the family Didelphidae (opossum) and these remains are evenly distributed throughout all of the operations, each containing two remains, but only within 7 out of the 24 units. Spatial autocorrelation found th e pattern of opossum remains to 96

PAGE 97

be clustered with a less than 5% chance of this being a random distribution. The limited distribution of remains in only se ven units did not allow for cokrig ing to be calculated Through a visual analysis the majority of remains were located on the right (n=6, 75%) over the left (n=2, 25%) side, in the southern (n=6, 75%) rather th an the northern (n=2, 25 %) areas, and the light (n=6, 75%) over dark (n=2, 25%) pa rts of the cave. The elements of opossums identified at the site include, in order of frequency, cranial (n=4, 50%), axial (n=3, 37.5%), and humerus (12.5%) remains. Seven of the remains were sided, in cluding five right (71.4% ) and two left (28.6%) sided elements. The small sample size and even distribution of remains throughout the site make it difficult to identify specific pa tterns of spatial distribution. Dasypus novemcintus Dasypus novemcintus (nine-banded armadillo) represented the highest amount of specimens from all of the taxonomic categories, with a total of 102 remain s. These remains were distributed in all of the operations and within 19 out of the 24 units. Spatial autocorrelation identified the pattern of armadillo remains as being random and neither clustered nor dispersed within the site. Cokrigin g identified through interpolation that there was a higher probability for armadillo remains in the north over the south, the left over the right, and the dark over the light regions. Visual analysis confir med that there was a slightly higher concentration of remains within the right (n=52, 51%) over the left (n=50, 49%) side of the cave and the southern (n=52, 51%) rather than the northern (n=50, 49%) parts of the cave. These are the same areas that cokriging identified. However, more armadillo remains were found in the light (n=78, 76.5%) over the dark (n=24, 23.5%) areas of CBR than thos e identified in the patterns from cokriging. This may be due to the larger number per few ex cavation units of armadillo remains identified in the dark region than in the light regions of the cave. 97

PAGE 98

The number of armadillo specimens at the site (NISP) is an overestimate of the actual concentration of animals because a single armadill o can has an estimated 50 to 75 scutes on their nine bands for a total of betw een 450 and 625 scutes (McBee and Baker 1982). The sample from CBR contained 75 scutes (73.5%). Only 27 (26.5%) ar madillo specimens were identified to other elements. These non-scute elements are found in Operation 1A, 1B, and 1D. The non-scute elements included metapodials (n=15), vertebrae (n=6), radii (n=2), fibula (n=2), calcaneum (n=2), humeri (n=1), ulna (n=1), tibia (n=1), an d astragalus (n=1). Eleven elements were sided and seven were from the right side of the body (63.7%) while four were from the left (36.3%). The right elements are located in Operations 1A and 1D, while the left elements are located in Operations 1A and 1C. The sample size is too small to consider a specific meaning for the lack of both sided elements in a ll regions within the cave. Chiroptera Only a single Chiroptera (bat) bone was identified at the site of Caves Branch Rockshelter. This was found in one operation (1B) and unit (23J ) (Figure 6-1). Due to the lack of Chiroptera remains from the site none of the spatial analyses were possible. I was al so unable to identify a specific pattern from the single le ft innominate fragment located in a unit that was part of the right side, southern, and light regions of the cave. CBR is the onl y site from this study that lacks a larger sample of bat remains since caves and rockshelters are known ha bitation sites for these animals. Canidae A total of three bones were iden tified to the family Canidae (coyotes, dogs, and foxes). At CBR, all three bones we re identified as Canis lupus familiaris (domestic dog). The bones were located in Operations 1A and 1C and within onl y three units, but they do not seem to form a specific pattern. Due to the small sample size no GIS spatial analyses, spa tial autocorrelation and 98

PAGE 99

cokriging, were performed. The distribution of remains were concentrated on the left (n=2, 66.7%) over the right (n=1, 33.3%) side, the nor thern (n=2, 66.7%) over the southern (n=1, 33.3%) portions, and the light (n =3, 100%) over the dark (n=0 0%) of the cave. The dog elements included a maxilla, cranium, and metata rsus. These parts of the body may be more of a representation of dog skin rather than the remains from a meal. Only two of these bones were identified to a specific side of th e body, and each side was represented. Procyonidae A total of four specimens were identified to Procyonidae (raccoon) family. These were identified in two out of the four operations and within three of the 24 units. Two of these remains were located in Operation 1A a nd the other two were located in Operation 1D. Due to the small number of identified remains, spatial correla tion and cokriging were not performed by ArcGIS. Operation 1A is identified as being part of th e left (n=2, 50%) and north ern (n=2, 50%) side of the site, and Operation 1D is in the dark (n=2, 50%) region of the site. All of the remains were evenly distributed in each of the dichotomies analy zed in this site. The elements identified to the raccoon family include two uln ae (50%) and two tibiae (50%). Three of the raccoon remains were from the left side (75%) of the body, while only one (25%) of the remains was from the right side. These element sides did not form any patterns in their distribution because they are even distributed within all of the spatially distinct areas. Artiodactyla A total of eighteen remains were identified to the order Artiodactyla (even-toed ungulates). The general taxonomic group of ar tiodactyls, including the Taya ssuidae (peccary) and Cervidae (deer) families were used for analysis because the larger grouping provided a larger sample for analysis. They are an appropriate grouping because these animals together represent the largest and most commonly used species in the Maya region. Artiodactyls were found in all of the 99

PAGE 100

operations and within eleven of the 24 units. The spatial autocorrelation identified their distribution pattern as random and neither dispersed nor clustered within the site (i.e., evenly distributed). Cokriging identified areas of interpolation around Operation 1A in the north over the south direction and on the left over the right side. Cokriging also identifi ed the dark region of Operation 1D being more clustered than the li ght region of CBR. There were a higher number of artiodactyl remains in Operation 1A (n=9, 50%) than there are at the rest of the site. Spatially, there was an even distribution of specimens on th e left (n=9, 50%) and right (n=9, 50%) sides of the cave and also an even number of those rema ins in the north (n=9, 50%) and south (n=9, 50%) directions. There were more Artiodactyls in th e light (n=16, 88.9%) than in the dark (n=2, 11.1%). This distribution pattern may have been caused by the larger numbe r of units that were excavated in the light than in the dark parts of CBR. The two major family groups, Tayassuidae and Cervidae, have been analyzed separately be low and that section includes a discussion on the types of elements and siding of these elements. Tayassuidae A total of three Tayassuidae (or peccary) family specimens were located in three out of the four Operations (Op 1A, Op 1B, and Op 1C) and within three out of th e 24 units. Due to the small sample size and limited distribution of rema ins, spatial autocorrelat ion and cokriging were not performed. A majority of the peccary remain s were located on the right and southern (n=2, 66.7%) over the left and northern (n=1, 33.3%) side s of the cave. Peccary specimens were only found in the light (n=3, 100%) region and not the dark region (n=0, 0 %). All of the peccary remains at CBR were identified as tooth fragments. Two of these teeth were sided to the left side of the body. 100

PAGE 101

Cervidae A total of eleven remains were identified to the Cervidae (deer) family specimens. These were located in all four operations and in nine out of the 24 units. There are two major genera of deer species in the Maya region including Odocoileus virginianus (white-tailed deer) (n=9) and Mazama sp. (brocket deer) (n=1). Cervidae and white -tailed deer remains had sample sizes large enough for spatial autocorrelation and both were iden tified as randomly distributed at the site. Brocket deer remains were not spatial autocorrelat ed because of their small sample size. None of the Cervidae assemblages, incl uding the two separate genera, were large enough for cokriging. Each operation contained at least one remain from the Odocoileus virginianus but only one Mazama sp. remain was identified at the site, and it was found in Operation 1A. Only the remains from the white-tailed deer are addressed in relation to the spatial separations of space. A higher number of specimens were located on the right (n=5, 55.6%) over left (n=4, 44.4%) sides of the cave, the southern (n =5, 55.5%) over the northern (n=4, 44.4%) regions of the cave, and the light (n=7, 77.8%) over the dark (n=2, 22.2%) pa rts of the cave. The elements identified for the white-tailed deer included appendicular (n= 6, 66.7%), cranial (n=2, 22.2%), and axial (n=1, 11.1%) body portions. There was an even number of left (n=4, 50%) and right (n=4, 50%) sided elements from the white-tailed deer. The brocket deer element was a tarsal centrali 4 from the right side of the body located in unit 12G in Op 1A. The element identifie d only as Cervidae was a metapodial located in unit 21O in Op 1D or the cave. Rodentia The number of small Rodentia (o r rodent) remains should be hi gh at most cave sites both because caves serve as a home for these species and because many cave dwelling animals also consume them. However, a total of only 30 spec imens were identified at CBR. These remains were distributed throughout all of the Operati ons, with the largest concentration (n=14, 46.6%) 101

PAGE 102

found in Op 1B located on the right side, southe rn, and light parts of the cave. Rodent remains were also identified in fift een out of the 24 units. Spatial autocorrelation identified the distribution of rodent remains as random and being neither clustered nor dispersed within the site. Cokriging identified the c oncentration of rodent remains to be greater around Op 1B and Op 1C (right, south, and light regions ) than any other part of the cave. By the numbers, rodent remains were more likely to be located in the right (n=22, 73.3%) over left (n=8, 26.7%) side of the site, on the south (n=22, 73.3%) over the nort h (n=8, 26.7%) direction, and the light (n=26, 86.7%) over the dark (n=4, 13.3%) regions of CBR. Rodent remains were located in the appendicular (n=19, 63.3) and th e cranial (n=11, 36.7%) parts of the body. Twenty-six of these elements were sided and the la rger amount of rodent remains were from the left (n=19, 73%) than the right (n=7, 27%) side of the body. Agoutidae and Dasyproctidae The combined appearance of Agoutidae (paca) and Dasyproctidae (agou ti or tepescuintle) families are some of the largest rodents found within the Maya region. There were a total of 24 remains identified to these families. All four ope rations within eleven units contained Agoutidae and Dasyproctidae remains. Spatial autocorrela tion identified the dist ribution of remains as random, but did find some clusteri ng attributed to random chance. Cokriging identified a high interpolation around Operation 1D, which includes the area identified as being on the right side, the south direction, and the dark region of CBR. Visual patterning iden tified that the remains were evenly distributed on the right (n=12, 50%) and the left (n=12, 50%) sides and in the northern (n=12, 50%) and the southern (n=12, 50%) directions. There were more remains identified in the light (n=18, 75%) and than in the dark (n=6, 25%) regions of the cave. The majority of elements were tooth fragments (n= 18, 75%) and the rest of the remains were from 102

PAGE 103

the appendicular (n=6, 25%) part of the body. Ther e were slightly more left-sided (n=12, 52.2%) than right-sided (n=11, 47.8%) elements. Sylvilagus sp. A total of one Sylvilagus sp. (rabbit) specimen was identifi ed in the collection from CBR. This single bone was located in Op eration 1B and is in one out of the 24 units. Due to the low number of specimens and the limited distribution of remains, neither spat ial autocorrelation nor cokriging were processed. The rabbit remain was located on the right hand side, the southern region, and the light area of the cave. The single bone was identified as a metatarsal for which side could not be determined. Stela Cave Stela Cave (STC) contained ten excavation units of various sizes and only eight of these units contained faunal material. Chamber 1 cons isted of units 1-5 and 7, while Chamber 3 contained units 8 to 11 with only eight and elev en containing faunal remains. At the STC I was able to look at multiple spatial dichotomies includ ing the north versus the south, the east versus the west, the light versus the da rk, and the open versus the rest ricted areas within the cave. Chamber 1 is identified as the part of the cave within the left, north, ea st, light, and open area. Chamber 3 is identified as the section within th e right, south, west, dark, and restricted part of STC (Figure 6-2, Table 6-3). With only eight uni ts containing faunal remains, cokriging was unable to run a formal analysis because it needs ten units to run in ArcGIS. Spatial autocorrelation was used in those samples th at contained larger numbers of remains. A total of 529 faunal remains were identif ied into 21 specific taxonomic groups. This represents 30.8% of the 1718 remains identifi ed from the 2004 season. These groups include rodents (12.92%), armadillos (4.66%), turtles (2.50%), birds (2.27%), artiodactyls (1.75%), snakes (1.05%), deer (0.93%), bats (0.76%), opossums (0.76%), crustaceans (0.64%), pacas and 103

PAGE 104

agoutis (0.47%), peccaries (0.41%), ray-finned fishes (0.23%), dogs (0.23 %), white-tailed deer (0.23%), brocket deer (0.23%), squirrels ( 0.23%), cotton-tailed rabbi ts (0.23%), raccoons (0.17%), amphibians (0.06%), and lizards (0.06%) (Table 6-1). E ach of the taxonomic groups is discussed in taxonomic order below. Crustaceans A total of eleven remains were identified as crustaceans. All crustacean remains were located in Chamber 1, from Units 1 to 5 and 7 at STC (six out of eight units) (Figure 6-2). Spatial autocorrelation identified the distribution of re mains as being randomly distributed and neither clustered nor dispersed within the site. This may have been due to the small sample size and the need for more than thirty samples to get a mo re accurate identification of spatial patterning. However, it is interesting that the entire crustacean remains (n=11, 100%) were located within the part of chamber identified as being on the left side, in the northern an d eastern part of the cave, and in the light and open regions. Most of these remains were iden tified as dactyl (n=10, 90.9%) with only one remain (9.1%) identified as a shell fragment. These dactyl were sided to both the right (n=6, 66.7%) and the left (n=3, 33.3%). Actinopterygii A total of four remains were identified as ray-finned fishes. These four remains were located in Units 1, 2, 3, and 8 (four out of eight units) (Figure 6-2). With such a small sample size, spatial autocorrelation was unable to proces s. The remains were not specifically patterned within STC, but the majority of remains we re located in Chambe r 1 (n=3, 75%) and are associated with the left, north, east, light, and open area. while only one (25%) was identified in Chamber 3 which includes those remains identifi ed within the right, south, west, dark, and restricted areas of the cave. The elements identi fied for the ray-finned fi sh includes dentary (n=2, 104

PAGE 105

50%), a vertebrae (n=1, 25%), a nd an unidentifiable fragment (n=1, 25%). Both of the dentary remains were sided and there was a single right-s ided and left-sided bone in the collection. Amphibia One amphibian remain was identified at STC, located in Chamber 1, Unit 7 (Figure 6-2). Spatial autocorrelation was unable to process. The bone was located in Chamber 1 which associates it with left, north, east, light, a nd open regions of the cave. The specimen was identified as a left humerus. Testudines A total of 43 turtle remains were identified and all of these remains were located in Chamber 1, Units 1 to 5 and 7 (6 out of 8 units ) (Figure 6-2). Spatial autocorrelation identified the distribution of turtle remains as somewhat dispersed, but the pattern was still random in its distribution. The remains were located only in Ch amber 1 which was identified as being on the left, north, east, light, and open pa rts of the cave. The turtle remains were almost exclusively plastron and carapace fragments (n=42, 97.7%) with only one (3.3%) bone identified as a humerus. There were twenty elements that were sided and there was an even distribution of left (n=10, 50%) and right (n=10, 50%) sided elements. Sauria A total of one lizard remain was identified in Chamber 1, Unit 3 (Figure 6-2). Spatial autocorrelation was unable to process. The lizard remain was located in Chamber 1 and therefore within the left, north, east, light and open regions of the cave. This remain was identified as a mandible fragment and it could not be sided. Serpentes A total of eighteen specimens were identified as snake remains which were located in Chamber 1 (Units 2, 3, 5, and 7) and Chamber 3, Un it 8 (five of eight units) (Figure 6-2). Spatial 105

PAGE 106

autocorrelation identified the pattern somewh at clustered, but still random in its overall distribution. Most of the remain s (n=17, 94.4%) were identified in Chamber 1, in the north, east, left, open, and light regions of the cave. All of the serpent remains were vertebrae (n=18, 100%) and therefore from the axial part of the body. Aves A total of 39 remains were identified as bird and all of these remains were located in Chamber 1, Units 1 to 5, 7 (six out of eight un its) (Figure 6-2). Spatial autocorrelation identified the remains as the pattern bei ng random and neither dispersed nor clustered. All of the bird remains (n=39, 100%) are in Chamber 1, the north, east, left, open, and light part of the cave. The elements identified from the bird remains include limb elements (n=27, 69.2%), axial (n=9, 23.1%), unidentifiable (n=3, 7.7%). A total of 27 bird specimens were sided, and the right-sided (n=16, 59.3%) elements slightly outnumbered the leftsided (n=11, 40.7%) elements. Didelphidae A total of thirteen specimens were identified to the opossum family and they were all located in Chamber 1, Units 2 to 5, and 7 (f ive out of eight units) (Figure 6-2). Spatial autocorrelation identified the distribution of re mains as random and being neither dispersed nor clustered. Since all of these remains were identifie d in Chamber 1 they were associated with the northern, eastern, left, light, and open regions of the cave. Th e opossum materials were found from two portions of the body, including the cranial (n=7, 53. 8%) and the axial (n=6, 46.2%) regions. Six specimens were sided, all of which were mandible fragments, this included four left (66.7%) sided and two right (33.3%) sided elements. Dasypus novemcinctus A total of 80 specimens were identified to the nine-banded armadillo and were located in Chamber 1, Units 1 to 5, 7, and Chamber 3, Unit 8 (seven of eight units ) (Figure 6-2). Spatial 106

PAGE 107

autocorrelation identified the armadillo material as being somewhat clustered, but the pattern may have still been due to random chance. A majo rity of the remains were scutes (n=51, 63.8%), and these bones can number 450 to 625 scutes per armadillo (McBee and Baker 1982) and can therefore skew the representation of remains within the cave. Visually the armadillo material was distributed with 78 (97.5%) remains being loca ted in the left, north, eastern, light, and open regions of the cave, while 2 (2.5%) remains were located in the right, south, west, dark, and restricted parts of the cave. The two remains located in Chamber 3 were scutes and may have been unintentionally brought to this part of the cave, since these bones are small, light and easily transported. Along with the 51 scutes identified at the site there were ot her elements identified including vertebrae (n =4), rib (n=1), scapula (n=1), humeri (n=2), ulna (n=3), metapodials (n=7), innominate (n=1), patella (n=1), tibia (n=4), fibul a (n=2), astragalus (n=1), and calcaneus (n=1). All of these elements were located in Chamber 1. A total of 15 specimens were sided and the distribution of sided remains in cludes 9 left (60%) sided bone s and 6 right (40%) sided bones. Chiroptera Only thirteen remains were identified to order Chiroptera and all of these remains were located in Chamber 1, Units 1-5, and 7 (six of eight units) (Figure 6-2). Spatial autocorrelation found the pattern to be random, showing neither a clustered nor dispersed pattern. At STC, a large number of bats currently live in Chamber 3 and therefore it is here that we would have expected to find a larger amount of speci mens (personal communication Griffith 2009). However, bat specimens were only located in Ch amber 1, associated with the left, north, east, light, and open parts of the cave. The bat elements included a small number of cranial elements (n=3, 23.1%) and the other elements were front limb bones (n=10, 76.9%). There were seven bones that were sided; al l from the left side. 107

PAGE 108

Canidae A total of four remains were identified to th e canid family; including two remains from the domestic dog and two remains from the gray fox. The canid assemblage from STC was found in Chamber 1, Units 2, 3, and 4 and Chamber 3, Unit 11 (four of eight units) (Figure 6-2). Due to the small sample size spatial autocorrelation failed to process the canid remains. Three (75%) of the remains, two of which were fox remains, we re found in the left, nor th, east, light, and open regions of the cave and one (25%) specimen a dog remains, was found in the right, south, west, dark, and restricted areas. The elements incl uded three (75%) limb bones and one (25%) cranial fragment from a fox. Three of the canid bones we re sided and there were two (66.7%) left-sided, one of these bones was from the fox and one (33.3%) right-sided elements in STC. Procyonidae A total of three remains were identified to th e raccoon family, and all of these remains are located in Chamber 1, Units 3 and 5 (two of ei ght units) (Figure 6-2). Spatial autocorrelation failed because of the small sized sample of racc oon materials at STC. All three of the remains were located in Chamber 1, associated with the le ft, north, east, light, and open parts of the cave. The elements included one (33%) cranial fr agment, one (33%) metapodial, and one (33%) scapula. Two of these remains were sided, both to the left side. Artiodactyla A total of 30 artiodactyls remains located in both Chamber 1, Units 1 to 5, and 7 and Chamber 3 Unit 8 (seven of eight units) (Fi gure 6-2). Spatial autocorrelation identified the pattern of artiodactyl remains as random, not showing a dispersed or clustered pattern. The majority of remains were in Chamber 1 (n=27, 90%) rather than in Chamber 3 (n=3, 10%) and this may be due to the different proportion of excavation units in both of these areas. With the majority of artiodactyls in Chamber 1 these rema ins were identified with the left, north, east, 108

PAGE 109

light, and open regions of the cave. The elements were from the limb (20, 66.7%), cranial (9, 30%), and axial (n=1, 3.3%) parts of the body. A tota l of nineteen elements were sided, including eight (42.1%) left-sided and eleven (57.9%) right-sided bones. Th e artiodactyl materials were separated into two families, the peccary and deer, which are discussed below. Tayassuidae A total of seven specimens were identified to the peccary family and all of them were located in Chamber 1, Units 1, 2, 3, and 5 (four of ei ght units), therefore associated with the left, north, east, light, and open parts of caves (Fig ure 6-2). Spatial autocorrelation identified the peccary material as being randomly distributed within the site. The elements included three (42.8%) cranial fragments, two (28.6%) teeth fr agments, and two (28.6% ) metapodials. There were five sided elements, including two (40%) left-sided and three (6 0%) right-sided bones. Cervidae Sixteen specimens were identified for the deer family and these remains are located in both Chamber 1, Units 1 to 5, and 7, and Chamber 3, Unit 8 (six of eight units) (Figure 6-2). Of these, four were identified as white-tailed deer, located in Chamber 1, Units 1, 5, and 7, and four were identified as the brocket deer, located in Ch amber 1, Units 2, 4, 5, and 7. Spatial autocorrelation found the distribution of deer remains as bei ng neither clustered nor dispersed, and randomly distributed at the site. However, spatial auto correlation failed during processing for both the white-tailed deer and the brocket deer materi al because the sample size was too small for analysis. Most deer remains, and all securely identified white-tail and brocket deer, were concentrated in Chamber 1 (n=15, 93.8%), spatiall y associated with the left, north, east, light, and open areas of the cave. The elements were from the cranium (n=5, 31.3%), hind limb (n=6, 37.5%), front limb (n=2, 12.5%), metapodials (n= 3, 18.8%), and a scapula (n=1, 6.3%). A total of eleven deer elements were sided, seven (63.6 %) from the right side and four (36.4%) from the 109

PAGE 110

left side. There was a higher proportion of right (n=3, 75%) to left (n=1, 25%) sided elements from the white-tailed deer and from the broc ket deer (right n=2, 66.7%, left n=1, 33.3%). Rodentia The largest concentration of remains at the STC, both numerically a nd spatially, was from the rodent order with a total of 222 remains. These were distributed throughout all of the units in Chamber 1 and Chamber 3 (eight of eight units ) (Figure 6-2). Spatial autocorrelation of the rodent specimens identified them as being random distributed and neither dispersed nor clustered. A total of 181 (81.5%) re mains were located in areas id entified as left, north, east, light, and open parts of the cave, while 41 (18.5 %) specimens were located to the right, south, west, dark, and restricted regions of the cave. The rodent remains were from the hind limb (n=120, 54.1%), cranium (n=67, 30.2%), front limbs (n=22, 9.9%), and axial region (n=13, 5.8%). A total of 213 elements were sided to the left (n= 114, 53.5%) and right (n= 99, 46.5%) sides. Scuiridae Four specimens of the family Scuiridae (squi rrels) were found in Chamber 1, Units 1 and 4 (two of eight units) (Figure 6-2). With only four remains identified to the squirrel family, spatial autocorrelation failed to process. Since the rema ins were only located in Chamber 1 they were distributed in the left, north, east, light, and open regions of the cave. The elements included a mandible (25%), two femurs (50%), and 1 tibia ( 25%). The majority of the elements were sided to the left (n=3, 75%) over the right (n=1, 25%). Agoutidae and Dasyproctidae A total of eight remains were identified to the Agoutidae and Dasyproctidae families in both Chamber 1, Units 1 to 5 and Chamber 3, Unit 8 (six of eight units) (Figure 6-2). These two families were processed using spatial autoco rrelation and the remains were found to be 110

PAGE 111

somewhat dispersed, but pattern was still due to random chance. The majority of remains (n=7, 87.5%) were identified to the left, north, east, light, and open parts of the cave. The element distribution includes five (62.5%) teeth, one (12.5%) ulna, one (12.5 %) tibia, and one astragalus (12.5%). A total of seven elements were sided; there were four (57.1%) left-sided elements and three (42.9%) right-sided elements. Sylvilagus sp. A total of four remains were identified as rabbit, and were located in both Chamber 1, Units 1, 2, and 4 and Chamber 3, Unit 8 (four of ei ght units) (Figure 6-2). Spatial autocorrelation failed to process these remains. The rabbit remain s were mainly concentrated (n=3, 75%) in the left, north, east, light, and open re gions of the cave. All four bones were sided and there was an even distribution of two (50%) right-sided and two (50%) left-sided elements. Cueva de El Duende Cueva de El Duende (CD) consists of three excavated operations that were analyzed for this study including CD1 located ne ar a collapsed area, CD2 located within an open area, and CD3 which is located within the dark regions of the cave. The separati ons of space within CD include the north versus the sout h, the east versus the west, the dark versus the light, and the open versus the restricted regions of the cave (Figure 6-3). Each separation is described using the specific units that were excavated in this cave. The northern part of the cave is represented by CD1-1, CD1-2, CD1-3, CD1-4, CD2-0-3, and CD3-1-1 (75%), while the southern part of the cave includes CD2-1 and CD2-2 (25%). Split along th e other directional axis the western part of the cave is represented by CD1-1, CD1-2, CD1-3, CD1-4, and CD3-1 (62.5%), while the eastern part of the cave includes CD2-0-3, CD2-1 a nd CD2-2 (37.5%). The light and open regions include CD2-0-3, CD2-1, and CD2-2 (37.5%), those units located in the dark and restricted regions include CD1-1, CD1-2, CD1-3, CD 1-4, and CD3-1 (62.5%) (Table 6-4). 111

PAGE 112

There were a total of eighteen taxonomic gr oups and 2496 total specimens identified at the site of Cueva de El Duende. These groups incl ude bats (25.32%), ray-finned fishes (15.99%), snakes (2.00%), birds (1.48%), even-toed ungulat es (1.20%), opossums (1.16%), canids (0.88%), rodents (0.84%), deer (0.72%), white-tailed deer (0.60%), tur tles (0.60%), amphibians (0.44%), cats (0.32%), pacas and agoutis (0.32%), lizards (0.20%), armadillos (0.08%), brocket deer (0.04%), and crustaceans (0.04%) (Table 6-1). Each of these taxonomic groups is discussed in detail below. When applicable, information a bout the elements or body portions and element siding are included. Faunal remains were only identified in eight excavation units and therefore cokriging, which requires ten units, was not processed. Spatia l autocorrelation was also limited because of the small sample sizes since this program also needs at least thirty samples to function accurately. Therefore, most of the analysis fr om Cueva de El Duende relied on the visual analysis of the site and the overall dist ribution of these remains at the site. Crustacean One crustacean specimen was identified and it was located in CD2-1, in the southern, eastern, light, and open region of the cave (one of eight units) (Figure 6-3) .The single crustacean remain was a claw fragment that could not be sided. With only one specimen, little can be said about this finding. Cueva de El Duende is located near the El Duende pyramid and is part of an underground river system which connects five of the cave sites at Dos Pilas (Minjares 2003) and the crustacean fragment maybe from this nearby river. Actinopterygii A total of 399 remains were identified to the ray-finned fish class. All of these remains were located in excavation units near the ope ning of the cave in CD1-1, CD1-4, CD2-0-3, and CD2-1 (four of eight units) (Figure 6-3). Spatia l autocorrelation identified the pattern of 112

PAGE 113

distribution of the fish remains as being random and neither clustered nor dispersed at the site. There were differences in the di stribution of fish remains within the cave, with the majority located in the southern (n=359, 90%) over the northern (n=40, 10%), and the eastern/light/open part (n=377, 94.5%) over the west ern/dark/restricted part (n =22, 5.5%). Although there were many fish remains identified in this cave, very few (n= 48, 12%) of these were identified into more specific taxonomic groups. Of these, 39 (9.8%) were identified to the family Lepisosteidae (or gars) which were only found in CD2-0-3, and CD2-1 (Figure 6-3). The gar remains identified include a majority of ganoid scales (n-25), cranial fragments (n=10) and vertebrae (n=4) fragments. There were also eight (2%) remains from the order Siluriformes (or catfish) located in CD1-1, CD1-4, and CD2-1 (Figure 6-3). Both gars and catfish are freshwater species and were probably from the surrounding areas, but ar e not inhabitants of the underground rivers connecting five of the caves at Dos Pilas. The fish remains identified at CD included unidentifiable remains (n=148, 37.1%), cranial (n=51, 12.8%), ri bs (n=76, 19%), vertebrae (n=39, 9.8%) ganoid scales (n=25, 6.3%), spines (n=19, 4.8%), and axial fragments (n=60, 15%). A total of 43 fish remains were sided and these bones included 25 (58.1%) right-sided and 18 (41.9%) left-sided fish bones. Amphibia The class Amphibia (or Amphibians) was repr esented by very few remains (n=11), most from the order Anura or frogs and toads. Am phibian remains were found in CD1-4 and CD2-1 (two of eight units) (Figure 63). Most were found in the so uth/east/open/light region (n=7, 63.6%) as opposed to north/west/restricted/ dark region (n=4, 36.4%). The breakdown of amphibian elements identified at CD include 4 (36.4%) innominates, 3 (27.2%) humeri, 2 (18.2%) long bones, 1 (9.1%) vertebra, and 1 (9.1 %) maxilla. A total of eight elements were 113

PAGE 114

sided and these were evenly distributed between the right (n= 4, 50%) and left (n=4, 50%) sides of the body. Testudines A total of 15 remains were identified as turtle at CD and these were located in CD2-1 (n=13, 86.6%), CD2-2 (n=1, 6.7%), and CD1-3 (n= 1, 6.7%) (three of eight units) (Figure 6-3). The majority of turtle remains (n=14, 93.3%) were located in CD2-1 and CD2-2 which are associated with the southern, eas tern, light and open regions of the cave. Although this is a small sample size, they do seem to cluster in this part of the cave. The elements identified from turtles includes the carapace (n=12, 80%), coracoid (n=1, 6.7%), humerus (n=1, 6.7%), and innominate (n=1, 6.7%). From these elements, nine of the tu rtle specimens were sided including six (66.7%) to the right side and three (33.3%) to the left sides of the body. Sauria A total of five lizard remains were identified in a single unit, CD2-1, which is located in southern, western, light, and open ar ea of the cave (Figure 6-3). Th e five lizard remains were all identified as dentary bones. Out of the five lizar d remains, two (40%) were identified as being from the iguana family. Three specimens were sided including two left (66.7%) and one right (33.3%). Serpentes A large number of snake remains (n=50) were identified at the site. The largest concentration was located in CD2-1 (n=45, 90%), and the other remains were identified in CD20-3 (n=2, 4%) and CD1-4 (n=3, 6%) (Figure 6-3). The highest concentrati on of snake materials was located spatially in the southern (n=45, 90% ) and eastern/light/open (n=47, 94%) regions of the cave. The snake remains included 38 (76%) ribs and 12 (24%) vertebrae. None of the snake materials were sided. 114

PAGE 115

Aves There are 37 remains that have been identified to the class of Aves (or bird). These remains were located within CD2-1 (n=32, 86.5%), CD 1-1 (n=1, 2.7%) and CD1-4 (n=4, 10.8%) (Figure 6-3). The highest concentration of the remains was in CD2-1 (n=32, 86.5%) which is located within the southern, eastern, light and open pa rt of the cave. The body portions of the bird remains from CD include hind limbs (n=15, 40.5%), front limbs (n=10, 27%), axial (n=6, 16.2%), cranial (n=3, 8.1%), and long bone fragme nts (n=3, 8.1%). A total of 25 elements were sided including 13 (52%) left-sided and 12 (48%) right-sided elements. Didelphidae A total of 29 opossum remains were identifie d at Cueva de El Duende. There were 18 (62.1%) identified within CD2-1 and a total of 11 (37.9%) fro m CD1. CD1-1 had three (10.3%) and CD1-4 had eight (27.6%) remains (Figure 6-3). The opossum remains were more highly concentrated in the southern, eastern, light, and open (n=18, 62 .1%) than the northern, western, dark, and restricted (n=11, 37.9 %) areas of the cave. The body portions represented by the opossum remains include the cranial (n=14, 48.3%), axial (n=5, 17.2%) front limb (5, 17.2%), and hind limb (n=5, 17.2%). A total of 26 remain s were sided and they were fairly evenly distributed between the left side (n=1 4, 53.8%) and the right side (n=12, 46.2%). Dasypus novemcinctus A total of two armadillo remains were identi fied from the collection at CD. These two bones were scutes and they were located in CD2-1 (Figure 6-3). CD2-1 is located in the southern, eastern, light, and open regions of the cav e. The small size of scutes and their mobility through water movement and trampling may have affected their dist ribution at the site. 115

PAGE 116

Chiroptera The largest proportion of identified remains fr om the faunal assemblage of Cueva de El Duende was from the order Chiroptera with a total of 632 identified specimens. These remains were found in the following Operations CD2-1 (n=509, 80.5%), CD2-0-3 (n=117, 18.5%), CD14 (n=5, 0.8%), and CD1-1 (n=1, 0.2%) (Figure 6-3). Spatial autocorrelation found the pattern of faunal remains to be somewhat clustered, but sti ll due to random chance. Th e largest part of the sample was located within the northern (n=509, 80.5%) rather than in the southern (n=123, 19.5%) part of the cave. There was a higher number of bats remains (n=626, 99.1%) in the eastern, light and open parts of the cave. The bat body portions identified from this faunal assemblage includes phalanges (n=309, 48.9%), front limbs (n= 224, 35.4%), axial (n=38, 6%), hind limb (33, 5.2%), and cranial (n=28, 4.4%) re gions of the body. A total of 116 (55.5%) bat elements were left-sided while 93 (44.5%) were right-sided. Canidae A total of 22 specimens were identified to the canid family, none to a more specific genus. These remains were located in units CD2-1 (n=18, 81.8%), CD2-2 (n=2, 9.1%), CD1-4 (n=1, 4.5%), and CD3-1-1 (n=1, 4.5%) (Figure 6-3). Spa tial autocorrelation identified the pattern of canid remains as random, and neither clustered nor dispersed. The majority of remains were located in CD2-1 and CD2-2 (n=20, 90.0%) which were both located in the southern, eastern, light, and open areas of CD. Th ere was one canid bone located in CD3-1-1 which is located in the deepest part of the cave that was excavat ed, therefore dark, restricted, and also on the northern and western parts of the cave. The elemen ts identified include te eth (n=8, 36.4%), front limbs (n=6, 27.2%), hind limbs (n=4, 18.1%), inno minates (n=2, 9%), a cranial fragment (n=1, 4.5%), and an atlas vertebrae (n=1, 4.5%). A to tal of 21 remains were sided, and these included ten (47.6%) left-sided and eleven (52.4%) right-sided remains. 116

PAGE 117

Felidae A total of eight Felidae (cat) remains were identified and all of these were located in CD21 (one out of eight units) associ ated with the southern, easter n, light, and open regions of the cave (Figure 6-3). All eight (100%) of the remain s were cranial fragments. A total of five (62.5%) of the cat remains were fr om the right side of the body, while only three (37.5%) were from the left side of the body. Artiodactyla A total of 30 specimens were identified to the artiodactyl order, the largest sized mammals at the site. The artiodactyl remains were distribute d within four units at the site including CD2-1 (n=14, 46.7%), CD1-1 (n=2, 6.7%), CD1-2 (n= 4, 13.3%), and CD1-4 (n=10, 33.3%) (Figure 63). Spatial autocorrelation identified the patt ern of artiodactyls remains as being random; showing neither clustered nor dispersed patter ning. The remains were distributed throughout the site in nearly even amounts with 16 (53.3%) sp ecimens located in the north, west, dark, and restricted parts of the cave, and the other 14 (46. 7%) remains located in the south, east, light, and open areas within the cave. The body portions of the artiodactyls at CD included the hind limbs (n=11, 36.7%), distal appendages (n=8, 26.7%), ax ial (n=5, 16.7%), front limb (n=4, 13.3%), and cranial (n=2, 6.7%) portions of th e body. A total of 13 elements were sided, and this included ten (76.9%) left-sided and three ( 23.1%) right-sided elements. Cervidae Within the artiodactyls, there were 18 bone fragments identified to the deer family, and none specifically identified to the peccary famil y. Deer materials were id entified in units CD2-1 (n=11, 61.1%), CD1-2 (n=1, 5.6%), and CD1-4 (n =6, 33.3%) (Figure 6-3). Within the deer family, fifteen specimens were identified as wh ite-tailed deer and one wa s identified a brocket deer. There were eleven (61.1%) cervid remain s within the southern, eastern, light, and open 117

PAGE 118

areas, while only seven (38.9%) remains were found in the other spatial distinctions including northern, western, dark, and restri cted areas within the cave. Th e body portions identified for the deer specimens include the hind limbs (n=9, 50% ), front limbs (n=4, 22.2%), distal appendages (n=2, 11.1%), cranial (n=2, 11.1%), and axial (n=1, 5.6%) portions. There was a total of eleven remains that were able to be sided including eight (72.7%) left-sided and three (27.3%) rightsided elements. The distribution of white-tailed d eer includes nine (60%) of the remains located in the southern, eastern, light, and open areas, while six of the remains were found in the northern, western, dark, and restri cted. The distribution of body el ements from the white-tailed deer include cranial (n=1), axia l (n=1), front limb (n=2), hind limbs (n=8), and distal (n=2). There was one brocket deer remain, a metacarpal, identified in the southern, eastern, light, and open regions of the cave. Rodentia Rodent remains were found in very small numb ers at Cueva de El Duende with a sample of twenty-one specimens. The rodent materials were identified in units CD1-2 (n=12, 57.1%), CD2-1 (n=5, 23.8%), and CD1-4 (n=4, 19.1%) (Figur e 6-3). The majority of Rodentia remains (n=16, 76.2%) were within the northern, western, dark, and restricted parts of the cave. The body portions of rodents within CD include the hi nd limbs (n=9, 42.9%), axial (n=7, 33.3%), cranial (n=4, 19%), and front limb (n=1, 4.8%) of the b ody. The numbers of sided elements were very similar with eight (53.3%) left-sided and seven (46.7%) righ t-sided elements identified. Agoutidae and Dasyproctidae A total of eight remains were identified to both Agoutidae and Dasyproctidae families. The majority of remains were located in CD2-1 (n=6, 75%). The remainder were located in CD1-4 (n=1, 12.5%) and CD3-1-1 (n=1, 12.5%) (Figure 6-3). The majority of these remains (n=6, 75%) were located within the southern, eastern, lit, and open parts of the cave. The elements included 118

PAGE 119

the tibia (n=3, 37.5%), humerus (n=2, 25%), femu r (n=2, 25%), and ulna (n=1, 12.5%). Six out of eight (75%) remains were sided to the right side while only two (25%) were sided to the left. Cueva de Sangre Cueva de Sangre (CS) is a formed by severa l long and thin passageways with multiple entrances into the cave (Figure 6-4). There were a large number of surf ace collections and some large soil samples that were defl occulated from the site by Jame s Brady and Anne Scott (1997). A total of 81 of these collecti on areas were found to contain faunal materials. Due to the large number of surface collections cokr iging analysis was able to process for those taxonomic groups with large samples. However, due to the shape of this cave many of the spatial separations were unable to process. A visual analysis was also conducted on the remain s from this site. For analysis of the left versus right side of the cave, only remains identified from CS1 (n=57) are used because the maps provided the best information for this division. The separation of space includes the left (n=26, 45.6%) versus the right (n=31, 54.4%) sides of th e cave, the northern (n=64, 79%) versus the southern (n=17, 21%) regions, and the eastern (n=20, 24.7%) versus western (n=61, 75.3%) parts of the cave (Figure 6-4) (Table 6-5). A total of sixteen taxonomic groups were identified from CS. These include rodents (10.07%), pacas and agoutis (4.61%), turtles ( 3.60%), even-toed ungulates (3.55%), opossums (3.29%), deer (2.97%), canids ( 2.07%), brocket deer (1.85%),wh ite-tailed deer (1.01%), bats (0.64%), crustaceans (0.42%), peccary (0.26%), birds (0.21%), snakes (0.16%), armadillos (0.08%), and ray-finned fishes (0.11%) (Table 6-1) Each of these taxonomic groups is discussed in detail below and includes elements or body por tions and side of elements when applicable. Crustaceans Eight crustacean specimens were identified w ithin CS1-7-1, CS1-9-1, and CS1-78-1 (3 of 81 surface collections) (Figure 6-4). Due to the lim ited distribution and small sample size of the 119

PAGE 120

crustaceans at CS, both spatial autocorrelation and cokriging failed to process. All of the crustacean remains were located on the right side, in the northern and eastern regions of Cueva de Sangre. These remains were also located in an area of the cave that was known to have a high amount of run-off and clayey soil at the site. All of the eight bones identified to crustacean were claw fragments and the siding of these remains include three (60%) right -sided and two (40%) left-sided claws. Actinopterygii A total of two remains were identified to the ray-finned fish and were located in a single surface collection, CS1-9-1 (Figure 6-4). The fish materials failed spatial autocorrelation and cokriging because of the small sample size and limited distribution. The small size and brittleness of fish remains may be a reason for these bones not being found in Cueva de Sangre. They were also not found in the large mud samp les that were later deflocculated, which was water screened through a fine mesh that was 4 mm in size (Brady and Scott 1997). The two bones identified as fish remains included one ve rtebra and one postcranial element. Neither of these bones could be sided. Testudines A total of 70 remains were identified as turtles, which were present within 30 out of the 81 surface collections at Cueva de Sangre (Figure 6-4) Spatial autocorrelation identified the pattern as being random and showing neither clustered nor dispersed distributi ons. Cokriging was only able to be processed for the right versus the left side of the cave, and the highest concentrations were located near the main entrance of the cave in CS1 (Figure 6-9). All of the turtle remains were located within the northern and eastern part of the cave while there were more remains located on the right (n=41, 58.6%) than the left (n=29, 41.4%) sides of the cave. Most of the turtle remains were from the carapace and pl astron (n=69, 98.6%) and there was also a single 120

PAGE 121

vertebra (n=1, 1.4%) fragment as well. Eleven remains were side d and they included both leftsided (n=6, 54.5%) and right-sided (n=5, 45.5%) elements. Serpentes A total of three snake remains were identified in Cueva de Sangre and they were found in surface collections of CS1-9-1 and CS1-13-1 (Fi gure 6-4). Both spatial autocorrelation and cokriging failed due to small sample size. There were slightly more snake remains located on the right (n=2, 66.7%) than the left (n=1, 33.3%) side s of the cave. All of the snake remains were located on the northern and eastern part of the ca ve, nearest to the main cave entrance of CS1. The snake remains were also from the axial part of the body (vertebrae) and none of these were sided. Aves A total of four remains were identified to the class of birds, and all were found in surface collections of CS1-77-2, CS1-78-1, and CS6-81 (Figure 6-4). Spatial autocorrelation and cokriging were both unable to process this taxonomic group. The majority of remains were located on the left (n=2, 66.7%) rather than the ri ght (n=1, 33.3%) sides of the cave. All of the remains were located in the north, and more were concentrated in the western (n=3, 75%) than the eastern (n=1, 25%) parts of the cave. The four remains were identified as one humerus, one ulna, one vertebra, and one unidentifiable fragme nt. There were only two remains that could be sided, including a left-sided hum erus and a right-sided ulna. Didelphidae A total of 62 specimens were identified to the family Didelphidae and found in five collection units including CS 1-9-1, CS1-46-1, CS1-78-1, CS1-772, and CS11-4-1 (Figure 6-4). Spatial autocorrelation identified 5 to 10% likelihood that this cluste red pattern is the result of random chance. Most of the remains were located in the right (n=33, 54.1 %) over the left (n=28, 121

PAGE 122

45.9%), and the north and west (n=61, 98.4%) over the south and east (n=1, 1.6%), respectively. The opossum elements identified at the site in cluded vertebrae (n=28, 45.2%), teeth (n=11, 17.7%), mandible (n=8, 12.9%), cranium (n=5, 8.1%), humerus (n=5, 8.1%), radius (n=2, 3.2%), scapula (1, 1.6%), ulna (n=1, 1.6%), and astr agalus (n=1, 1.6%). Ther e were a total of 32 remains sided, and these remains were evenly distributed between the left (n=16) and right (n=16) sides. Dasypus novemcinctus A total of three armadillo remains were identified within three surface collections including CS1-10-1, CS1-13-1, and CS6-8-1 (Figure 6-4). Due to the limited distribution and small sample size autocorrelation and cokriging could not be processed. The remains were only located on the left (n=2, 100%) side of the passage and the northern (n=3, 100%) regions of the cave, but the majority of remains were locate d on the western (n=2, 66.7% ) not the eastern (n=1, 33.3%) side of the cave. The armadillo elemen ts included two (66.7%) scutes and one (33.3%) caudal vertebrae. None of the armadillo remains from CS could be sided. Chiroptera A total of 12 bat remains were identified a nd were located in a two surface collections, CS1-78-1 and CS6-3-1 (Figure 6-4) Both spatial autocorrelation and cokriging were unable to process. Most of the remains were from surface collection CS1-78-1 (n=9) which includes the area that was deflocculated and water screened th rough a fine mesh size. The bat remains at CS were only located on the right (n =9, 100%) side and northern (n= 12, 100%) sections of the cave, and more were found in the western (n=9, 75%) than in the eastern (n=3, 25%) parts of this cave. The elements identified from the bat remains include humeri (n=4, 33.3%), ulnas (n=4, 33.3%), femurs (n=1, 8.3%), and phalanx (n=3, 25%). Five specimens were sided and their distribution was similar including left (n=3, 60%) and right (n=2, 40%) sides. 122

PAGE 123

Canidae Cueva de Sangre contained the largest amount of canid remains from the caves in this study, with a total of 39 remains lo cated within 9 out of the 81 surface collections (Figure 6-4). Due to the larger sample size, spatial autocorrelation identified the pattern as being somewhat clustered, but the pattern may be due to ra ndom chance. Cokriging was unable to process because only nine and not the necessary ten su rface collections included canid remains. The majority of remains were located on the righ t (n=11, 64.7%) over the left (n=6, 35.3%) sides of the cave, the northern (n=38, 97.4%) over the sout hern (n=1, 2.6%) regions, and the western (n=23, 59%) over the eastern (n=16, 41%) regions of the cave. One of the remains was identified as Urocyon cinereoargentes and the rest were identified as Canis lupus familiaris (n=38). Most of the canid specimens were identified as teeth fragments (n=37, 94.8%). The rest of the elements include one (2.6%) cranium, identified as Urocyon cinereoargentes and one (2.6%) domestic dog bacculum. The only specimens side d were the teeth and these included 17 (53.1%) teeth sided to the left and 15 (46.9%) sided to the right. Artiodactyla A total of 67 specimens were id entified as artiodactyls and thes e were located within 26 of the 81 surface collections with fa unal material (Figure 6-4). Spatial autocorrelation identified the pattern of remains as random and being neither clustered nor dispersed. Cokriging analysis found a higher concentration around surf ace collections CS7-3-1 and CS111-1, for all of the spatial distinctions including left versus right, north versus south, and eas t versus west. Visual analysis of this material identifies a slightly higher concen tration of remains on the right (n=15, 57.7%) over the left (n=11, 42.3%), the south (n=40, 59. 7%) over the north (n=2 7, 40.3%), and the east (n=41, 61.2%) over the west (n=26, 38.8%). Specif ic elements and the side of elements distributions are discussed for each family separately. 123

PAGE 124

Tayassuidae A total of five specimens were identified to the peccary family and were located in surface collections of CS1-9-1, CS1-27-1, CS9-2-1, CS11-2-1, and CS11-6-1 (5 out of 81 surface collections) (Figure 6-4). Both spatial autocorrel ation and cokriging failed for the analysis of peccary remains. These remains were visually as sessed and it was found that all of the remains were located on the left (n=2, 100%) side of CS1. There was a higher amount of peccary on the southern (n=3, 60%) versus northern (n=2, 40 %) areas and on the western (n=3, 60%) versus eastern (n=2, 40%) regions. The elements identi fied as peccary included teeth (n=3, 60%), a mandible (n=1, 20%), and a metacarpal (n=1, 20%). Three of these elements were sided and two (66.7%) were sided to the right and one (33.3%) was sided to the left. Cervidae The majority of artiodactyl remains are from the cervid family (n=56) and these remains were located within 19 of the 81 surface collectio ns with faunal materials (Figure 6-4). Deer remains were identified as being somewhat clus tered, the pattern may be due to random chance. Cokriging analysis identified the same probab ility distribution of remains around CS7-3-1 and CS11-1-1 for the relationship between left versus right, north versus south, and east versus west (Figure 6-10). Upon a visual analysis of the site the majority of remains in CS1 were located on the right (n=13, 65%) versus the left (n=7, 35%), in the sout h (n=35, 62.5%) versus the north (n=21, 37.5%), and in the east (n=36, 64.3%) versus the west (n=20, 35.7%). Out of the 56 remains identified to cervids, 48 of these rema ins were sided and the distribution was even between the left (n=24, 50%) a nd the right (n=24, 50%) sides. Between the two major subgroups of the cervids the white-tailed deer accounted for 19 remains and was distributed within 16 surface collections. The brocket deer accounted for 35 specimens but was only found in four of the surface collections. Spatial autocorrelation identified 124

PAGE 125

both the white-tailed deer and brocket deer patt erns as being random a nd neither clustered nor dispersed. Due to the limited distribution of rema ins only the white-tailed deer was able to be processed using cokriging and the results from th is analysis identify a large concentration of remains within CS1 for the relati onship between north versus south and east versus west. It also identified a pattern for left vers us right at the mouth of the ca ve within the CS1 area. Brocket deer material were not well-represented in CS1 (n=1) so the analysis of their distribution is inconclusive. However, there was a higher concentration of remains within the south (n=33, 94.3%) than the north (n=2, 5.7%) and in the ea st (n=34, 97.1%) than the west (n=1, 2.9%). White-tailed deer the remains were somewhat evenly distributed in CS1 in relation to the left (n=10, 58.8%) and right (n=7, 41.1%) sides of the cave. However, there was a higher concentration in the north (n =17, 89.5%) over the south (n=2, 10.5%) and in the west (n=17, 89.5%) over the east (n=2, 10.5%).Visually the brocket deer material was identified at three of the cave entrances within sections CS1, CS6, CS7, and CS11. That is an interesting pattern considering that the white-tailed deer remains ar e well distributed further into the cave at CS1 and also within CS11. Rodentia Accounting for the largest number of specimens (n=190) of faunal remains from Cueva de Sangre, rodent remains were only located within ten of the 81 surface collections (Figure 6-4). Spatial autocorrelation identified the distribution of remains as random, being neither clustered nor dispersed. Cokriging analysis identified the highest concentrations centered on CS6-8-1 and CS6-3-1 to varying degrees for left versus right, the north versus the south, and the east versus the west (Figure 6-11). The highest concentrat ion of remains were identified for CS6-8-1 (n=112, 58.9%). In CS1 a total of 22 remains we re identified, most on the left (n=18, 81.8%) rather than the right (n=4, 18.2%) sides. Ther e were more remains in the north (n=187, 98.4%) 125

PAGE 126

than the south (n=3, 1.6%) and the east (n =168, 88.4%) than the west (n=22, 11.6%). The body portions identified for the rodent remains in cludes cranial (n=54, 28.4%), axial (n=55, 28.9%), front limbs (n=28, 14.7%), hind limbs (n=40, 21.1 %), and distal (n=13, 6.8%) parts of the body. A total of 124 remains were sided and most were from the left (n=79, 63.7%) side rather than the right (n=45, 36.3%) side. Agoutidae and Dasyproctidae A total NISP of 87 was identified for the mediumand large-sized rodents including the families Agoutidae and Dasyproctidae within ninet een of the surface collections (Figure 6-4). Spatial autocorrelation identified the pattern as being neither clustered nor dispersed, but it was considered a random distribution. Cokriging an alysis was performed for the Agoutidae and Dasyproctidae specimens and alt hough there were no areas of hi gh concentrations identified by specific surface collections, they were cente red within the lower region of CS1, around collection CS1-84-1. Visual analysis identified a slightly higher concentration of Agoutidae and Dasyproctidae material in the le ft (n=22, 29.7%) over the right (n =52, 70.3%) sides of the cave. There was a disproportionate number of remain s in the northern (n=76, 87.4%) versus the southern (n=11, 12.6%) half of th e cave and within the western (n=74, 85.1%) versus the eastern (n=13, 14.9%) half. The body portions identified for these two families include cranial (n=61, 70.1%), front limb (n=15, 17.2%), hind limb (n=6, 6.9%), distal (n=4, 4.6%), and innominate (n=1, 1.1%) parts of the body. There were an even proportion of sided elem ents including the left side of 39 (48.8%) remains and the right side with 41 (51.3%) remains. Naj Tunich Excavations at the site of Naj Tunich were completed in the early 1980s and it was here that some of the first hypotheses were develope d for the ritual use of cave sites in the Maya region. The faunal remains from the site of Naj Tunich were identified by Susan Colby at UCLA 126

PAGE 127

in 1984 (Brady 1989). This published resource is us ed to show how more information can be obtained from the analysis of olde r research projects, particularly when well-scaled maps are also available. As in the other sites, there was a vari ed sample size for taxonomic groups at the site so spatial autocorrelation analysis varied within the site. However, due to the large number of surface collections and some large sample si zes, Naj Tunich provided enough information for cokriging analysis to be performe d at the site. All identified patterns through visual analysis were completed with the recreation of the map with th e identified surface collection faunal assemblage from Operation IV. Some patterns were iden tified by Brady in 1989, and I plan on adding to these initial observations by identifying the spatia l patterns of these faunal remains using left versus right sidedness, directiona lity, and light versus dark. A total of 44 surface collections from Operati on IV at Naj Tunich contained faunal remains (Figure 6-5). These surface collections are refe rred to by Lot by the researchers. The spatial distinctions identified for thes e Lots with Operation IV incl ude the left (n=30, 68.2%) versus right (n=14, 31.8%) sides, the north (n=33, 75%) versus south (n=11, 25%), the east (n=32, 72.7%) versus west (n=12, 27.2%), and the li ght (n=13, 29.5%) versus dark (n=31, 70.5%) regions of the cave (Figure 6-5) (Table 6-6). A total of twenty taxonomic groups were identified within Operation IV at Naj Tunich. These taxonomic groups include even-toed ungulates (16.91%), deer (14.32%), white-ta iled deer (11.83%), birds (2.70 %), peccary (2.59%), pacas and agoutis (2.59%), rodents ( 2.07%), brocket deer (1.66%), Meleagris gallopavo (domestic turkey) (1.45%), Tapirus bairdii (Bairds tapir) (1.14%), opossums (1 .04%), turtles (0.83%), crustaceans (0.62%), bats (0.62%), armadillos (0.62%), prim ates (0.62%), canids (0.41%), cats (0.31%), raccoons (0.31%),and cottontail rabbit (0.21%) (Table 6-1). Each of these taxonomic groups is 127

PAGE 128

discussed in detail below and includes elemen ts or body portions and side of elements when applicable. Crustacean There was a small sample (n=6) of crustacean remains identified in Operation IV at Naj Tunich and they were located in Lots 12, 45, 50, and 51 (Figure 6-5). Spatial autocorrelation identified the pattern as somewhat clustered, but this pattern was probably due to random processes. Due to the limited distribution into only four Lots, cokrig ing was unable to be performed. There was an even distribution of thes e remains, three in each region, in relation to the left and right sides and the north and south di rections. However, all of the crustacean remains were located within the dark part of the cave a nd on the eastern section. All of the remains were identified as claws and n one of them were sided. Testudines A total of eight turtle remains were identifie d within Operation IV. The turtle materials were identified in Lots 2 (n=1, 12.5%), 50 (n=3, 37.5%), and 51 (n=4, 50%) (Figure 6-5). Due to the small sample size and limited distribution, both spatial autoco rrelation and cokriging failed to process the spatial patterns for the turtle specim ens. There are four (50%) remains found in both the left and right sides and the north and south directions, while all turtle remains (n=8, 100%) were found within the dark regions and eastern portion of the cave. A ma jority of the remains were identified as carapace (n =6, 75%) with some long bones (n =2, 25%) also being identified. None of the bones identified as turtle were sided. Aves Twenty-six avian remains were identified w ithin Operation IV in Lots 2, 3, 10, 13, 14, 15, 16, 22, 34, 36, 42, 46, 50, and 51 (Figure 6-5). These bird remains included both possible cave dwellers, including Buteo platypterus (broad-winged hawk), and t hose not known to live within 128

PAGE 129

cave habitats, including Ardeidae (herons and bitterns family), Ramphastidae (toucans family), Psittacidae (parrots and cockatoos family), Columbidae (doves and pigeons), and Meleagris gallopavo (Brady 1989). Spatial autocorrelation identi fied the pattern as being random and neither clustered nor dispersed w ithin Operation IV. Cokriging anal ysis identified the pattern of distribution as being concentr ated around IV 3, on the northern over the southern, the western over the eastern, the left over the right sides, and the dark regions over the light regions (Figure 6-12). A majority of the Aves remains are located in the left (n=19, 73%) versus the right (n=7, 27%), to the north (n=22, 85%) versus the sout h (n=4, 15%), to the eas t (n=19, 73%) versus the west (n=7, 27%), and within the dark (n=21, 81%) versus the light (n=5, 19%) areas of the cave. A large number of the bird remains (n=14, 53.8%) were identified as Meleagris gallopavo (domestic turkey) in Lots 3, 10, 13, 15, 16, 34, 42, and 46 (Figure 6-5). These remains were found to occur in larger numbers in the left (n=11, 78.6%) versus the ri ght (n=3, 21.4%), in the north (n=12, 85.7%) versus the south (n=2, 14.3% ), and in the dark (n=12, 85.7%) versus the light (n=2, 14.3%) regions. There was an even number of remains found in the eastern and western parts of the cave. The el ements identified from the bird remains in Operation IV include coracoids (n=4, 15.4%), humeru s (n=4, 15.4%), femur (n=4, 15.4 %), tibiotarsus (n=4, 15.4%), radius (n=2, 7.7%), keel (n=1, 3.8%), cran ium (n=1, 3.8%), phalanx (n=1, 3.8%), and unidentifiable fragments (n=5, 13.9%). There were fifteen bird remains that were sided with a majority of remains being from the right (n=10, 66.7%) rather than the le ft (n=5, 33.3%) side of the body. Didelphidae A total of ten opossum remains were identified in Operation IV and were located within Lots 3, 34, 48, 49, 50, and 51 (Figure 6-5). Spatia l autocorrelati on identified the pattern as 129

PAGE 130

somewhat clustered, but that the pattern was st ill random in distribution. Cokriging analysis failed to process. These ten remains were all loca ted in the dark regions of the cave. A majority of the remains were located in the left/north (n=7, 70%) over the right/south (n=3, 30%), and in the eastern (n=9, 90%) over the western (n=1, 10%) part of the cave. The elements included mandibles (n=4, 40%), scapulas (n=2, 20%), hu meri (n=2, 20%), maxilla (n=1, 10%), and a molar (n=1, 10%). Nine of these opossum remains were sided and there were four (44.4%) leftsided and five (55.6%) right-sided elements. Dasypus novemcintus A total of six armadillo remains were identified within Operation IV at Naj Tunich and were located in Lots 1, 2, and 50 (Figure 65). Due to the small sample size and limited distribution both spatial autocorrelation and cokriging failed analysis for armadillo remains. The remains were more concentrated in the right/sou th (n=4, 66.7%) than the left/north (n=2, 33.3%). All of the bones were collected in the eastern a nd dark regions of the cav e. Unlike the other sites analyzed, all of the materials identified were appendages and a cranial fragment and no scute remains were identified at the site. The elements identified as armadillo included femurs (n=2, 33.3%), a cranium (n=1, 16.7%), a humerus (n=1, 16.7%), an ulna (n=1, 16.7%), and a tibia (n=1, 16.7%). Five of these elements were side d including three (60%) le ft-sided and two (40%) right-sided elements. Chiroptera A total of six remains were id entified as bats and these rema ins were located in Lots 24, 25, 42, and 45 (Figure 6-5). Spatial autocorrelation identified the pattern as being random and neither clustered nor dispersed at the site. Cokrig ing was unable to process because of the limited spatial distribution of these remains. The bat spec imens were only located in the eastern part of the cave. There was a higher distribution of rema ins on the right (n=4, 66 .7%) over the left (n=2, 130

PAGE 131

33.3%) side of the cave. There was also an even distribution between the northern (n=3, 50%) and southern (n=3, 50%) and also the light (n=3 50%) and dark (n=3, 50%) parts of the cave. The elements included crania (n=2, 33.3%), radii (n=2, 33.3%), an ulna (n=1, 16.7%), and a metapodial (n=1, 16.7%). There was an even dist ribution of right (2, 50%) and left (2, 50%) sided elements from bats within Operation IV. Primates Naj Tunich is the only site under analysis that contained primate ma terial. Two specimens identified as Alouatta sp. or howler monkey, were identified from a single surface collection (Lot 1), on the left, north, east, and da rk parts of the cave (Figure 65). Spatial autocorrelation and cokriging both failed analysis for the primate remains. The materials were identified as a humerus and a tibia. Both of these remains were also left-sided. Canidae The Canidae family consisted of only four id entified specimens in Operation IV, and those were within Lots 3, 28, and 51 (Figure 6-5). Th e small number of canid remains and limited distribution meant neither spatial autocorrela tion nor cokriging were processed. The canid remains were distributed in only the left and nor thern part of the cave, while there were higher concentrations of remains in the western/dark (n=3, 75%) regi ons of the cave. The elements included a premolar (n=1, 25%), a radius (n= 1, 25%), an ulna (n=1, 25%), and a metapodial (n=1, 25%). Two out of the four canid remains were sided, and both of these bones were leftsided. Felidae There were three cat specimens identified in Operation IV within Lots 1, 6, and 34 (Figure 6-5). Due to the small sample size and limited di stribution spatial autoco rrelation and cokriging both failed. All of the remains were located in th e left, north, and dark parts of the cave, while 131

PAGE 132

there was a higher concentrati on in the east (n=2, 66.7%) than the west. The three elements identified include a radius (n=1, 33.3%), a humerus (n=1, 33.3%), and a metapodial (n=1, 33.3%). These three elements were sided and included two (66.7%) left-sided and one (33.3%) right-sided bone Procyonidae A total of three raccoon remains were identified at the site and with in Operation IV they were located in Lots 3, 4, and 37. Due to the smal l sample size, spatial correlation and cokriging both failed to process. All of the remains were with in the left side, and th e north and dark regions of the cave, and the majority of remains were located within the west ern (n=3, 66.7%) area. The elements included a radius (n=1, 33.3%), a sc apula (n=1, 33.3%), and a femur (n=1, 33.3%). These remains were sided, and the left-sided (n=2, 66.7%) outnumbered the right-sided (n=1, 33.3%) elements. Tapirus bairdii The Tapirus bairdii (Bairds tapir) was not identified at any of the other cave sites examined in this study. A total of eleven tapir remains were recorded at Naj Tunich, Operation IV in Lots 1, 3, 16, 24, 42, and 57 (Figure 6-5). Spatial autocorrelation iden tified the pattern as random and being neither clustere d nor dispersed. Due to the limited distribution of remains, cokriging was not able to process. A majority of these remains were located in the left (n=7, 63.6%) over the right (n=4, 36.4%), the north (n =8, 72.7%) over the south (n=3, 27.3%), east (n=10, 90.9%) over the west (n=1, 9.1%), and the dark (n=8, 72.7%) over the light (n=3, 27.3%) regions of the cave. Many of the remains were distal appendages from the body including the carpal or tarsal bones and the metapodials. The el ements identified include an atlas vertebra (n=1, 9.1%), a coccyx (n=1, 9.1%), a sca pula (n=1, 9.1%), a humerus (n=1, 9.1%), 132

PAGE 133

carpals/tarsals (n=2, 18.2%), long bones (n=2, 18.2%), and metapodials (n=3, 27.3%). Only two out of the eleven specimens were sided a nd both were identified as being left-sided. Artiodactyla The largest identified taxonomic group within Naj Tunich was the artiodactyls that had an total of 163 remains within Oper ation IV and were located in 26 of the 44 Lots (Figure 6-5). Spatial autocorrelation identified the pattern as clustered with only 5-10% likelihood that this clustered pattern is the result of random chance. The patterns identified by cokriging had the highest concentration of remain s centered on Lots 44 and 18 for the relationships between the left and right sides, the north and south, the east and west, and the light and dark regions. The majority of artiodactyls remains were located in the left (n=115, 70.6%) rather than on the right (n=48, 29.4%), on the northern (n=142, 87.1%) rather than on the southern (n=21, 12.9%) side, on the western (n=137, 84.1%) rather than on the eastern (n=26, 15.9%) region, and the dark (n=137, 84.1%) rather than the light (n=26, 15.9%). There was an even distribution of sided elements including rights (n=58) a nd left (n=60) sided remains. Specific elements are discussed below for the peccary and deer families. Tayassuidae A total of 25 peccary remains were located in Operation IV within 12 of the 44 Lots (Figure 6-5). Spatial autocorrela tion identified the pattern as be ing random neither clustered nor dispersed within Operation IV. Cokriging iden tified a high concentration of peccary materials within several clusters around Lots 1, 16, and 24, and also around Lot 57. Visual analysis identified the majority of peccary remains on th e left/north (n=17, 68%) over the right/south (n=8, 32%) regions, in the dark (n=16, 64%) over the light (n=9, 36%), and the eastern (n=20, 80%) over the western (n=5, 20%) parts of Opera tion IV. The body portions for the peccary from Naj Tunich include the cranial (n=5, 20%), the axial (n=6, 24%), the fr ont limb (n=9, 36%), the 133

PAGE 134

hind limb (n=4, 16%), and the metapodials (n=1 4%) from the body. There were also eight (72.2%) right-sided elements and only three (27.3%) left -sided elements identified at the site. Cervidae Within the artiodactyls, 138 remains were iden tified to the deer family in Operation IV within 23 of the 44 Lots (Figure 6-5). Of these, most were identified as white-tailed deer (n=114, 82.6%) while a smaller number of remains were within the brocket deer genus (n=16, 11.6%). Spatial autocorrelation iden tified the pattern of deer as clustered with on ly 5-10 % likelihood that this pattern was the resu lt of random chance. Within this family, white-tailed deer were identified as being somewhat clustered but the pattern wa s probably due to random chance. Brocket deer distribution was also described as being random, and neither clus tered nor dispersed. Due to the high number of remains identified wh ite-tailed deer cokriging analys is are only described for this specific taxonomic group. Cokrigi ng analysis for all four spatial distinctions identified the highest concentrations for the patterning of remains around Lo ts 16, 18, and 42 (Figure 6-13). Visual analysis was completed for only the whit e-tailed deer specimens at the site. A higher concentration of remains was found on the left (n=82, 71.9%) over right (n=32, 28.1%) sides, the north (n=105, 92.1%) over the south (n=9, 7.9 %), the east (n=97, 85.1%) over the west (n=17, 14.9%), and the dark (n=100, 87.7%) over the lig ht (n=14, 12.3%) regions. The higher amounts of faunal materials in specific parts of the cave may signify the ritual si gnificance for the deer remains within the site of Naj Tunich. The body portions from the white-tailed deer include the cranial (n=14, 12.3%), the axial (n=51, 44.7%), the front limbs (n=19, 16.7%), the hind limbs (n=27, 23.7%), and the distal (n=3, 2.6%) parts of the body. A total of 62 white-tailed deer remains were sided and slightly more remains were from the left (n=35, 56.5%) than the right (n=27, 43.5%) side of the body. 134

PAGE 135

Rodentia Only a total of twenty rodent remains were identified from the faunal assemblage from Operation IV in Lots 1, 15, 27, 48, 51, and 53 (Figur e 6-5). Spatial autoco rrelation identified the pattern as random and found that it was neither clustered nor disperse d at the site. Due to the low number of lots with rodent remains, cokriging fa iled. Visual analysis did identify some patterns of distribution of rodent remains in Naj Tuni ch including the higher amount of remains in the left/north (n=17, 85%) over the right/south (n=3, 15%) sides, th e west (n=13, 65 %) over the east (n=7, 35%) sides, and the light (n=14, 70%) over the dark (n= 6, 30%) regions. The elements identified from the rodent remains includes mandibles (n=3, 15%), cr ania (n=3, 15%), an innominate (n=1, 5%), humeri (n=4, 20%), femora (n=4, 20%), and tibiae (n=5, 25%). Out of the twenty remains a total of seventeen remains we re sided and this incl uded six (35.3%) left and eleven (64.7%) right sides. Agoutidae/Dasyproctidae The larger rodent remains include the families Agoutidae and Dasyproctidae which had an NISP of 25 within Operation IV and within 14 of the 34 Lots at the site (Figure 6-5). These remains were identified by spatial autocorrelatio n as having a random pattern of distribution that was neither clustered nor dispersed. Cokriging for the families Agoutidae and Dasyproctidae indicated there were some differences in the area of concentrations for the dichotomies under analysis. Both the left versus right and the north versus south, and the east versus west had a higher concentration around Lot 57. However, the shape of these concentrations differed for these areas under analysis. The concentrations for light versus dark regions were located in two areas including around Lot 57 and Lots 1, 16, and 42. The distributi on of these two families were higher on the left (n=14, 56%) side than the ri ght (n=11, 44%) side of the cave, on the northern (n=16, 64%) than the southern (n=9, 36%), in th e eastern/dark (n=22, 88%) than the western/lit 135

PAGE 136

(n=3, 12%) part of the cave. The body portions identified include the cranial (n=11, 44%), the axial (n=4, 16%), the front limb (n=4, 16%), a nd the hind limbs (n=6, 24 %) of the body. A total of fourteen remains were sided and the distri bution included eight (57.1 %) right-sided and six (42.9%) left-sided elements. Sylvilagus sp. Only two specimens were identified to the genus Sylvilagus sp. in Operation IV in Lots 29 and 49 (Figure 6-5). These remains failed spatial autocorrelation and cokriging because of the small sample size and limited distribution. The co ttontail rabbit remains were evenly distributed with one bone located in the le ft and the right, the east and the west, the north and the south, the light, and the dark regions of the cave. The elemen ts identified as cottonta il rabbit include a tibia and a humerus. Both of these elements are right-sided. Summary The distribution of the faunal remains by taxono mic groups varied in concentrations and placement within each cave site. GIS and mapping of the site assisted in both the spatial analyses performed by the computer system and also with the visual assessment. The roles of each group within all of the caves are discussed in the ne xt chapter (Chapter 7, In terpretations). It is important to realize that these differences and si milarities are all based on the ritual placement, both intentional and unintentional, of faunal remains within these caves. 136

PAGE 137

Figure 6-1. Separation of Space at Ca ves Branch Rockshelter, Belize 137

PAGE 138

Figure 6-2. Separation of Space at Stela Cave, Belize 138

PAGE 139

Figure 6-3. Separation of Space at Cueva de El Duende, Guatemala 139

PAGE 140

140 Figure 6-4. Separation of Space at Cueva de Sangre, Guatemala

PAGE 141

Figure 6-5. Separation of Space at Naj Tunich, Guatemala 141

PAGE 142

Figure 6-6. Caves Branch Rockshelter, Crustace an Remains Ordinary Cokriging Light versus Dark Regions 142

PAGE 143

Figure 6-7. Caves Branch Rockshelter, Testudin es Remains Ordinary Cokriging Light versus Dark Regions 143

PAGE 144

Figure 6-8. Caves Branch Rockshelter, Serpen tes Remains Ordinary Cokriging Light versus Dark Regions 144

PAGE 145

Figure 6-9. Cueva de Sangre, Testudines Remains Ordinary Cokriging Left versus Right Sides 145

PAGE 146

Figure 6-10. Cueva de Sangre, Cervidae Rema ins Ordinary Cokrigi ng North versus South Regions 146

PAGE 147

Figure 6-11. Cueva de Sangre, Rodentia Rema ins Ordinary Cokriging Light versus Dark Regions 147

PAGE 148

Figure 6-12. Naj Tunich, Aves Remains Ordinary Cokriging Light versus Dark Regions 148

PAGE 149

149 Figure 6-13. Naj Tunich, Cervidae Remains Ordi nary Cokriging Light versus Dark Regions

PAGE 150

Table 6-1. Relative Frequency Va lues for Five Cave Sites. CBR STC CD CS NT Gentaxa NISP RF NISP RF NISP RF NISP RF NISP RF Crustaceans 65 5.09 11 0.64 1 0.04 8 0.42 6 0.62 Actinopterygii (ray-finned fishes) 13 1.02 4 0.23 399 15.99 2 0.11 N/A N/A Amphibia (amphibian) 5 0.39 1 0.06 11 0.44 N/A N/A N/A N/A Testudines (turtle) 42 3.29 43 2.50 15 0.60 70 3.71 8 0.83 Sauria (lizard) 6 0.47 1 0.06 5 0.20 N/A N/A N/A N/A Serpentes (snake) 37 2.90 18 1.05 50 2.00 3 0.16 N/A N/A Aves (bird) 21 1.65 39 2.27 37 1.48 4 0.21 26 2.70 Meleagris gallopavo (domestic turkey) N/A N/A N /A N/A N/A N/ A N/A N/A 14 1.45 Didelphidae (opossums) 8 0.63 13 0.76 29 1.16 62 3.29 10 1.04 Dasypus novemcinctus (nine-banded armadillo) 102 7.99 80 4.66 2 0.08 3 0.16 6 0.62 Chiroptera (bats) 1 0.08 13 0.76 632 25.32 12 0.64 6 0.62 Primates (primates) N/A N/A N /A N/A N/A N/ A N/A N/A 2 0.21 Canidae (coyotes, dogs, foxes, jackals, and wolves) 3 0.24 4 0.23 22 0.88 39 2.07 4 0.41 Felidae (cats) N/A N/A N/A N/A 8 0.32 N/A N/A 3 0.31 Procyonidae (raccoons) 4 0.31 3 0.17 N/A N/A N/A N/A 3 0.31 Tapirus bairdii (Bairds tapir) N/A N/A N/A N/A N/ A N/A N/A N/A 11 1.14 Artiodactyla (even-toed ungul ates) 18 1.41 30 1.75 30 1.20 67 3.55 163 16.91 Tayassuidae (peccary) 3 0.24 7 0.41 N/A N/A 5 0.26 25 2.59 Cervidae (deer) 11 0.86 16 0.93 18 0.72 56 2.97 138 14.32 Odocoileus virginianus (white-tailed deer) 9 0.71 4 0.23 15 0.60 19 1.01 114 11.83 Mazama sp. (brocket deer) 1 0.08 4 0.23 1 0.04 35 1.85 16 1.66 Rodentia (rodent) 30 2.35 222 12.92 21 0.84 190 10.07 20 2.07 Scuiridae (squirrel) N/A N/A 4 0.23 N/A N/A N/ A N/A N/A N/A Agoutidae (pacas) / Dasyproctidae (agoutis + acuchis) 24 1.88 8 0.47 8 0.32 87 4.61 25 2.59 Sylvilagus sp. (cottontail rabbit) 1 0.08 4 0.23 N/A N/A N/A N/A 2 0.21 Total Identified Remains 380 498 1270 547 299 Total NISP 1276 1718 2496 1887 964 150

PAGE 151

Table 6-2. Separation of Space for Ca ves Branch Rockshelter, Belize Op Unit L1 R2 N1 S2 E1 W2 Light1 Dark2 Open1 Restr2 1A 10F 1 1 1 1 N/A 1A 10G 1 1 1 1 N/A 1A 11F 1 1 1 1 N/A 1A 11G 1 1 1 1 N/A 1A 12F 1 1 1 1 N/A 1A 12G 1 1 1 1 N/A 1A 13F 1 1 1 1 N/A 1A 13G 1 1 1 1 N/A 1B 23H 2 2 1 1 N/A 1B 23I 2 2 1 1 N/A 1B 23J 2 2 1 1 N/A 1B 23K 2 2 1 1 N/A 1B 22K 2 2 1 1 N/A 1B 24H 2 2 1 1 N/A 1B 24I 2 2 1 1 N/A 1B 24J 2 2 1 1 N/A 1B 24K 2 2 1 1 N/A 1D 21N 2 2 1 2 N/A 1D 21O 2 2 1 2 N/A 1D 21P 2 2 1 2 N/A 1C 29G 2 2 1 1 N/A 1C 29H 2 2 1 1 N/A 1C 30G 2 2 1 1 N/A 1C 30H 2 2 1 1 N/A Table 6-3. Separation of Space for Stela Cave, Belize Chamber Unit L1 R2 N1 S2 E1 W2 Light1 Dark2 Open1 Restr2 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 3 1 1 1 1 1 1 4 1 1 1 1 1 1 5 1 1 1 1 1 1 7 1 1 1 1 1 3 8 2 2 2 2 2 3 9 2 2 2 2 2 3 10 2 2 2 2 2 3 11 2 2 2 2 2 151

PAGE 152

Table 6-4. Separation of Space for Cueva de El Duende, Guatemala Provenience L1 R2 N1 S2 E1 W2 Light1 Dark2 Open1 Restr2 CD2-0-3 N/A 1 1 1 1 CD2-1 N/A 2 1 1 1 CD2-3 N/A 2 1 1 1 CD2-2 N/A 2 1 1 1 CD1-1 N/A 1 2 2 2 CD1-2 N/A 1 2 2 2 CD1-3 N/A 1 2 2 2 CD1-4 N/A 1 2 2 2 CD3-1-1 N/A 1 2 2 2 152

PAGE 153

Table 6-5. Separation of Space for Cueva de Sangre, Guatemala Provenience L1 R2 N1 S2 E1 W2 Light1 Dark2 Open1 Restr2 Provenience L1 R2 N1 S2 E1 W2 Light1 Dark2 Open1 Restr2 CS1-1-1 1 1 2 N/A N/A CS1-58-1 2 1 2 N/A N/A CS1-1-2 1 1 2 N/A N/A CS1-62-1 2 1 2 N/A N/A CS1-1-3 1 1 2 N/A N/A CS1-66-2 2 1 2 N/A N/A CS1-2-1 2 1 2 N/A N/A CS1-66-3 2 1 2 N/A N/A CS1-3-1 1 1 2 N/A N/A CS1-68-1 2 1 2 N/A N/A CS1-4-1 2 1 2 N/A N/A CS1-70-2 1 1 2 N/A N/A CS1-5-1 1 1 2 N/A N/A CS1-71-1 1 1 2 N/A N/A CS1-6-1 2 1 2 N/A N/A CS1-74-1 1 1 2 N/A N/A CS1-7-1 2 1 2 N/A N/A CS1-76-1 2 1 2 N/A N/A CS1-8-1 2 1 2 N/A N/A CS1-77-2 1 1 2 N/A N/A CS1-9-1 2 1 2 N/A N/A CS1-78-1 2 1 2 N/A N/A CS1-10-1 1 1 2 N/A N/A CS1-84-1 1 1 2 N/A N/A CS1-12-1 2 1 2 N/A N/A CS1-85-1 1 1 2 N/A N/A CS1-13-1 1 1 2 N/A N/A CS1-89-1 2 1 2 N/A N/A CS1-15-1 2 1 2 N/A N/A CS1-90-1 2 1 2 N/A N/A CS1-16-1 1 1 2 N/A N/A CS1-99-1 2 1 2 N/A N/A CS1-19-1 1 1 2 N/A N/A CS2-10-1 1 2 N/A N/A CS1-20-1 2 1 2 N/A N/A CS3-04-1 1 2 N/A N/A CS1-21-1 1 1 2 N/A N/A CS3-05-1 1 2 N/A N/A CS1-26-1 1 1 2 N/A N/A CS3-07-1 1 2 N/A N/A CS1-27-1 2 1 2 N/A N/A CS5-08-1 2 1 N/A N/A CS1-28-1 1 1 2 N/A N/A CS6-03-1 1 1 N/A N/A CS1-32-1 2 1 2 N/A N/A CS6-06-1 1 1 N/A N/A CS1-33-1 1 1 2 N/A N/A CS6-08-1 1 1 N/A N/A CS1-34-1 1 1 2 N/A N/A CS7-02-1 2 1 N/A N/A CS1-35-1 1 1 2 N/A N/A CS7-03-1 2 1 N/A N/A CS1-37-1 2 1 2 N/A N/A CS7-04-1 2 1 N/A N/A CS1-39-1 2 1 2 N/A N/A CS7-05-1 2 1 N/A N/A CS1-40-1 1 1 2 N/A N/A CS9-02-1 2 1 N/A N/A CS1-43-1 1 1 2 N/A N/A CS9-03-1 2 1 N/A N/A CS1-44-1 2 1 2 N/A N/A CS9-04-1 2 1 N/A N/A CS1-45-1 2 1 2 N/A N/A CS9-06-1 2 1 N/A N/A CS1-46-1 2 1 2 N/A N/A CS9-07-1 2 1 N/A N/A CS1-50-1 2 1 2 N/A N/A CS9-08-1 2 1 N/A N/A CS1-51-1 2 1 2 N/A N/A CS11-01-1 2 1 N/A N/A CS1-52-1 1 1 2 N/A N/A CS11-02-1 2 1 N/A N/A CS1-53-1 2 1 2 N/A N/A CS11-03-1 2 1 N/A N/A CS1-54-2 1 1 2 N/A N/A CS11-04-1 2 1 N/A N/A CS1-55-1 1 1 2 N/A N/A CS11-05-1 2 1 N/A N/A CS1-56-1 2 1 2 N/A N/A CS11-06-1 2 1 N/A N/A CS1-57-1 2 1 2 N/A N/A 153

PAGE 154

154 Table 6-6. Separation of Space for Naj Tunich, Guatemala Op Lot L1 R2 N1 S2 E1 W2 Light1 Dark2 Open1 Restr2 IV 1 1 1 1 2 N/A IV 2 1 1 1 2 N/A IV 3 1 1 2 2 N/A IV 4 1 1 2 2 N/A IV 5 1 1 2 2 N/A IV 6 1 1 2 2 N/A IV 7 1 1 2 2 N/A IV 10 1 1 2 2 N/A IV 11 1 1 1 2 N/A IV 12 1 1 1 2 N/A IV 13 1 1 1 2 N/A IV 14 1 1 1 2 N/A IV 15 2 2 1 1 N/A IV 16 1 1 1 2 N/A IV 17 2 2 1 1 N/A IV 18 1 1 1 2 N/A IV 21 1 1 1 2 N/A IV 22 2 1 1 1 N/A IV 24 2 2 1 1 N/A IV 25 2 2 1 1 N/A IV 26 2 2 1 1 N/A IV 27 1 1 2 1 N/A IV 28 1 1 2 1 N/A IV 29 1 1 2 1 N/A IV 30 2 2 1 1 N/A IV 31 1 1 2 1 N/A IV 34 1 1 1 2 N/A IV 36 1 1 1 2 N/A IV 37 1 1 1 2 N/A IV 40 1 1 1 2 N/A IV 41 1 1 1 2 N/A IV 42 2 1 1 2 N/A IV 43 2 1 1 2 N/A IV 44 1 1 1 2 N/A IV 45 1 1 1 2 N/A IV 46 2 2 1 1 N/A IV 47 2 2 1 1 N/A IV 48 2 2 1 2 N/A IV 49 2 2 1 2 N/A IV 50 2 2 1 2 N/A IV 51 1 1 1 2 N/A IV 53 1 1 2 2 N/A IV 54 1 1 2 2 N/A IV 57 1 1 1 2 N/A

PAGE 155

CHAPTER 7 INTERPRETATIONS Introduction This research project attempted to reconstruct the ancient Maya ritu al use of caves through the analysis of the spatial dist ribution of faunal remains using bot h GIS and visual analysis. My study relies on some of the major tenets of cognitive archaeology, or the study of reconstructing and understanding the ancient mind (Fla nnery and Marcus 1996, 1998), including reconstructing the intentional placement of artifacts within the landscape using such technological advances as GIS. The use of lands capes and the distribution of sites and artifacts across these landscapes is an important part of understanding an ancien t peoples use of their surroundings; however, there is mo re than just subsistence unde rlying these sometimes complex patterns. The landscape represents both the functional place for food and shelter, and a sacred landscape to the ancient Maya. The sacred landscape of the ancient Maya is based on both the vertical, or layering of space, and the horizontal, or card inal directionality, planes (Mat hews and Garber 2004).The most relevant plane to be analyzed in the cave and this research project is the horizontal plane, which includes the quadripartite separation of space and directionality in the form of north, south, east, west, and the center (Garcia-Zambrano 1994; Mathews and Garber 2004; Moyes 2001, 2002, 2004) and also the sidedness of space between the right and left (Brown 2004; Coggins 1980; Palka 2002). Here I have attempted to reconstruct the ancient Maya mind in relation to the faunal materials in a ritual setting. The known ritual us e of caves and the ethno graphic, ethnohistoric, and iconographic information available about the symbolic meaning of a variety of animals (often spatially associated) allowed me to test various spatial models of the ancient Maya universe and symbolic meaning of animals within cave rituals. 155

PAGE 156

GIS was used to help facilitate the analysis of the use of space within the cave along with a visual analysis of the site. Cokriging and Spa tial Autocorrelation tools were used to define significant patterning or pattern loci of taxonomic groups in each cave. Spatial autocorrelation identifies the spatial distribution of taxonomic gr oups within the cave as being either dispersed, random, or clustered within the cave. Cokriging was used to identify the spatial patterning by analyzing the relationship between taxonomic groups and the five dichotomies for the separations of space. The patterns formed from c okriging identify the areas within the caves that higher concentration of these taxonomic groups should be identified within. Due to the small sample sizes and limited distribution of remains a nd areas of excavations the GIS spatial analysis tools were not always helpful in identifying patterns Therefore, visual analysis of each site was also conducted on the taxonomic groups. Visual assessment was also used to determine patterning including the relations hip between elements or body por tions, sides of elements, and when applicable, the burning or charring of these elements under analysis. Limitations to Spatial Pattern Analysis Spatial analysis and computer-generated fi nding always have a minimum of samples needed based on the programs and systems used for analysis. It is always important in any scientific study to address these limitations ear ly on and use them to help with the overall assessment of the site and those fi ndings at these sites. There were some limitations in the spatial patterns analysis that need to be addressed fo r both the GIS and visual assessments in this research project. The limitations are addressed generally for all of the sites, and then specific site issues are discussed. The main problems that aros e from both the GIS and visual analyses can be traced back to the reconstruction of the sites based on maps, placement of the excavation units, the small sample size of faunal remains from these sites, and the taphonomic influences, both natural and cultural, on these faunal remains. 156

PAGE 157

Many of these sites and the excavation units wi thin them were reconstructed from maps that were drawn to scale. However, there wa s some difficulty in placing these sites in georeferenced areas with known coordinates, wh ich is a necessity when working with GIS. There were also some difficulties in redrawing a nd rescaling these maps; however, I was able to get the maps in ArcGIS to reflect the scales and orientations provided by researchers from these sites. Caves Branch Rockshelter and Stela Cave both had images of the sites that had been previously constructed and digitized by their respective researchers, Gabriel Wrobel (Wrobel and Tyler 2006; Wrobel 2008) and Cameron Griffith (Ishihara and Griffith 2004). Cueva de Sangre and Cueva de El Duende were both reconstructed from printed materials from the site and then digitized by myself (Brady 1997; Brady et al. 1997; Brady and Rodas 1992; Brady and Scott 1997; Minjares 2003). Operation IV at Naj Tunich was also taken from Bradys (1989) dissertation and digitized with al l of the surface collection sites included in this reformatted map. The only problem encountered duri ng the digitizing or prepping of these maps for GIS analysis was that the placement of excavation units and the use of scale were sometimes faulty because of the copies of the maps used. Therefore, it is sa fe to assume that these reconstructed maps and placement of excavation units are at the sc ale provided by the initial researchers. The two GIS spatial analysis tools, Spatial Autocorrelation and Cokriging, needed specific sample sizes and distributions which hindered the ability to process these tools. Spatial autocorrelation required a sample size of over 30 samples to run effectively, although there were discrepancies in these results even with 30 samples. This di screpancy in effective outcomes included that most of the samples were random ev en when visually they seemed clustered in a specific area within the cave. The limitation with the cokriging program was that a sample of over ten areas with remains was needed. The use of ten units or surface co llections still did not 157

PAGE 158

produce accurate measurements from cokriging and these results were inaccurate when put against visual analysis. The sites that were unable to process cokrig ing because of limited distribution included Cueva de El Duende and Stela Cave, both of which only had eight units with faunal remains. Visual analysis helped in th e identification of some patterns; however, there were also limitations in the size of the sample s that dictated the observable results. Smaller samples were difficult to identify during this re search study and the limite d distribution of some units in specific spatial regions also skew ed those findings. For example, Caves Branch Rockshelter included a very small dark region (n=3) of the site sample and a very high amount from the light region (n=21) which skewed some of the findings since there would always be a high preponderance of light ove r dark parts of the cave. Excavation methods were different within each of the caves. There were those caves whose remains were collected from excavated units, including CBR, CD, and STC, while there were also sites with mainly surface collections and very few excavated areas including, CS and NT. These differences in collection methods can be seen in the differences in how both GIS and visual analysis were able to pr ocess. CS and NT contained the highest amount of samples within this study and these remains were well distribute d for both GIS and visual analysis. There were however issues with the placement of these surfa ce collections and the higher amount of remains being identified in some areas than in other areas i.e. more surface collec tions in the light than the dark regions of the cave. Each cave also had a different taphonomic im pact that affected the visual and GIS analyses. These taphonomic influences include to movement of artifacts by both animals, water, and human movements within the cave. For ex ample, CBR was a known ancient Maya cemetery and the soil at this site was a mixed grave fill which means that the soil had been physically 158

PAGE 159

churned and dug within over time as new burials were being interred at the site. At CBR, which is located next to a river, ther e is a large area of r unoff and displacement of artifacts from the seasonal rains within the area. Cueva de Sangre and Cueva de El Duende both had a high amount of sedimentation from seasonal flooding during the rainy season that caused the movement of artifacts and animal remains. At both STC and CBR there were also eviden ce of looting that may have caused the destruction and displacement of animal remains within the site. The taphonomic movement and inclusion of animal remains may have skewed both the GIS and visual analysis findings. These impacts are discussed in the section below. Taxonomic Trends Trends that were identified for some species and taxonomic groups in terms of the spatial divisions tested in this thes is are highlighted and discusse d in detail below based on the taxonomic groups used for analysis. Additional variables for each taxonomic group such as body portion and side distributions are considered. In each case, additional information on the habitats, habits, and symbolic importance of the animal group are used to consider the possible meaning of the distributions revealed. Within the caves analyzed for and include d within this study the collection methods, excavations placements and taphonomy are also all important in understanding the overall patterns a nd these are also discussed in the sections below. It is important to keep in mind that each cave is unique, and for this study many different shapes and types of caves were considered. Therefore, alt hough the separations of space, or dichotomized relationships considered include th e left versus right sides, nort h versus south, east versus west, light versus dark, and open versus restricted parts of the cave, the proportion of these spaces in each cave does vary as does the propor tion of each space that was excavated. Certain caves show interesting patterns for so me species, but general patterning replicated in all caves was not identified. Reasons for the lack of generalized patterning and the 159

PAGE 160

distributions or lack of certain species are di scussed in detail below in terms of the possible taphonomic and sample size limitations Crustaceans Crustacean remains were expected to be well-di stributed in cave sites that are near a river or body of water and those that might contain wa ter seasonally or year round within the cave. This is because crabs, even the terrestrial crabs, require water to complete their life cycle since all species have a planktonic stage that involves larval metamorphic changes to the crustaceans body before adulthood is reached (Hickman 1967). Ca ves Branch Rockshelter is located adjacent to a river (Bonor 1998, 2002; Glassman and Bonor Villarejo 2005). The ca ves from Dos Pilas, Cueva de Sangre and Cueva de El Duende, ar e interconnected by an underground river system with the three other caves identi fied at this site (Minjares 2003:24). Naj Tunich does contain areas with pools of water that were both natura lly and culturally constructed within the cave and may have served different ceremonies at the si te (Brady 1989). Stela Cave has some seasonal flooding, but areas of year-round water are not identified (Ishihara and Griffith 2004). All five of the cave sites contained crusta cean remains, and only at CBR were a high proportion (n=65, 5.09%) of crustacean specimens id entified (Table 7-3). This high prevalence of crustacean remains at CBR may be due to the clos e proximity of a river at the site, but there is otherwise no correlation between caves with crustaceans and proximity of water sources, suggesting that their presence may be intentional. Crabs might have been brought to the caves by predators (perhaps from smaller water source s or by flying predator s), but the body portion analysis does not support this interpretation. In all caves, the predominant body portion was the claw in mainly complete forms and this may be consistent with human use. The claw is not better preserved or preferred by natural crab predat ors. Even in CBR, where the crabs might be a natural component, we see a predominance of crab claws. Therefore, we can assume that the 160

PAGE 161

crabs were brought to the caves and used there by humans in most cases. There is no evidence for a predominance of right-sided over left-sided crab claws in these samples. In two of the five caves the crustacean remains were found to be clustered (CBR, NT) (Tables 7-3, 7-17), in one cave no patterning was found (STC) (Table 7-5), and in the other two caves spatial autocorrelation was not possibl e (Tables 7-8, 7-12). At CBR, cokriging found significant clustering in Operation 1A associated with the left, north, a nd light regions of the cave, in the other four sites, cokriging was not possible for cr ustaceans because of the limited distribution of remains (Table 7-3). Visual comparison of the remains showed that in three out of the five caves, there was a predominance of crustacean remains in the light over the dark regions of the cave. The lack of remains within the dark regions of the caves may be due to the size and larger number of excavation units in the light than in the dark re gions of these caves. If considered as a proportion of number of remains found in each of these regions however, we s ee that there are over twice as many (n=3.4) per unit in the light region than those in the dark re gion (n=1.5) per unit. Therefore, the crabs are correlated with the light area of the cave. There is a predominance of left over right sides of the cave in th ree out of the four caves. There ar e also more remains located in the north (three out of five) and th e east (four out of the five). Water has been identified as a sacred part of caves (Ashmore 1989, 1991; Brady 1997; Brady and Ashmore 1999; Prufer and Kindon 2005; Stone 1995). Crustaceans are also known to live in water both within and su rrounding many of these cave sites. No other researchers have identified that crustaceans are symbolically associated with water. A single source about the noises from animals within caves can be used to understand the role of crustaceans within the caves (Bruchez 2007). In 2007 (p.52), Margaret S. Bruchez suggested that the noises produced 161

PAGE 162

by crustaceans within cave could be mistaken fo r larger animal specie s including crocodiles and jaguars by the ancient Maya. In su mmary then, crabs were found to be clustered in, or associated with, the north, east, and light areas of the caves in this sample. The clustering does not seem to have any links to the natural habits or habitats of the crabs or their predat ors, nor any explanation based on taphonomy or excavation practice. Ray-Finned Fishes Ray-finned fish remains were found within four out of the five cave sites. The distribution of ray-finned fish remains varied throughout all of the caves within this study. The largest sample of fish came from Cueva de El Duende (n=399) (Table 7-8) while only three other sites (CBR, STC, CS) contained a total of 19 fish re mains (Tables 7-3, 7-5, 7-12). This disparity between sites is interesting cons idering that the site with the closest water source was CBR, and it had only 13 fish remains. From these very small sample sizes of fish remains, little can be identified in the patterning of their distribution. Cokringing was unabl e to process at all four sites with fish remains and spatial au tocorrelation identified the pattern of fish remains as random at both CBR and CD. Cueva de El Duende did contain a majority of fish remains with the highest majority of remains being identified from the southern, eastern, light a nd open parts of the cave (Tables 7-10, 7-11, 7-12). However, this pattern may be due to the higher number of excavated units in these areas than in the other areas and also due to the separations of space that I identified. At the sited of Cueva de Sangre, the small size and brittleness of fish remains may be a reason for these bones not being found. They were also not found in the large mud samples that were later deflocculated, which was water screen ed through a fine mesh that was 4 mm in size (Brady and Scott 1997). Fish refuse and the caching of fish remains ar e important at many of the ceremonial centers and within burials and caches at elite cen ters (Beaubien 2004; Moholy-Nagy 2004). The Maya 162

PAGE 163

located at inland sites have been associated with the cult of the sea and held marine fish and other animals as showing prestige among the an cient Maya (Pohl 1983). Fish and other marine fauna may have been used to represent water and th is association is seen in the representation of faunal remains in layered caches (Kunen et al. 2001; Mathews and Garber 2004). Only one marine fish remain was identified in all of the caves, this is a cf. Sparisoma sp. (spotlight parrotfish) pharyngeal jaw from Caves Branch Rock shelter. Parrotfish are tropical fish located along shallow reef systems. CBR is located at an inland forested area. Th is indicates that the parrotfish, or this parrotfish pharyngeal jaw, had to have been imported from the coast to the site. Fish are also depicted as offe rings within the Maya codices an d they have been identified as offerings within the Year Bearer ceremonies (Thompson 1972; Bill et al. 2000; Taube 1988; Tozzer 1941) and Burner Cycles (Bricker 1991; Thompson 1972). Within the yearbearer ceremonies there have been fish remains identif ied with both the southern (Bill et al. 2000; Taube 1988; Tozzer 1941) and western (Thompson 1972) directions. The fish remains were sparsely distributed at CBR, STC, and CS, but were identified in large concentrations at CD. The ma jority of the remains at CD were associated with the southern and eastern directions, and their placement within these parts of the cave may have been from the yearbearer ceremonies (Tables 7-6, 7-7). However, the lack of a la rge number of fish remains at all of the sites and there only being one marine specimen identified within these four collections may be an indication of either poor preservation or collection methods at these sites. Fish and marine fish in general may be more closely a ssociated with rituals at above ground than in underground ancient Maya sites. Amphibians There were very few amphibian remains id entified within the cave sites. Amphibians would be found naturally within th e forested habitats of all five cave sites and some may even 163

PAGE 164

have been inhabitants of these caves. Three out of the five sites (CBR, STC, CD) under study had amphibian remains and only a total of 17 spec imens were identified. Due to the small sample size and limited distribution of remains at all three sites, both spatial autocorrelation and cokriging failed to process. Visual analysis did identify a majority of remains being located in the light and open regions of the cave which woul d be expected since these animals are known to live at the mouth of caves. Most of these remain s (n=16) were identified from the order Anura (frogs and toads). Frogs and toads are associat ed with water and are found to live within and surround both caves and cenotes. From these close a ssociations with entran ces to the underworld and the fact that these animals live in watery envi ronments they have been identified as having a relationship with Chac, the rain god (Pohl 1983). Th ey have also been identified as guards to these entrances, caves and cenotes, into the underworld. Toads and frogs were sacredly associated with the underworld and may be the re ason for the low number of remains within the cave sites. The lack of amphibian remains c ould also be due to the small size and low preservation of these animal remains within cave contexts. Turtles Turtles represent an importan t food source for the ancient Maya for both the elite and nonelite residents of many sites. Turtle foods may ha ve also been used as offerings in the Maya codices as tortoise breads are found in many depictions throughout these sacred books (Bricker 1989). Turtle remains were identified at all five of the caves. Spatial auto correlation was able to process at three out of the five sites and the distribution of these remains was identified as random (Tables 7-3, 7-5, 7-12). Cokriging was only able to process at two of these sites and the clusters identified at CBR (Table 7-3) were located around Operations 1A, 1C, and 1D and at CS (Table 7-12) they were concentrat ed at the mouth of the cave. Duri ng visual analysis of the sites there was a higher concentration of turtle remains a ssociated with the eastern part of three out of 164

PAGE 165

the five caves and the open over restricted pa rts in two out of the five caves. There was inconclusive evidence for the relationship between the northern versus southern and light versus dark. The depictions of turtles as offerings have been identified in the Ma ya codices and there is a reference to offering of tortoise bread to the eastern direction during the summer rainmaking ceremony in D.29b-30b (Bricker 1991; Thompson 1972). All of the turtle remains were identified to be freshwater species which have been known to be utilized both for subsisten ce and utilitarian uses (Emery 1997) At the site of Caves Branch Rockshelter a large number of Kinosternidae (mud and musk turtles) and a few Dermatemys mawii remains were identified. The mud and musk tu rtles are believed to ha ve served as a food source for the Maya (Emery 1997), but others have identifies them as non-food sources and only used for ritual events (Pohl 1983). However, the more desired species of food used by the Maya elite was the Dermatemys mawii (Emery 1997:132) which was lacki ng in large numbers at the all of the cave sites in this study. There was an association with the type of turt le bones, carapace and plas tron or turtle shell, recovered from all of the cave sites. The high amount of turtle shell remains could be due to two major factors, either th e subsistence of turtles as a viable food resource and the method of cooking turtles in their shells that has been identified by ot her researchers (Hamblin 1984; Kozelsky 2005) or the use of turtle shell as dr ums during ceremonial ev ents (Pohl 1983). Since none these remains were identified as being burnt, it would be safe to assume that these turtle shells may have been used or associated with ancient instruments. Turtles are also associated with ancient Maya mythology, with the birth of th e maize god from a turtles back and the use of a turtle shell to represent the primordial world of the Maya. This association with turtles and the 165

PAGE 166

creation of the Maya is an important part of thei r use in ceremonies and their close associations with water, earth, and also abunda nce (Kozelsky 2005; Pohl 1983). Lizards There was only a total of 12 lizard remains iden tified in three out of the five caves (CBR, STC, CD) analyzed in this study. Lizards are know n inhabitants of forested areas and would also be expected to live within the opening of caves Spatial autocorrelation wa s only able to process at CBR and identified the lizard distribution as ra ndom within the site (Table 7-3). Due to the low number of specimens and their limited distri bution, cokriging failed to process at all three sites. The low number of lizard remains is interesting considering the strong association of iguanas with ritual offerings of the ancient Maya. Iguanas were offered as breads (Thompson 1972) and tamales (Taube 1988:238) throughout th e Maya codices. The i guana offerings from the Maya codices have been shown to have a stro ng association with ritual events and offerings towards the western direction (B ricker 1991). Only one of the caves, CD, had an association between the western direction and the distribution of five lizard remains in this study (Table 7-7). The lack of strong associations with these rema ins and the western direction in these caves can be due to the preservation of these remains within the caves and also with their actual placement within the caves. Lizards are know n forest inhabitants in the areas surrounding these cave sites. Most of the remains might actually be remnants of these animals as either cave dwellers or refuse from predator consumption. The Maya codices depict iguanas as being parts of breads or tamales. Cooking of these animals would help to quicken the degradation of these remains in a cave setting. The lack of iguana remains in caves may be more associated with the preparation of these materials as offerings than thei r normal preservation within the site. 166

PAGE 167

Snakes Snake remains were associated in four out of the five caves (CBR, STC, CD, CS), but with largely varying amounts of these sp ecimens within these cave sites. Snakes are also inhabitants of caves and their surrounding areas. Spatial au tocorrelation was only able to process the snake remains in two cave sites, and the distribution was considered random (Tables 7-3, 7-5). Cokriging was only able to process at CBR and the concentration of remains was identified around Operation 1D, the small cave at the site (Tab le 7-3). The majority of snake remains were located within the light region for three of the four caves with specimens. However, there was a large amount of snake remains f ound within the dark region of the small cave located at CBR, which is further interesting because these remains were also burned (Table 7-2). Snake handling and the use of snakes in ritual has been identified for the cuch ritual and also in bloodletting ceremonies (Pohl 1983). Snakes are associated with fertility, which may be due to their behavior and the f act that snakes are seen shortly after it rains (Pohl 1981). The snake remains within the caves ma y actually be more of a represen tation of a natural distribution of these animals since they live in these habitats. All of the snake remains were from the axial part of the body including the vertebrae and ribs of these animals. This is an expected finding for these animals since snakes are mainly vertebrae, ribs, and the cranium. However, at many of these sites the snake remains were very similar in size and may actually represent a few intact snakes that died by natural causes. Birds Bird remains have been an important part of the deposition of Maya ritual behavior because more bird remains, excluding turkeys, ar e recovered in ritual de posits than in trash middens (Pohl 1983). There are many raptoral and predatory birds that can be identified as normal cave inhabitants, including owls and hawks. However, there are some examples of cave 167

PAGE 168

faunal assemblages that contain those birds not normally located in areas around caves, for example the bird assemblage from Naj Tunich included Ardeidae (herons and bitterns family), Ramphastidae (toucans family), Psittacidae (parro ts and cockatoos family), Columbidae (doves and pigeons), and Meleagris gallopavo (domesticated turkeys) wh ich are not known inhabitants of caves (Brady 1989). Although my analysis only identif ied birds to the gene ral class of Aves, these remains were located in all of the cave sites, but in varying amounts. Spatial autocorrelation was able to process at three of the sites (CS, CBR, and NT), and the distribution of remains was identified as being randomly distributed at all of the sites. Cokriging processe d at two sites, CBR had a higher concentration in Operation 1A (Table 7-3) and NT had a larger concentration around Lot 3 (Table 7-17). The majority of sites, three out of the five caves (CBR, STC, CD) contained a high amount of bird remains within the light region. Howe ver, there was little to no true patterning of remains in relation to directionality or sidedness. Mary Pohl (1983) identi fied the right side of the body as being associated with the rising s un and the left side of the body with the underworld. Using a small sample (n=14) of bi rd remains from Eduardo Quiroz Cave, Pohl (1983) identifies the larger number of left-sided bones (n=11) as being an example of selective sidedness for bird remains within cave sites. However, the idea that left-sided bird bones were cached within cave sites does not work well within most of the caves used in this study. A total of 71 bones were sided from all of the five cave sites, within these remains a total of 36 were left-sided and 35 were right-sided. This similarity in sidedness does not he lp to prove that there was a selection for specific sides of bird remains. Opossums The opossum remains were identified at all five of the cave sites in this analysis. Spatial autocorrelation was able to process at four out of the five sites, with three sites showing 168

PAGE 169

clustering (Tables 7-3,7-12, 7-17) and one site having a random distribution (Table 7-5). Cokriging was unable to process at any of the sites. The amount of these animals identified at all the sites varied, and the highest concentration we re identified in Cueva de Sangre (n=62). The distribution of opossum remains doe s not adhere to any specific patterns. These remains are distributed throughout the caves in relation to sidedness and directionality. However, within three out of the five caves (CBR, STC, NT), opossum remains were located in the light region and mouth of these caves. This would be expected since opossums tend to live in forested areas and may have used the caves as a place of ref uge. There are no specific findings of opossums in the archaeological record. Howeve r, there are depictions of the Bacabs or the Opossum actors in codices that are f ound in the relation to New Year s and Uayeb rites (Thompson 1972). The depiction of these animals dr essed in human clothing is one of the only depictions of these animals within the codices. Armadillos Armadillos are found throughout the Maya region and are known inhabitants of forested areas and agricultural field where they burrow their homes into. Armadillo remains were found in all five cave sites, but ther e were drastic differences in the amount of remains identified in those caves from Belize (CBR and STC) and those from Guatemala (CD, CS, and NT). The amounts of remains identified in the Belizean collections are 182 remains while only 11 remains were identified at the Guatemalan sites. The di sparity in numbers of specimens between regions is unclear but an inte resting pattern considering how la rge the differences are. Spatial autocorrelation was only able to process for the two caves with large sample sizes, and CBR (Table 7-3) had a random distri bution while STC (Table 7-5) wa s somewhat clustered. Cokriging was only able to process at CBR and the highest concentrations were located around Operation 1A (Table 7-3). 169

PAGE 170

The majority of remains were located in the light region; these remains were specifically located in three out of the five sites including both Belize sites. Ther e was no real spatially distinct distribution of the armadillo remains in relation to the sidedness and directionality within these caves. Armadillos represent a known food s ource to the Maya and they are depicted with only a few in the hunting and trapping almanacs of the codices. The armadillo is associated with fertility throughout Mesoamerica (Pohl 1983). At the site of CBR, there is a large concentration of armadillo remains (n=102) and this may be related to large number of known armadillo burrows at the site. The armadillo would find th e mixed grave fill soils from CBR as hospitable places to live. The movement of armadillos within archaeological sites has been shown to affect the distribution of artifacts a nd their movement through the soils may have caused some impacts on the archaeological record (Araujo and Marcelino 2003). Bats Bats represented a link to the underworld to the ancient Maya, an association that is easily understood because they are natural inhabitants of caves. Bat remain s were identified at all five of the sites, but the amounts of these remains were low at four out of the five sites. These low numbers of remains are unexpected since these site s serve as home for many bat species in the Maya region. Spatial autocorrelation was identified the pattern as random at three out of the five (STC, CD, NT) sites that it was able to proc ess (Tables 7-5, 7-8, 7-17) Cokriging failed to process at all five sites. Cueva de El Duende was the only site to contain a high number of remains (n=632). The site with the lowest amount (n=1) was Caves Bran ch Rockshelter. The other sites only contained a few samples each including Stela Cave (n=13) Cueva de Sangre (n=12), and Naj Tunich (n=6). The small sample sizes did not help in producing any significant patterns. The sidedness of elements included more left-sided than right sid ed elements at four of the five sites. The most 170

PAGE 171

significant cave with a large number of left sides (n=7) with no ri ght sides was Stela Cave (Table 7-5). The distribution of these remains was only lo cated in Chamber 1 which was associated with the light and open regions of the cave. An unexp ected result considering that the bats were known to live in the back part of the cave in Chamber 3. At Cueva de El Duende, the majority of remains over 99% were located in the open and light region as well (Table 7-7). The natural occurrence of bats in all of the cave sites makes their distribution seem more natural than cultural in origin. Primates Primates were closely associated with deer by the ancient Maya and were represented pictorially in the Maya codices with deer parts, including antler s. Deer were depicted with monkey attributes, including long ta ils (Pohl 1983). Monkeys were al so associated with the arts and creation myths of the ancient Maya (Baker 1992). Primates from the Maya region include two monkey species, the howler monkeys ( Aloutta palliate and Alouatta pigra ) and the spider monkey ( Ateles geoffroyi ). There were only two monkey remains identified at one out of the five sites from this study. These two remains from Naj Tunich were in a single unit and could not be analyzed for any type of patterning (Table 7-17). The lack of primate remains in the cave sites mirror those at surface sites because monkey remain s were identified rarely in the archaeological record. This may have something to do with th e close association of humans and monkey. The ancient Maya identified the phyl ogenetic relationship between humans and monkeys, and this may be the reason they were not be ing used in cave ritual events. Canids The Canidae family includes coyotes, dogs, foxe s, jackals, and wolves and within this larger group two main species of canids fr om the Maya region were identified including Canis familiaris (domestic dog) and Urocyon cinereoargenteus (gray fox). Canid remains were 171

PAGE 172

identified in all five of th e cave sites in this study. Howeve r, there were not many remains identified at three of these sites, CBR (n=3), STC (n=4), and NT (n=4). Since dogs are known to be ritually significant animals it would be expect ed that larger numbers of canid remains would be located in these caves. However, the canid remains from these ritually significant contexts were lacking in numbers to support their known ritual importance to the ancient Maya. Cokriging failed to process at a ny of these sites, and spatial au tocorrelation was identified as being random at two of the five sites. The domestic dog represents the only animal, be sides the turkey, that was a domesticated species in Mesoamerica. Dogs, and in general al l canids, are believed to be guides on the river that is in the underworld and they are also well known ritual offerings as both sacrifice and as burial offerings (Pohl 1983). The inclusion of the domestic dog at many of these sites represents a human placed species that was ritually interr ed in these areas. In contrast to the known inhabitants, such as the opossum, bat, and ar madillo, the dog remains in these caves could only have been placed through human processes. At th e site of Cueva de Sangre a large number of dog teeth were identified at the site, this is very similar to the large sample of dog teeth found at Actun Polbilche (Pendergast 1974). Dogs have also been identified as offerings in Yearbearer ceremonies in both the Maya codices, specifically in the Madrid Codex (pages 34 to 37) (Bill et al. 2000; Bricker 1989; Love 1986; Taube 1988; Vail 2004; Vail a nd Bricker 2004) and ethnographic accounts from Landa (Tozzer 1941). D og as sacrificial offerings make the low number of canid remains in the caves even more significant. Felids Another important carnivore family located in some of the caves includes the felid or cat family. There were only two caves with identified cat remains. This is very significant because the largest cat in the Maya region, the jaguar, is a symbol of wealth and cultural distinction in the 172

PAGE 173

Maya region and Mesoamerica in general. Jaguar pelts and paws were worn and used to show distinction among Maya hierarchy. Therefore the absence of their remains at these sites is an interesting occurrence. Large cat remains are usually found in a ssociation with burials that include pelts, paws, and also teeth. The pelts a nd paws may have also played a role in the accession ceremonies for kings that have been hypothesized to have occurred within the caves (Pohl 1983). The mediumand large-sized cats of the Maya region are known inhabitants of caves, and have been traditionally identified with the setting sun, the underworld, and with water (Pohl 1983). The small number of feline remains (n=11) is interesting because it would be expected that a larger amount of ritually and elite-centric remains would be associated with these sites. Naj Tunich remains (n=3) include a radius, a humerus, and a metapodial, which may have all been associated with cat paws. The remains at Cueva de El Duende (n=8) were all cranial fragments from a single unit, another interesting finding sin ce these remains may represent either a skull or multiple cat skulls (Table 7-8). With a larger sa mple more patterning might be identified, but due to the small sample size no distinct distribution patterns can be identified. Raccoons The Procyonidae includes raccoons, coatis, a nd kinkajous. These animals are known as a food source to the ancient Maya and symbolically represented th e negative aspects of people including thievery and gluttony (Emery 1997). Ther e were very small samples of remains (n=3 to 4) from three out of the five sites (CBR, ST C, NT). These remains were located in different places within all the caves and no specific spatia l relationship can be identified for the raccoon family. These remains may be from natural contex ts since known predators of the raccoons lived in caves and also because no known ritual signi ficance has been assigned to this family. 173

PAGE 174

Tapirs The largest mammal in the Maya region is the Bairds tapir ( Tapirus bairdii). Due to the illusive and timid behavior of this animal, the tapi r is rarely identified in the caves. In this study only one site, Naj Tunich has a sa mple of tapir remains. These remains were associated with the left side of the cave, the nort h, east, and dark regions of the cave (Tables 7-13, 7-14, 7-15, 7-16). The elements identified also varied, but the majority of metapodials and tarsals/carpals bone identified may be indications that the lower le g and lower arm bones may have been distributed at the site (Table 7-17). Ther e are no known or described associ ations with the tapir in Maya ritual caches or within other cave sites. Artiodactyls The artiodactyls, or even-toed ungulates, represent so me of the major food sources to the ancient Maya. The artiodactyls include the peccary and deer families. These two families are known to have played a major role in both subsis tence and also in ritu al. This large taxonomic grouping was identified at all five of the sites. It is not known if the an cient Maya identified the close taxonomic relationship of this large group but both the peccary and deer were analyzed separately and are discu ssed in detail below. Peccaries There are two major peccary species identifi ed in the Maya region including the whitelipped peccary (Tayassu pecari ) and the collared peccary ( Tayassu tajacu ). Due to the very similar nature of peccary remains, these two groups were not individually identified in this study. Peccaries are depicted in the Maya codices, but th ese depictions are rare and only found with the animal trapping almanacs. The most important el ement of the peccary was its cranium, and there are a majority of carved craniums that have be en identified in the archaeological record. The most interesting being the distribution of peccar y remains on the island of Cozumel, with the 174

PAGE 175

postcranial remains identified only in Postclas sic burials, while the sk ulls were found only in housemounds (Hamblin 1984; Pohl 1983). Peccary remains were identified in four out of the five cave sites in this study. Spatial autocorrelation at Stela Cave a nd Naj Tunich identified the patte rn of peccary remains as random (Table 7-5, 7-17). Only at Naj T unich was cokriging able to proc ess and at the site the highest concentration of remains was identified at Lots 1, 16, 24, and 57 (Table 7-17). All of the remains were identified in varying parts of the caves. However, there wa s an interesting trend in the elements identified at all of the sites. Within th ree of the sites the majority of remains included cranial and teeth remains, out of the fifteen remain s identified twelve were cranial in origin. Naj Tunich included the largest sample from all of th e sites (n=25), and at this site the majority of remains were limb elements (Table 7-17). This may be due to the use of pe ccary skulls at surface sites instead of within caves. By placing animal remains within cave site, the Maya would be taking a valuable product out of their ritual economy (Brady 2005). The rarity or sacredness of peccary skulls, many of which are carved, at surface sites may have been the limiting factor in their addition to ritual cave faunal assemblages. Deer There are two major species of deer identified and utilized by the Maya including the white-tailed deer ( Odocoileus virginianus ) and the brocket deer ( Mazama sp.). In Pohls (1983) groundbreaking analysis of ritual us e and association of animal remains, deer were the first and most important animal species analyzed. Deer rema ins at archaeological sites in the Maya region usually represent the largest identified species in a collection. Deer served as a major source of protein to the Maya and were ri tually significant offerings for th e elite and gods of the ancient Maya. Deer also represented the sun and as a source of renewal. The Maya codices contain many different depictions of deer haunch offerings to both the gods and the ruling elites. Deer remains 175

PAGE 176

represented the largest gr oup of artiodactyls at all five cave sites. The numbers of deer remains were usually higher for white-tailed deer than brocket deer at all but one of the sites, Cueva de Sangre. Although this discrepancy is due to a large number of teeth, 31 out of the 35, remains identified to brocket deer at the site. For deer remains at the cave sites, there were few patterns identified. Deer are associated with the sun, more specifically the rising sun. This association would place the directional importance of deer with the east a nd also with the sun at it apex, or the north. Close associations with the eastern and northern directions were identified in the codices for the New Years or Uayeb ceremonies and also with some of the burner cycle ceremonies (Bricker 1991, 1997; Thompson 1972). The only site to show a greater concentration of deer remains in the north (92.1%) and east (85.1%) than the south and west was at the site of Naj Tunich (Tables 7-10, 711). However, none of the other sites with the lo wer amounts of deer remains showed this same pattern. The use of deer haunche s as ritual offerings would in volve a larger number of deer remains being associated with these elements A high amount of forelimbs and hindlimbs were identified within three sites containing the sm aller-sized samples, CBR (7 out of 11 remains) (Table 7-3), STC (11 out of 16 remains) (Table 75) and at CD (15 out of the 18 remains) (Table 7-8). However, in the larger samples there were fewer numbers of limb and distal elements. Rodents Rodents are known in habitants of caves and also known food sources for those animals that inhabit these caves. It woul d be expected that the second la rgest amount of remains, behind bats, identified within cave sites would be rodents. However, the amount of rodent remains varied throughout the five sites. The concentrations of rodent rema ins at the sites from largest to smallest include, Stela Cave (n=222), Cueva de Sangre (n=190), Caves Branch Rockshelter (n=30), Cueva de El Duende (n=21), and Naj Tunich (n=20). Spatial autocorrelation was able to 176

PAGE 177

process at four out of the five sites and the patt erns of remains were identified as random (Tables 7-3, 7-5, 7-12, 7-17). Cokriging wa s only able to process at two sites and at CBR these remains were concentrated around Operations 1B and 1C (Table 7-3), while at CS the highest concentration of remains were aroun d CS6-8-1 and CS6-3-1 (Table 7-12). The lack of rodent remains from some of th ese sites, including Naj Tunich and Cueva de El Duende, may be due to the collection met hods. Rodents, like other small animals, are associated with otherworldliness and also the underworld (Pohl 1983). The patterns of distribution of remains varied th roughout these sites and failed to produce a possible pattern for their spatial distribution. Small-si zed rodents are usually identifie d as intrusive species in the zooarchaeological record at mo st sites; however, new ethnogra phic and ethnohistoric research has identified a possible trad e of and utilization of th ese small protein sources. Squirrels Squirrel remains were only found at one site, St ela Cave, and in very limited numbers with a total of four specimens. However, there are strong ethnographic and epig raphic identifications of squirrels as ritual offerings. In Muluc years, squirrels were r ecorded as being offered in both Landas Relacion de las Cosas de Maya (Tozzer 1941) and also within the yearbearer pages of the Madrid Codex (M.36) (Love 1986). Muluc years are associated with the eastern direction and these four remains were identified in the eastern part of the cave. However, it is difficult to assume the ritual significance of this small sample size. These remains were located in Chamber 1 of STC which falls within the open and light pa rt of the cave (Table 7-4). The squirrel remains may actually represent remains fr om carnivore droppings or possibl y the natural death of these animals within the cave. 177

PAGE 178

Agoutis and pacas The largest and most well-known food source fro m the rodent order includes the mediumsized rodents from the Agoutidae (pacas) and Dasyproctidae (agoutis) families. Agoutis and pacas were identified at all five of the cave s ites in various amounts. Spatial autocorrelation identified the pattern of distribut ion as random at four out of the five sites it was able to process at (Tables 7-3, 7-5, 7-12, 7-17). Cokriging was able to process at th ree sites including CBR around Operation 1D (Table 7-3), at CS around lower regions of CS1 (Table 7-12), and at NT concentrated around Lot 57 (Table 7-17). These two families have not been identified to have specific ritual significance to the ancient Maya by previous research projects. There were no specific spatial patterns identified for their distribution within thes e caves because all aspects of the separations of space varied. Cottontail rabbit Finally, the last taxonomic group to be discussed is the cottontail rabbit. The rabbit has been identified as the second most consumed an imal after deer (Emery 1997). Symbolically, the rabbit has been associated with fertility and vegetation because of its known preponderance to reproduce rapidly (Emery 1997). The rabbit, like th e deer, is also associ ated with the ancient Maya sky. The moon is believed to be the profile of a rabbit in the s ky (Emery 1997). The rabbit has also been pictorially repres ented as a scribe in the Maya codices. There were a few rabbit (n=7) remains identified within three of the five caves (CBR, STC, NT). Due to the small sample size spatial patterning for these remains could not be identified. Summary GIS and visual analysis were able to identify patterning for the following taxonomic groups, including the remains from crustaceans, turtles, peccary, deer, and squirrel remains. These patterns were formed using small samp le sizes and should be taken as tentative 178

PAGE 179

identifications for the intentiona l distribution of faunal remains within cave site. Animals have strong sacred associations with Maya rituals and in caves, these associations may be the reasoning for and lack of some taxonomic groups A further discussion of these results is discussed in the next chapter. 179

PAGE 180

Table 7-1. Separation of Space Su mmaries for Left versus Right Sides and North versus South Directions at Caves Branch Rockshelter, Belize. Region of the cave (n=380) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP Left/North Crabs 65 33 21.0% 50.8% (n=157) Fish 13 3 1.9% 23.1% Amphibia 5 2 1.3% 40.0% Turtles 42 14 8.9% 33.3% Iguana 6 3 1.9% 50.0% Snakes 37 6 3.8% 16.2% Birds 21 11 7.0% 52.4% Opossums 8 2 1.3% 25.0% Armadillos (w/ scutes) 102 50 31.8% 49.0% Armadillos (non-scutes) 31 21 13.4% 67.7% Bats 1 0 0.0% 0.0% Dogs 3 2 1.3% 66.7% Raccoons 4 2 1.3% 50.0% Artiodactyls 18 9 5.7% 50.0% Peccaries 3 1 0.6% 33.3% Deer 9 4 2.5% 44.4% Rodents 30 8 5.1% 26.7% Agoutis and pacas 24 12 7.6% 50.0% Rabbit 1 0 0.0% 0.0% Total: 380 157 100.0% Right/South Crabs 65 32 14.3% 49.2% (n=223) Fish 13 10 4.5% 76.9% Amphibia 5 3 1.3% 60.0% Turtles 42 28 12.6% 66.7% Iguana 6 3 1.3% 50.0% Snakes 37 31 13.9% 83.8% Birds 21 10 4.5% 47.6% Opossums 8 6 2.7% 75.0% Armadillos (w/ scutes) 102 52 23.3% 51.0% Armadillos (non-scutes) 31 10 4.5% 32.3% Bats 1 1 0.4% 100.0% Dogs 3 1 0.4% 33.3% Raccoons 4 2 0.9% 50.0% Artiodactyls 18 9 4.0% 50.0% Peccaries 3 2 0.9% 66.7% Deer 9 5 2.2% 55.6% Rodents 30 22 9.9% 73.3% Agoutis and pacas 24 12 5.4% 50.0% Rabbit 1 1 0.4% 100.0% Total: 380 223 180

PAGE 181

181 Table 7-2. Separation of Space Su mmaries for Light versus Dark and Open versus Restricted Regions at Caves Branch Rockshelter, Belize. Region of the cave (n=380) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP Light/Open Crabs 65 62 20.1% 95.4% (n=309) Fish 13 7 2.3% 53.8% Amphibia 5 5 1.6% 100.0% Turtles 42 36 11.7% 85.7% Iguana 6 5 1.6% 83.3% Snake 37 22 7.1% 59.5% Birds 21 21 6.8% 100.0% Opossums 8 6 1.9% 75.0% Armadillos (w/ scutes) 102 78 25.2% 76.5% Armadillos (non-scutes) 31 26 8.4% 83.9% Bats 1 1 0.3% 100.0% Dogs 3 3 1.0% 100.0% Raccoons 4 2 0.6% 50.0% Artiodactyls 18 16 5.2% 88.9% Peccaries 3 3 1.0% 100.0% Deer 9 7 2.3% 77.8% Rodents 30 26 8.4% 86.7% Agoutis and Pacas 24 18 5.8% 75.0% Rabbit 1 1 0.3% 100.0% Total: 380 309 100.0% Dark/Restricted Crabs 65 3 4.2% 4.6% (n=71) Fish 13 6 8.5% 46.2% Amphibia 5 0 0.0% 0.0% Turtles 42 6 8.5% 14.3% Iguana 6 1 1.4% 16.7% Snake 37 15 21.1% 40.5% Birds 21 0 0.0% 0.0% Opossums 8 2 2.8% 25.0% Armadillos (w/ scutes) 102 24 33.8% 23.5% Armadillos (non-scutes) 31 5 7.0% 16.1% Bats 1 0 0.0% 0.0% Dogs 3 0 0.0% 0.0% Raccoons 4 2 2.8% 50.0% Artiodactyls 18 2 2.8% 11.1% Peccaries 3 0 0.0% 0.0% Deer 9 2 2.8% 22.2% Rodents 30 4 5.6% 13.3% Agoutis and Pacas 24 6 8.5% 25.0% Rabbit 1 0 0.0% 0.0% Total: 380 71 100.0%

PAGE 182

Table 7-3. Summary of NISP, % Taxa, Left Sided Elements, Body portions, % NISP Burned, Spatial Auto Co rrelation and Cokriging at Caves Branch Rockshelter, Belize. Taxonomic Group NISP (n=545) % Taxa % of Left sided elements Body Portion comment % NISP Burned Spatial Autocorrelation Cokriging Crabs 65 17.1% 28/56 = 50% 98.5% claws None Clustered, not random chance Op 1A cluster (Left/North/Light) Fish 13 3.4% 1/1 = 100% cranial 4, axial 4, UID 6 None Slight clustering No clusters Amphibia 5 1.3% 2/3 = 66.7% appendicular 4, axial 1 None Didnt run Didnt run Turtles 42 11.1% 2/11 = 18.2% 92.8% carapace/plastron, 5, 11.9% axial, 1, 2.4% long bone 2/42, 4.8% Random Cluster between Op 1A, 1B, + 1D Iguana 6 1.6% 1/4 = 25% Appendicular 1, cranial 2, axial 3 None Random Didnt run Snakes 37 9.7% N/A All verts 13/37, 35.1% Not clustered, not dispersed Op 1D cluster Birds 21 5.5% 1/4 = 25% App 14, axial 4, UID 3 1/21, 4.8% Slight clustering Op 1A cluster Opossums 8 2.1% 2/7 = 28.6% Cranial 4, axial 3, humerus 1 None Clustered, not random chance Didnt run Armadillos 102 26.8% See Below Scutes 71, Read below 13/102, 12.8%, 10/ 102, 9.8% scute Random, not clustered or dispersed Higher probability for N/S, L/R, D/Light Armadillos (non-scute) 31 4/11 = 36.3% metapodial 15, vertebrae 6, radius 2, fibula 2, calcaneum 2, humerus 1, ulna 1, tibia 1, astragalus 1 Random, not clustered or dispersed Higher probability for N/S,L/R, D/Light Bats 1 0.3% 1/1 = 100% 1 innominate None Didnt run Didnt run Dogs 3 0.8% 1/2 = 50% 2 crania, 1 distal None Didnt run Didnt run Raccoons 4 1.1% 3/4 = 75% 2 ulnae, 2 tibiae None Didnt run Didnt run Artiodactyls 18 4.7% 4/8 = 50% 6 appendicular, 2 cranial, 1 axial 2/18, 11.1% Random, neither dispersed nor clustered Clustering in Op 1A and 1D Peccaries 3 0.8% 2/3 = 66.7% All teeth 1/3, 33.3% Didnt run Didnt run Deer (9 wtd) 11 2.9% 4/8 = 50% App 6, cranial 2, axial 1 None Random Didnt run Rodents 30 7.9% 19/26 = 71% App (19) cranial (11) 1/30, 3.3% Random, not clustered nor dispersed Clustering around 1B and 1C Agoutis and pacas 24 6.3% 12/23 = 52.2% Tooth (18), App (6) 4/24, 16.7% Clustering, random chance Clustering at 1D Rabbit 1 0.3% n/a metatarsal None Didnt run Didnt run 182

PAGE 183

Table 7-4. Separation of Space Summ aries for Left versus Right Si des, North versus South, East versus West Directions, Light versus Dar k, and Open versus Restricted Regions at Stela Cave, Belize. Region of the cave (n=498) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP Left/ North/East/ Light/Open Crabs 11 11 2.5% 100.0% (n=447) Fish 4 3 0.7% 75.0% Amphibians 1 1 0.2% 100.0% Turtles 43 43 9.6% 100.0% Lizards 1 1 0.2% 100.0% Snakes 18 17 3.8% 94.4% Bird 39 39 8.7% 100.0% Opossums 13 13 2.9% 100.0% Armadillo 80 78 17.4% 97.5% Bats 13 13 2.9% 100.0% Dogs 4 3 0.7% 75.0% Racoons 3 3 0.7% 100.0% Artiodactyls 30 27 6.0% 90.0% Peccary 7 7 1.6% 100.0% Deer 16 15 3.4% 93.8% White-tailed deer 4 4 0.9% 100.0% Brocket deer 4 4 0.9% 100.0% Rodent 222 181 40.5% 81.5% Squirrel 4 4 0.9% 100.0% Agoutis and pacas 8 7 1.6% 87.5% Rabbits 4 3 0.7% 75.0% Total: 498 447 100.0% Right/South/West/Dark/Restricted Crabs 11 0 0.0% 0.0% (n=51) Fish 4 1 2.0% 25.0% Amphibians 1 0 0.0% 0.0% Turtles 43 0 0.0% 0.0% Lizards 1 0 0.0% 0.0% Snakes 18 1 2.0% 5.6% Bird 39 0 0.0% 0.0% Opossums 13 0 0.0% 0.0% Armadillo 80 2 3.9% 2.5% Bats 13 0 0.0% 0.0% Dogs 4 1 2.0% 25.0% Racoons 3 0 0.0% 0.0% Artiodactyls 30 3 5.9% 10.0% Peccary 7 0 0.0% 0.0% Deer 16 1 2.0% 6.3% White-tailed deer 4 0 0.0% 0.0% Brocket deer 4 0 0.0% 0.0% Rodent 222 41 80.4% 18.5% Squirrel 4 0 0.0% 0.0% Agoutis and pacas 8 1 2.0% 12.5% Rabbits 4 1 2.0% 25.0% Total: 498 51 100.0% 183

PAGE 184

Table 7-5. Summary of NISP, % Taxa, Left Sided Elements, Body portions, % NISP Burned, Spatial Auto Co rrelation and Cokriging at Stela Cave, Belize. Taxonomic Group NISP (n=498) % Taxa % of Left sided elements Body Portion comment % NISP burned Spatial Autocorrelation Crabs 11 2.2% 3/9 = 33% 10, 91% claw, 1, 9% body frag None Random, neither clustered nor dispersed Fish 4 0.8% 1/2 = 50% 2, 50% dentary, 1, 25% vert, 1, 25% UID None Didnt run Amphibians 1 0.2% 1/1 = 100% 1, 100% humerus None Didnt run Turtles 43 8.6% 10/20= 50% 42, 97 .7% plastron/carapace, 1, 2.3% humerus None Dispersed, random chance Lizards 1 0.2% N/A 1, 100% mandible frag None Didnt run Snakes 18 3.6% N/A 18, 100% vertebrae 1 blackened Clustering, random chance Bird 39 7.8% 10/26 = 38.5% 27, 69.2% limb, 9, 23.1% axial, 3, 7.7% UID None Random, neither clustered nor dispersed Opossums 13 2.6% 4/6 = 67% 7, 53.8% cranial, 6, 46.2% axial None Random, neither clustered nor dispersed Armadillo 80 16.1% 9/15 = 60% 51, 63.8% scutes, 4 verts, 1 rib, 1 scapula, 2 humerus, 3 ulna, 7 metapodials, 1 inn. 1 patella, 4 tibia, 2 fibula, 1 astragal, 1 calcaneus 1 calcined (scute), 2 blackened (long bones) Clustering, random chance. Bats 13 2.6% 7/7 = 100% 10, 76.9% front limb, 3, 23.1% cranial None Random, neither clustered nor dispersed Dogs 4 0.8% 1/3 = 33% 3 limb, 1 cranial 1 calcined gray Didnt run Racoons 3 0.6% 2/2 = 100% 1 cranial frag, 1 metapodial, 1 scapula None Didnt run Artiodactyls 30 6.0% 8/19 = 42.1% 20, 66.7% limb, 9, 30% cranial, 1, 3.3% axial 5 blackened, 1 calcined gray, 3 calcined white Random, neither clustered nor dispersed Peccary 7 1.4% 2/5 = 40% 3 cranial, 2 teeth, 2 metapodial s 3 blackened Random, neither clustered nor dispersed Deer 16 3.2% 4/11 = 36.4% 5, 31.3% cranium, 6, 37.5% hindlimb, 2, 12.5% front limb, 3, 18.8% metapod, 1, 6.3% scapula 2 blackened, 1 calcined gray, 1 calcined white Random, neither clustered nor dispersed White-tailed deer 4 0.8% 1/4 = 25% 2 astragal us, 1 metacarpal, 1 mandible None Didnt run Brocket deer 4 0.8% 1/3 = 33.3% 1 ulna, 1 molar, 1mtt, 1mtp 1 calcined gray, 1 calcined white Didnt run Rodent 222 44.6% 114/213 = 53.5% 120, 54.1% hind limb, 67, 30.2% cranium, 22, 9.9% front limbs, 13,5.8% axial None Random, neither clustered nor dispersed Squirrel 4 0.8% 3/4 = 75% 2 femurs, 1 mandible, 1 tibia None Didnt run Agoutis and pacas 8 1.6% 4/7 = 57.1% 5 (62.5%) teeth, 1 (12.5%) ulna, 1 (12.5%) tibia, 1 astragulus (12.5%), None Dispersed, random chance Rabbits 4 0.8% 2/4 = 50% 3 femurs, 1 mandible None Didnt run 184

PAGE 185

Table 7-6. Separation of Space Summaries for North versus South Directions at Cueva de El Duende, Guatemala. Region of the cave (n=1270) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP North Crabs 1 0 0.0% 0.0% (n=611) Fish 399 40 6.5% 10.0% Lepisosteidae (gars) 39 2 0.3% 5.1% Siluriformes (catfish) 8 4 0.7% 50.0% Amphibians 11 4 0.7% 36.4% Turtles 15 1 0.2% 6.7% Lizards 5 0 0.0% 0.0% Snakes 50 5 0.8% 10.0% Bird 37 5 0.8% 13.5% Opossums 29 11 1.8% 37.9% Armadillo 2 0 0.0% 0.0% Bats 632 509 83.3% 80.5% Dogs 22 2 0.3% 9.1% Cats 8 0 0.0% 0.0% Artiodactyls 30 16 2.6% 53.3% Deer 18 7 1.1% 38.9% White-tailed deer 15 6 1.0% 40.0% Brocket deer 1 0 0.0% 0.0% Rodent 21 16 2.6% 76.2% Agoutis and Pacas 8 2 0.3% 25.0% Total: 1270 611 100.0% South Crabs 1 1 0.2% 100.0% (n=659) Fish 399 359 54.5% 90.0% Lepisosteidae (gars) 39 37 5.6% 94.9% Siluriformes (catfish) 8 4 0.6% 50.0% Amphibians 11 7 1.1% 63.6% Turtles 15 14 2.1% 93.3% Lizards 5 5 0.8% 100.0% Snakes 50 45 6.8% 90.0% Bird 37 32 4.9% 86.5% Opossums 29 18 2.7% 62.1% Armadillo 2 2 0.3% 100.0% Bats 632 123 18.7% 19.5% Dogs 22 20 3.0% 90.9% Cats 8 8 1.2% 100.0% Artiodactyls 30 14 2.1% 46.7% Deer 18 11 1.7% 61.1% White-tailed deer 15 9 1.4% 60.0% Brocket deer 1 1 0.2% 100.0% Rodent 21 5 0.8% 23.8% Agoutis and Pacas 8 6 0.9% 75.0% Total: 1270 659 100.0% 185

PAGE 186

186 Table 7-7. Separation of Space Summaries for East versus West Directions, Light versus Dark, and Open versus Restricted Regions at Cueva de El Duende, Guatemala. Region of the cave (n=1270) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP East/ Light/Open Crabs 1 1 0.1% 100.0% (n=1182) Fish 399 377 31.9% 94.5% Lepisosteidae (gars) 39 39 3.3% 100.0% Siluriformes (catfish) 8 4 0.3% 50.0% Amphibians 11 7 0.6% 63.6% Turtles 15 14 1.2% 93.3% Lizards 5 5 0.4% 100.0% Snakes 50 47 4.0% 94.0% Bird 37 32 2.7% 86.5% Opossums 29 18 1.5% 62.1% Armadillo 2 2 0.2% 100.0% Bats 632 626 53.0% 99.1% Dogs 22 20 1.7% 90.9% Cats 8 8 0.7% 100.0% Artiodactyls 30 14 1.2% 46.7% Deer 18 11 0.9% 61.1% White-tailed deer 15 9 0.8% 60.0% Brocket deer 1 1 0.1% 100.0% Rodent 21 5 0.4% 23.8% Agoutis and Pacas 8 6 0.5% 75.0% Total: 1270 1182 100.0% West/Dark/Restricted Crabs 1 0 0.0% 0.0% (n=88) Fish 399 22 25.0% 5.5% Lepisosteidae (gars) 39 0 0.0% 0.0% Siluriformes (catfish) 8 4 4.5% 50.0% Amphibians 11 4 4.5% 36.4% Turtles 15 1 1.1% 6.7% Lizards 5 0 0.0% 0.0% Snakes 50 3 3.4% 6.0% Bird 37 5 5.7% 13.5% Opossums 29 11 12.5% 37.9% Armadillo 2 0 0.0% 0.0% Bats 632 6 6.8% 0.9% Dogs 22 2 2.3% 9.1% Cats 8 0 0.0% 0.0% Artiodactyls 30 16 18.2% 53.3% Deer 18 7 8.0% 38.9% White-tailed deer 15 6 6.8% 40.0% Brocket deer 1 0 0.0% 0.0% Rodent 21 16 18.2% 76.2% Agoutis and Pacas 8 2 2.3% 25.0% Total: 1270 88 100.0%

PAGE 187

Table 7-8. Summary of NISP, % Taxa, Left Sided Elements, Body portions, % NISP Burned, Spatial Auto Co rrelation and Cokriging at Cueva de El Duende, Guatemala. Taxonomic Group NISP (n=1270) % Taxa % of Left sided elements Body Portion comment % NISP burned Spatial Autocorrelation Crabs 1 0.1% N/A 1 claw None Didnt run Fish 399 31.4% 18/43= 41.9% 148, 37.1% UID, 15, 12.8% Cranial, 200, 50.1% Axial None Random, neither clustered nor dispersed Lepisosteidae (gar) 39 3.1% 5/8 = 62.5% 25, 64.1% scales, 4, 10.3% vertebrae, 10, 25.6% cranial None Didnt run Siluriformes (catfish) 8 0.6% 1/3 = 33% 3 cranial, 1 coracoid, 4 postcranial None Didnt run Amphibians Notes: 10 Anura 11 0.9% 4/8 = 50% 1, 9.1% vertebra, 1, 9.1% maxilla,3, 27.2% humerus, 4, 36.4% innominate, 2, 18.2% long bone None Didnt run Turtles 15 1.2% 3/9 = 33% 12, 80% carapace, 1, 6.7% coracoids, 1, 6.7% humerus, 1, 6.7% innominate None Didnt run Lizards 5 0.4% 2/3 = 66.7% 5, 100% dentary None Didnt run Snakes 50 3.9% N/A 38, 76% ribs, 12, 24%vertebrae None Didnt run Bird 37 2.9% 13/25 = 52% 15, 40.5% hind limbs, 10, 27% front limbs, 6, 16.2% axial, 8.1% cranial, 3, 8.1% long bone fragments None Didnt run Opossums 29 2.3% 14/26 = 53.8% 14, 48.3% cranial, 5, 17.2% axial, 5, 17.2% front limb, 5, 17.2% hind limb None Didnt run Armadillo 2 0.2% N/A 2, 100% scutes None Didnt run Bats 632 49.8% 116/209 = 55.5% 309, 48.9% phalanx fragments, 224, 35.4% front limbs, 38, 6% axial, 33, 5.2% hind limb, 28, 4.4% cranial None Clustering, random chance. Dogs 22 1.7% 10/21 = 47.6% 1, 4.5% cranial fragment, 8, 36.4% teeth, 1, 4.5% atlas vertebrae, 2, 9% innominates, 6, 27.2% front limbs, 4, 18.1% hind limbs 1 blackened Random, neither clustered nor dispersed Cats 8 0.6% 3/8 = 37.5% 8, 100% cranial fragments None Didnt run Artiodactyls 30 2.4% 10/13 = 76.9% 2, 6.7% cranial, 5, 16.7% axial, 4, 13.3% front limb, 11, 36.7% hind limb, 8, 26.7% distal appendages 3 blackened Random, neither clustered nor dispersed 187

PAGE 188

188 Table 7-8. Continued Taxonomic Group NISP (n=1270) % Taxa % of Left sided elements Body Portion comment % NISP burned Spatial Autocorrelation Deer 18 1.4% 8/11 = 72.7%, 2, 11.1% cranial, 1,5.6% axial, 4, 22.2% front limb, 9, 50% hind limb, 2,11.1% distal appendages 2 blackened Didnt run White-tailed deer 15 1.2% 7/10 = 70% 2 cranial, 1 axial, 2 front limb, 8 hind limbs, 2 distal 2 blackened Didnt run Brocket deer 1 0.1% N/A 1 metacarpal None Didnt run Rodent 21 1.7% 8/15 = 53.3% 4, 19% cranial, 7, 33.3% axial, 1,4.8% front limb, 9, 42.9% hind limb None Didnt run Agoutis and Pacas 8 0.6% 2 of 8 = 25% 2, 25% humerus, 1, 12.5% ulna, 2, 25% femur, 3, 37.5% tibia None Didnt run

PAGE 189

Table 7-9. Separation of Space Summaries for Le ft versus Right Sides at Cueva de Sangre, Guatemala. Region of the cave (n=299) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP Left Crustaceans 8 0 0.0% 0.0% (n=118) Fish 2 0 0.0% 0.0% Turtle 70 27 22.9% 38.6% Snake 3 1 0.8% 33.3% Bird 4 2 1.7% 50.0% Opossums 62 28 23.7% 45.2% Armadillo 3 3 2.5% 100.0% Bats 12 0 0.0% 0.0% Dogs 39 6 5.1% 15.4% Artiodactyls 67 11 9.3% 16.4% Peccary 5 5 4.2% 100.0% Deer 56 7 5.9% 12.5% White-tailed deer 19 10 8.5% 52.6% Brocket deer 35 0 0.0% 0.0% Rodent 190 18 15.3% 9.5% Agouti and pacas 87 22 18.6% 25.3% Total: 547 118 100.0% Right Crustaceans 8 8 4.4% 100.0% (n=181) Fish 2 2 1.1% 100.0% Turtle 70 41 22.7% 58.6% Snake 3 2 1.1% 66.7% Bird 4 1 0.6% 25.0% Opossums 62 33 18.2% 53.2% Armadillo 3 0 0.0% 0.0% Bats 12 12 6.6% 100.0% Dogs 39 11 6.1% 28.2% Artiodactyls 67 15 8.3% 22.4% Peccary 5 0 0.0% 0.0% Deer 56 13 7.2% 23.2% White-tailed deer 19 7 3.9% 36.8% Brocket deer 35 1 0.6% 2.9% Rodent 190 4 2.2% 2.1% Agouti and pacas 87 52 28.7% 59.8% Total: 547 181 100.0% 189

PAGE 190

Table 7-10. Separation of Space Summaries for No rth versus South Directions at Cueva de Sangre, Guatemala. Region of the cave (n=295) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP North Crabs 6 3 1.3% 50.0% (n=238) Turtle 8 5 2.1% 62.5% Bird 26 22 9.2% 84.6% Domestic turkey 14 12 5.0% 85.7% Opossums 10 7 2.9% 70.0% Armadillo 6 2 0.8% 33.3% Bats 6 3 1.3% 50.0% Primates 2 2 0.8% 100.0% Dogs 4 4 1.7% 100.0% Cats 3 3 1.3% 100.0% Raccoons 3 3 1.3% 100.0% Bairds tapir 11 8 3.4% 72.7% Artiodactyls 163 142 59.7% 87.1% Peccary 25 17 7.1% 68.0% Deer 138 125 52.5% 90.6% White-tailed deer 114 105 44.1% 92.1% Brocket deer 16 16 6.7% 100.0% Rodent 20 17 7.1% 85.0% Agouti and paca 25 16 6.7% 64.0% Rabbit 2 1 0.4% 50.0% Total: 295 238 100.0% South Crabs 6 3 5.3% 50.0% (n=57) Turtle 8 3 5.3% 37.5% Bird 26 4 7.0% 15.4% Domestic turkey 14 2 3.5% 14.3% Opossums 10 3 5.3% 30.0% Armadillo 6 4 7.0% 66.7% Bats 6 3 5.3% 50.0% Primates 2 0 0.0% 0.0% Dogs 4 0 0.0% 0.0% Cats 3 0 0.0% 0.0% Raccoons 3 0 0.0% 0.0% Bairds tapir 11 3 5.3% 27.3% Artiodactyls 163 21 36.8% 12.9% Peccary 25 8 14.0% 32.0% Deer 138 13 22.8% 9.4% White-tailed deer 114 9 15.8% 7.9% Brocket deer 16 0 0.0% 0.0% Rodent 20 3 5.3% 15.0% Agouti and paca 25 9 15.8% 36.0% Rabbit 2 1 1.8% 50.0% Total: 295 57 100.0% 190

PAGE 191

191 Table 7-11. Separation of Space Summaries for East versus West Directions at Cueva de Sangre, Guatemala. Region of the cave (n= 547) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP East Crustaceans 8 0 0.0% 0.0% (n=244) Fish 2 0 0.0% 0.0% Turtle 70 0 0.0% 0.0% Snake 3 0 0.0% 0.0% Bird 4 1 0.4% 25.0% Opossums 62 1 0.4% 1.6% Armadillo 3 1 0.4% 33.3% Bats 12 3 1.2% 25.0% Dogs 39 16 6.6% 41.0% Artiodactyls 67 41 16.8% 61.2% Peccary 5 3 1.2% 60.0% Deer 56 36 14.8% 64.3% White-tailed deer 19 2 0.8% 10.5% Brocket deer 35 34 13.9% 97.1% Rodent 190 168 68.9% 88.4% Agouti and pacas 87 13 5.3% 14.9% Total: 547 244 100.0% West Crustaceans 8 8 2.6% 100.0% (n=303) Fish 2 2 0.7% 100.0% Turtle 70 70 23.1% 100.0% Snake 3 3 1.0% 100.0% Bird 4 3 1.0% 75.0% Opossums 62 61 20.1% 98.4% Armadillo 3 2 0.7% 66.7% Bats 12 9 3.0% 75.0% Dogs 39 23 7.6% 59.0% Artiodactyls 67 26 8.6% 38.8% Peccary 5 2 0.7% 40.0% Deer 56 20 6.6% 35.7% White-tailed deer 19 17 5.6% 89.5% Brocket deer 35 1 0.3% 2.9% Rodent 190 22 7.3% 11.6% Agouti and pacas 87 74 24.4% 85.1% Total: 547 303 61.7%

PAGE 192

Table 7-12. Summary of NISP, % Taxa, Left Sided Elements, Body portions, % NISP Burned, Spatial Auto Correlation and Cokriging at Cueva de Sangre, Guatemala. Taxonomic Group NISP (n=545) % Taxa % of Left sided elements Body Portion comment % NISP Burned Spatial Autocorrelation Cokriging Crustaceans 8 1.5% 2/ 5 = 40% 8 claws None Didnt run Didnt run Fish 2 0.4% N/A 1, 50%vertebrae, 1, 50% postcranial None Didnt run Didnt run Turtle 68 12.5% 6/ 11 = 54.5% 69, 98.6% carapace/plastron, 1, 1.4% vertebra 4/68 = 5.9% blackened Random, neither clustered nor dispersed R/L high conc near mouth of cave Snake 3 0.6% N/A 3 vertebrae None Didnt run Didnt run Bird 4 0.7% 1/2 = 50% 1 vertebrae, 1 humerus, 1 ulna, 1 UID None Didnt run Didnt run Opossums 62 11.4% 16/32 = 50% 28, 45.2% vertebrae, 5, 8.1% cranium, 11, 17.7% teeth, 8, 12.9% mandible, 1, 1.6% scapula, 5, 8.1% humerus, 2, 3.2% radius, 1, 1.6% ulna, 1, 1.6% astragulus None Clustering, 5-10% chance it was random Didnt run Armadillo 3 0.6% N/A 2, 66.7% scutes, 1, 33.3% caudal vert None Didnt run Didnt run Bats 12 2.2% 3/5 = 60% 4, 33.3% humerus, 4, 33.3% ulnas, 1, 8.3% femurs, 3,25% phalanx None Didnt run Didnt run Dogs 39 7.2% 17/32 = 53.1% 37, 94.8% teeth, 1,2.6% cranium, 1, 2.6% baculum None Clustering, random chance Didnt run Artiodactyls 67 12.3% 25/51 = 49% 40, 59.7% cranial, 1, 1.5% axial, 6, 9% frontlimb, 7, 10.4% hindlimb, 13, 19.4% distal None Random, neither clustered nor dispersed L/R, N/S, E/W high conc CS7-31 and CS11-1-1 Peccary 5 0.9% 1/3 = 33.3% 3, 60% teeth, 1, 20% mandible, 1, 20% metacarpal None Didnt run Didnt run Deer 56 10.3% 24/48 = 50% 36, 64.3% cranial, 1,1.8% axial, 4, 7.1% frontlimb, 7, 12.5% hindlimb, 8, 14.3% distal None Clustering, random chance. L/R, N/S, E/W high conc CS7-31 and CS11-1-1 192

PAGE 193

193 Table 7-12 Continued. Taxonomic Group NISP (n=545) % Taxa % of Left sided elements Body Portion comment % NISP Burned Spatial Autocorrelation Cokriging Whitetailed deer 19 3.5% 4/11 = 36.4% 1, 5.3% cranial, 1, 5.3% axial, 4, 21.1% frontlimb, 6, 31.6% hindlimb, 7, 36.8% distal None Random, neither clustered nor dispersed N/S, E/W high conc CS1, L/R high conc mouth of CS1 Brocket deer 35 6.4% 18/35 = 51.4% 34, 97.1% cranial, 1, 2.9% hindlimb None Random, neither clustered nor dispersed Didnt run Rodent 190 34.9% 79/124 = 63.7% 54, 28.4% cranial, 55, 28.9% axial, 28, 14.7% front limbs, 40, 21.1% hindlimbs, 13, 6.8% distal None Random, neither clustered nor dispersed L/R, N/S, E/W high conc CS6-81 and CS6-3-1 Agouti and pacas 87 16.0% 39/80 = 48.8% 61, 70.1% cranial, 1, 1.1% innominate, 15, 17.2% frontlimb, 6, 6.9% hindlimb, 4, 4.6% distal None Random, neither clustered nor dispersed L/R, N/S, E/W high conc lower region of CS1 (CS1-84-1)

PAGE 194

Table 7-13. Separation of Space Summaries for Left versus Right Sides at Naj Tunich, Guatemala. Region of the cave (n=295) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP Left Crabs 6 3 1.5% 50.0% (n=204) Turtle 8 5 2.5% 62.5% Bird 26 19 9.3% 73.1% Domestic turkey 14 11 5.4% 78.6% Opossums 10 7 3.4% 70.0% Armadillo 6 2 1.0% 33.3% Bats 6 2 1.0% 33.3% Primates 2 2 1.0% 100.0% Dogs 4 4 2.0% 100.0% Cats 3 3 1.5% 100.0% Raccoons 3 3 1.5% 100.0% Bairds tapir 11 7 3.4% 63.6% Artiodactyls 163 115 56.4% 70.6% Peccary 25 17 8.3% 68.0% Deer 138 98 48.0% 71.0% White-tailed deer 114 82 40.2% 71.9% Brocket deer 16 13 6.4% 81.3% Rodent 20 17 8.3% 85.0% Agouti and paca 25 14 6.9% 56.0% Rabbit 2 1 0.5% 50.0% Total: 295 204 100.0% Right Crabs 6 3 3.3% 50.0% (n=91) Turtle 8 3 3.3% 37.5% Bird 26 7 7.7% 26.9% Domestic turkey 14 3 3.3% 21.4% Opossums 10 3 3.3% 30.0% Armadillo 6 4 4.4% 66.7% Bats 6 4 4.4% 66.7% Primates 2 0 0.0% 0.0% Dogs 4 0 0.0% 0.0% Cats 3 0 0.0% 0.0% Raccoons 3 0 0.0% 0.0% Bairds tapir 11 4 4.4% 36.4% Artiodactyls 163 48 52.7% 29.4% Peccary 25 8 8.8% 32.0% Deer 138 40 44.0% 29.0% White-tailed deer 114 32 35.2% 28.1% Brocket deer 16 3 3.3% 18.8% Rodent 20 3 3.3% 15.0% Agouti and paca 25 11 12.1% 44.0% Rabbit 2 1 1.1% 50.0% Total: 295 91 100.0% 194

PAGE 195

Table 7-14. Separation of Space Summaries for No rth versus South Direct ions at Naj Tunich, Guatemala. Region of the cave (n=295) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP North Crabs 6 3 1.3% 50.0% (n=238) Turtle 8 5 2.1% 62.5% Bird 26 22 9.2% 84.6% Domestic turkey 14 12 5.0% 85.7% Opossums 10 7 2.9% 70.0% Armadillo 6 2 0.8% 33.3% Bats 6 3 1.3% 50.0% Primates 2 2 0.8% 100.0% Dogs 4 4 1.7% 100.0% Cats 3 3 1.3% 100.0% Raccoons 3 3 1.3% 100.0% Bairds tapir 11 8 3.4% 72.7% Artiodactyls 163 142 59.7% 87.1% Peccary 25 17 7.1% 68.0% Deer 138 125 52.5% 90.6% White-tailed deer 114 105 44.1% 92.1% Brocket deer 16 16 6.7% 100.0% Rodent 20 17 7.1% 85.0% Agouti and paca 25 16 6.7% 64.0% Rabbit 2 1 0.4% 50.0% Total: 295 238 100.0% South Crabs 6 3 5.3% 50.0% (n=57) Turtle 8 3 5.3% 37.5% Bird 26 4 7.0% 15.4% Domestic turkey 14 2 3.5% 14.3% Opossums 10 3 5.3% 30.0% Armadillo 6 4 7.0% 66.7% Bats 6 3 5.3% 50.0% Primates 2 0 0.0% 0.0% Dogs 4 0 0.0% 0.0% Cats 3 0 0.0% 0.0% Raccoons 3 0 0.0% 0.0% Bairds tapir 11 3 5.3% 27.3% Artiodactyls 163 21 36.8% 12.9% Peccary 25 8 14.0% 32.0% Deer 138 13 22.8% 9.4% White-tailed deer 114 9 15.8% 7.9% Brocket deer 16 0 0.0% 0.0% Rodent 20 3 5.3% 15.0% Agouti and paca 25 9 15.8% 36.0% Rabbit 2 1 1.8% 50.0% Total: 295 57 100.0% 195

PAGE 196

Table 7-15. Separation of Space Summaries for Ea st versus West Directions at Naj Tunich, Guatemala. Region of the cave (n=295) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP East Crabs 6 6 2.5% 100.0% (n=237) Turtle 8 8 3.4% 100.0% Bird 26 19 8.0% 73.1% Domestic turkey 14 7 3.0% 50.0% Opossums 10 9 3.8% 90.0% Armadillo 6 6 2.5% 100.0% Bats 6 6 2.5% 100.0% Primates 2 2 0.8% 100.0% Dogs 4 1 0.4% 25.0% Cats 3 2 0.8% 66.7% Raccoons 3 1 0.4% 33.3% Bairds tapir 11 10 4.2% 90.9% Artiodactyls 163 137 57.8% 84.0% Peccary 25 20 8.4% 80.0% Deer 138 117 49.4% 84.8% White-tailed deer 114 97 40.9% 85.1% Brocket deer 16 13 5.5% 81.3% Rodent 20 7 3.0% 35.0% Agouti and paca 25 22 9.3% 88.0% Rabbit 2 1 0.4% 50.0% Total: 295 237 100.0% West Crabs 6 0 0.0% 0.0% (n=58) Turtle 8 0 0.0% 0.0% Bird 26 7 12.1% 26.9% Domestic turkey 14 7 12.1% 50.0% Opossums 10 1 1.7% 10.0% Armadillo 6 0 0.0% 0.0% Bats 6 0 0.0% 0.0% Primates 2 0 0.0% 0.0% Dogs 4 3 5.2% 75.0% Cats 3 1 1.7% 33.3% Raccoons 3 2 3.4% 66.7% Bairds tapir 11 1 1.7% 9.1% Artiodactyls 163 26 44.8% 16.0% Peccary 25 5 8.6% 20.0% Deer 138 21 36.2% 15.2% White-tailed deer 114 17 29.3% 14.9% Brocket deer 16 3 5.2% 18.8% Rodent 20 13 22.4% 65.0% Agouti and paca 25 3 5.2% 12.0% Rabbit 2 1 1.7% 50.0% Total: 295 58 100.0% 196

PAGE 197

197 Table 7-16. Separation of Space Summaries for Light versus Dark Regions at Naj Tunich, Guatemala. Region of the cave (n=295) Gentaxa Total NISP NISP per Region % NISP per Region % NISP per Total NISP Light Crabs 6 0 0.0% 0.0% (n=58) Turtle 8 0 0.0% 0.0% Bird 26 5 8.6% 19.2% Domestic turkey 14 2 3.4% 14.3% Opossums 10 0 0.0% 0.0% Armadillo 6 0 0.0% 0.0% Bats 6 3 5.2% 50.0% Primates 2 2 3.4% 100.0% Dogs 4 1 1.7% 25.0% Cats 3 0 0.0% 0.0% Raccoons 3 0 0.0% 0.0% Bairds tapir 11 3 5.2% 27.3% Artiodactyls 163 26 44.8% 16.0% Peccary 25 9 15.5% 36.0% Deer 138 17 29.3% 12.3% White-tailed deer 114 14 24.1% 12.3% Brocket deer 16 0 0.0% 0.0% Rodent 20 14 24.1% 70.0% Agouti and paca 25 3 5.2% 12.0% Rabbit 2 1 1.7% 50.0% Total: 295 58 100.0% Dark Crabs 6 6 2.5% 100.0% (n=237) Turtle 8 8 3.4% 100.0% Bird 26 21 8.9% 80.8% Domestic turkey 14 12 5.1% 85.7% Opossums 10 10 4.2% 100.0% Armadillo 6 6 2.5% 100.0% Bats 6 3 1.3% 50.0% Primates 2 0 0.0% 0.0% Dogs 4 3 1.3% 75.0% Cats 3 3 1.3% 100.0% Raccoons 3 3 1.3% 100.0% Bairds tapir 11 8 3.4% 72.7% Artiodactyls 163 137 57.8% 84.0% Peccary 25 16 6.8% 64.0% Deer 138 121 51.1% 87.7% White-tailed deer 114 100 42.2% 87.7% Brocket deer 16 16 6.8% 100.0% Rodent 20 6 2.5% 30.0% Agouti and paca 25 22 9.3% 88.0% Rabbit 2 1 0.4% 50.0% Total: 295 237 100.0%

PAGE 198

Table 7-17. Summary of NISP, % Taxa, Left Sided Elements, Body portions, % NISP Burned, Spatial Auto Correlation and Cokriging at Naj Tunich, Guatemala. Taxonomic Group NISP (n=295) % Taxa % of Left sided elements Body Portion comment % NISP Burned Spatial Autocorrelation Cokriging Crabs 6 2.0% N/A 6, 100% claws 1/6, 16.7% Clustering, random chance Didnt run Turtle 8 2.7% N/A 6, 75% carapace, 2, 25% long bones None Didnt run Didnt run Bird 26 8.8% 5/15 = 33.3% 1 Cranial, 4 coracoid, 1 keel, 4 humerus, 2 radius, 4 femur, 4 tibiotarsus, 1 phalanx, 5 UID 3/26, 11.5%, blackened Random, neither clustered nor dispersed concentrated around IV 3 Domestic turkey 14 4.7% 5/14 = 35.7% 3 coracoid, 2 humerus, 2 radius, 3 femur, 4 tibiotarsus 3/14,21.4%, blackened Random, neither clustered nor dispersed Didnt run Opossums 10 3.4% 4/9 = 44.4% 1, 10% maxilla, 4, 40% mandible, 1, 10% molar, 2, 20% scapula, 2, 20% humerus None Clustering, random chance Didnt run Armadillo 6 2.0% 3/5 = 60% 1, 16.7% cranium, 1, 16.7% humerus, 1, 16.7% ulna, 2, 33.3%femur, 1, 16.7% tibia None Didnt run Didnt run Bats 6 2.0% 2/4 = 50% 2, 33.3% cranium, 2, 33.3% radius, 1, 16.7% ulna, 1, 16.7% metapodial None Random, neither clustered nor dispersed Didnt run Primates 2 0.7% 2/2 = 100% humerus and tibia 1/2,50% blackened Didnt run Didnt run Dogs 4 1.4% 2/2 = 100% 1, 25% premolar, 1, 25% radius, 1, 25%ulna, 1, 25% metapodial None Didnt run Didnt run Cats 3 1.0% 1/3 = 33.3% 1, 33.3% radius, 1, 33.3% humerus, 1, 33.3% metapodial None Didnt run Didnt run Raccoons 3 1.0% 2/3 = 66.7% 1, 33.3% radius, 1, 33.3% scapula, 1, 33.3% femur None Didnt run Didnt run Bairds tapir 11 3.7% 2/2 = 100% 1, 9.1% atlas vertebra, 1, 9.1% coccyx, 1, 9.1% scapula, 1, 9.1% humerus, 2, 18.2% carpal/ tarsal, 3, 27.3% metapodial, 2, 18.2% long bone 1/11, 9.1% blackened Random, neither clustered nor dispersed Didnt run 198

PAGE 199

199 Table 7-17 Continued. Taxonomic Group NISP (n=295) % Taxa % of Left sided elements Body Portion comment % NISP Burned Spatial Autocorrelation Cokriging Artiodactyls 163 55.3% 60/118 = 50.8% 23, 14.1% cranial, 46, 28.2% axial, 9. 5.5% axial-pectorial, 4, 2.5% axial-pelvic, 33, 20.2% front limb, 41, 25.2% hind limb, 7, 4.3% distal 34/163, 20.9% Clustering, 5-10% chance it was random conc Lots 44 +18 Peccary 25 8.5% 3/11 = 27.3% cranial (n=5, 20%), axial (n=6, 24%), front limb (n=9, 36%), hind limb (n=4, 16%), metapodial (n=1, 4%) 2/25, 8% Random, neither clustered nor dispersed conc Lots 1, 16, 24, + 57 Deer 138 46.8% 40/78 = 51.3% 18, 13% cranial, 42, 30% axial, 7, 5.1% pectoral, 4, 2.9% pelvic, 24, 17.4% front limb, 37, 26.8% hind limb, 6, 4.3% distal 32/138, 23.2% Clustering, 5-10% chance it was random conc Lots 44 + 16 White-tailed deer 114 38.6% 25/62 = 14, 12.3% cranial, 51, 44.7% axial, 19, 16.7% front limb, 27, 23.7% hind limb, 3, 2.6% distal 20/114, 17.5% Clustering, random chance conc Lots 16, 18, and 42 Brocket deer 16 5.4% 3/11 = 27.3% 3, 18.8% cranial, 1, 6. 3% axial pectorial, 2, 12.5% front limb, 7, 43.8% hind limb, 3, 18.8% distal 4/16, 25% Random, neither clustered nor dispersed Didnt run Rodent 20 6.8% 6/17 = 35.3% 3, 15% mandible, 3, 15% cranium, 1, 5% innominate, 4, 20% humerus, 4, 20% femur, 5, 25% tibia None Random, neither clustered nor dispersed Didnt run Agouti and paca 25 8.5% 6/14 = 42.9% 11, 44% cranial, 4, 16% axial, 4, 16% front limb, 6, 24% hind limb 2/25, 8% Random, neither clustered nor dispersed L/R, N/S, E/W conc Lot 57 and l, Light/Dark Lots 57, 1, 16, and 42 Rabbit 2 0.7% 0/2 = 0% 1 tibia, 1 humerus None Didnt run Didnt run

PAGE 200

CHAPTER 8 CONCLUSIONS Trends The spatial patterning of faunal remains in anci ent Maya cave site was tested using both GIS and visual analysis of the remains. Work ing under the theoretical framework of cognitive archaeology, I attempted to correlate spatial dist ributions of animal remains in these ritual landscapes to models of the cognized Maya universe, and th ereby understand more fully the ancient Maya mind. I hypothesized that the dist ribution of faunal remains in the caves was caused by the ritual use of specifi c regions of the caves for certain activities associated with the symbolic meaning of those ritual spaces in the cognized universe or symbolic landscape of the Maya. The distinct placement of faunal remains as part of cave rituals could possibly be traced back to repeated ritual behaviors. However, do to the small sample sizes and limited distribution of remains there were few links identified to the ritual placement of faunal remains within the caves. The patterns that I identified by this analysis included the use of th e north/east/light areas of the cave for crab deposition, possibly linked to the symbolic associat ion of crabs and water within the caves. The large amount of unburned turt le shell remains within the site may be an indication of these remains being used as instru ments or offerings of instruments within the caves. Peccary remains were also found to contain more limb than cranial elements in the cave sites. This may have been due to the fact that carvings of peccary skulls were highly valued at above ground elite residences. A hi gher concentration of deer limbs, or haunches, was identified at three out of the five sites and maybe an indication of ritual offe rings of these ve ry sacred parts of the body. Squirrel remains were only identified in one cave site. However, their placement 200

PAGE 201

within the eastern part of the cave mirrors the codices and ethnographi c accounts for squirrel offerings during the Muluc New Years ceremonies (Love 1986; Tozzer 1941). No patterning was found for ray-finned fi shes, amphibians, snakes, birds, opossums, armadillos, primates, tapir, raccoons, rodents, agoutis and pacas, and cottontail rabbits. Snakes, amphibians, raccoons, rodents, agouti and pacas, and cottontail rabbits may represent inhabitants of the caves or may have been not used by the an cient Maya for ritual o fferings. Since the bird remains were identified to the class of Aves and not to individual families, patterning for those birds not normally living in caves could not be defined as Brady (1989) was able to accomplish within Naj Tunich. Opossums were located main ly in the light and open regions of the cave which is where they would naturally occur. Ar madillos identified at the cave sites might represent natural inclusions, since armadillos are known inhabitants and burrowers of soils in or around the cave sites in this study. The low number of primates and tapirs also seem strange considering their high prevalence in the fo rested habitats of all of the caves. Some expected species were missing, suggestin g another form of patterning. There was a lack of lizard, or iguana, remain s at all the cave sites which may be due to the use of iguana breads or tamales as offerings (Bricker 1991; Taube 1988; Thompson 1972). The cooking of these bones may have lessened their chance for pr eserving within the cave sites. Bat remains were only found at Cueva de El Duende in large quanitities which would ha ve been expected at all cave sites. The finding of 7 left sided remains at Stela Cave in the Chamber 1 is also a strange anomaly considering bats are known to live in Cham ber 3, located in the dark and restricted part of the cave. The lack of canid remains at many of the cave sites is a surprising finding, since dogs are well known ritual offerings to the ancien t Maya. There were also a low number of felid 201

PAGE 202

remains, which would have been expected to be in larger numbers since caves have been identified at the main place for the acce ssion rituals for Maya royalty (Brady 1989). Limitations The patterning revealed in this study was not easily quantified, nor was it statistically verifiable. There are four major reasons for the inability to clearly quantify or verify specific patterns for the distribution of animal remains within these five Maya cave sites. The first limitation on the analysis of the faunal remains at these sites was the mixture of both a small sample size for most of the taxonomic groups and also the limited distribution of the excavation units across the cave surfaces. GIS was us ed to help model and analyze the patterns that may have formed from the spatial relations hip of remains with these cave sites. However, the small samples for many of the taxonomic gro ups did not allow spatial autocorrelation to process the findings. There were i ssues with the results from thos e that could be processed since many of the samples were identified as random despite a visual a ssessment of patterning, suggesting that sample size was again impacti ng the GIS results. Cokriging was another GIS spatial analysis tool that could not be used at all in three site s and provided results only for rare groups for the other caves, again because of limited distributions at the other caves. A more complete collection pr ocess and robust sampling strate gy would have helped with the overall distribution of faunal remains; how ever, even these improvements may not have increased the number of faunal remains recove red from each site. In addition, many of the samples collected were from surface collections. This is because there were rarely deep contexts to excavate at many of these cave sites. The l ack of stratigraphic sa mples does not allow for specific chronological samples to be analyzed and the remains might represent multiple occupation periods. This is problematic because ritual practice and beliefs were undoubtedly not stagnant over the entire 2000 y ears of Maya occupation and use of the region. All of the sites 202

PAGE 203

date to part of the Preclassic pe riod, but most of the artifacts, cer amics, were from the Early to Late Classic at the sites. Finally, differential excavation and faunal mate rial recovery methods caused variations between sites. At the site of Cueva de Sangr e there were a few assemblages recovered by the deflocculation of 1 meter by 1 meter soil sample s. These samples produced the largest amounts of faunal remains at the site of CS and most of the findings were from these large soil samples. The findings at Naj Tunich were from surface co llections. The other sites, CBR, and STC were excavated using screens. For both CS and CD I am unsure of the specific recovery methods used; although I would expect screening wa s used. Screening and collection methods would impact many of the small animal findings, in cluding rodent and bat remains. The collection methods may be one of the reasons for the lack of bat remains at all of the sites except for CD. The second limitations on the results from th is study are due to th e taphonomic processes that occur in cave sites. Some taphonomic proc esses that may have caused the movement of remains within the caves are seasonal flooding and runoff from these flooding events. The caves with seasonal flooding and run-off include Caves Branch Rockshelter, Cueva de El Duende, and Cueva de Sangre. Stela Cave was identified as being relatively dry; however, this site was excavated before the rainy season and may in f act be affected by run-off. There were also indications at CBR that taphonomic disturbances may have been caused by the burrowing of armadillos at the site (Wrobel and Taylor 2006). Research has found that artifacts can be highly impacted by the movement of armadillo burrowing (Araujo and Marcelino 2003). Other taphonomic disturbances identified at most of the sites, except for Cueva de Sangre and Naj Tunich was looting. There were looters pits iden tified in Caves Branch Rockshelter which were 203

PAGE 204

found during the excavation of this site and areas of human-induced disturbances were found in Stela Cave and Cueva de El Duende. Other taphonomic processes that may have a ffected the deposition of animal remains within these cave sites are the cultural filters of animal remains over time. The Maya may have used the caves for areas of sacrificial offe rings; however, the consumption and subsequent deposition of these remains may have been outside of the cave. The offering of specific species within the caves may have occurred but their consumption in other areas would make these events untraceable in the archaeological recor d. Therefore, the patterning and use of specific animals with directional associations, sidedness, or light and dark regions of the cave could be lost in the faunal assemblages. The same can also be said of the cave inhabitants and the distribution of their remains which may have been more culturally than naturally dispersed. The types of offerings depicted in the Maya codice s of faunal remains are usually in the form of animal breads or tamales. The cooking of the hard parts of faunal remains will breakdown or disintegrates after these remains are cooked. The third source of bias in the analysis of the faunal remains fr om these five cave sites may be due to the different shapes of these caves. Caves Branch Rockshelter was a large rockshelter area with a small cave. The rockshelter contained a well known drip zone that may have caused runoff at the site. Cueva de Sangre was a 3.5 km multiple tunnel system that did not allow for easy separations of spaces within the cave. Naj Tunich was the largest ca ve in this study, and Operation IV was located at the mouth of this cave system, which would have left the site open to the elements and natural depos its of animal remains at the site. Stela Cave and Cueva de El Duende were both relatively small in size and c ontained large open chambers with smaller back chambers, therefore over representing material s from the light and open spaces. All of these 204

PAGE 205

varying sized caves were split into different sp atial categories based on the availability of space and the placement of excavat ion units at these sites. The final difficulty with this analysis was th e differential use of these cave sites. This analysis relies on the assumption that all the cav es would have been used for the same ritual purposes and therefore that the spatial distri butions would be consistent between caves. However, this was not likely the case and this study highlights the difficulty of comparing different areas even within a broad ritual s ubgroup. CBR is a rockshelter that served as an ancient Maya cemetery. The soil within this site was mixed grave fill and the site has been identified as a non-elite burial gr ound that may have been closel y associated with a few mounds located in front of the rockshelter. Stela Cave contains a chamber with modified architectural material, including a possible stela or altar, and therefore may have se rved a function as a nonelite area for cave rituals. Cueva de Sangre and Cu eva de El Duende are both located at the large site of Dos Pilas. These two caves had limited access and have been de scribed as elite cave contexts. Finally, the site of Naj Tunich, the largest cave site known in the Maya region, has been identified as an elite pilgrimage site because of the multitude of cave painting and hieroglyphics at the site. As archaeologists we hope that the rituals a nd cognitive worldviews of ancient people are stable, but in reality, they are a dynamic part of a people and the changes in ritual behaviors are large scale and vary over time. While the variations of sample size, excavat ion strategy, and recovery methods may have limited the ability of this analys is to discern specific patterni ng. The different uses and spatial layout between all of these caves may actually be the best indicators to understand the varied distribution of faunal remains at a ll of these sites. It may be that different caves were used in different ways and for different types of rituals, perhaps all associated with the Maya universe, 205

PAGE 206

but likely all representing that universe in diffe rent ways. The role in determining the spatial distributions may be affect by such differences as the status of the group using the site, the time period during which it was used, and even the season of the year it was used. Recommendations This research project provided the framew ork for conducting a futu re study involving the spatial distribution of faunal and other artifactual materials in both cave settings and also surface sites. The use of zooarchaeology and GIS in combination allows fo r both the analysis of spatial patterns and production of georefer enced databases with informati on that can be added to as research continues in the Maya region. In a future study there are recommendations I have for preparing a project for the analys is of faunal remains and GIS. Th e inclusion of GIS at the onset of a research project would help in the digitizing of cave site maps and would allow for a more three-dimensional perspective. With that said, th e use of a one meter by one grid system, like the one used at CBR, allows for a more precise collection method. Collection methodologies in cave site should be expanded to take the spatial distribution of remains into account. The expansion of collection areas may also increase the amount animal remains identified at the sites. Summary To the Maya, layering and direc tionality played an important part of their worldview. The layering of caches by the Maya may have repres ented their world and parts of their defined universe. The directional placement of buildings within sites, ceremonial events at specific directions, and the placement of these caches in specific directi ons all reinforced the ancient Maya world. The Maya worldview has been define d in relation to sidedne ss, directionality, and the use and access to space. Caves represented a microcosm for these worldviews and with their known associations with ritual behavior can serve as a model for understanding ancient Maya rituals. Animals and animal remains have always played a role in these rituals. Using GIS and 206

PAGE 207

207 faunal remains in caves was attempted to identif y and understand the ritual use of animal in relation to the relationships of sidedness, directionality, and acce ss to space. Hopefully future research studies are able to incorporate and identify the pattern of faunal remains in cave so that we can further understand the ancient Maya mind.

PAGE 208

LIST OF REFERENCES Anderson, Elyse 2009 Exploring Maya Ritual Fauna: Caves and the Proposed Link with Contemporary Hunting Ceremonialism MA thesis. University of Florida. Anschuetz, Kurt F., Wilshusen, Rich ard H. and Cherie L. Scheick 2001 An Archaeology of Landscapes: Perspectives and Directions. Journal of Archaeological Research 9(2):157-211. Araujo, Astolfo G. Mello and Jose Carlos Marcelino 2003 The Role of Armadillos in the Movement of Archaeological Materials: An Experimental Approach. Geoarchaeology. 18(4): 433. Arata, Sean T. 2008 Isla Cilvituk: Finding Primary Contexts using GIS Anthropology Department. Las Cruces, New Mexico, New Mexi co State University: 244. Ashmore,Wendy 1989 Construction and cosmology: Politics and id eology in lowland Maya settlement patterns. In Word and Image in Maya Culture: Explorations in Language, Writing and Representation edited by Hanks, W. F., and Rice, D. S., pp. 272. University of Utah Press, Salt Lake City. 1991 Site-planning principles and concepts of directionality among the ancient Maya. Latin American Antiquity 2: 199. 2002 Decisions and dispositions: Socializing spa tial archaeology. American Anthropologist 104: 1172. Ashmore, Wendy and Jeremy A. Sabloff 2002 Spatial orders in Maya civic plans. Latin American Antiquity 13: 201. Awe, Jaime J. 1998 The Western Belize Regional Cave Project: Objectives, Context, and Problem Orientation. In The Western Belize Regional Cave Project: A Report of the 1997 Field Season, edited by J. J. Awe, pp. 1-22, Department of Anthropology, Occasional Paper No. 1, University of New Hampshire, Durham. Awe, Jaime J., Sherry Gibbs, and Cameron Griffith 2005 Stelae and Megalithic Monuments in the Caves of Western Belize. In In the Maw of the Earth Monster: Mesoamerican Ritual Cave Use edited by James E. Brady and Keith M. Prufer, pp. 223-248, University of Texas Press, Austin. Baker, Mary 1992 Capuchin Monkeys ( Cebus capucinus ) and the Ancient Maya. Ancient Mesoamerica 3:219-228. Beaubien, Harriet F. 208

PAGE 209

2004 Excavation and Recovery of a Funera ry Offering of Marine Materials from Copan. In Maya Zooarchaeology: New Directions in Method and Theory edited by K. F. Emery. Institute of Archaeology, UCLA Press, Los Angeles. Belli, Francisco E. 1999 The archaeology of complex societies in s outheastern Pacific coastal Guatemala: a regional GIS approach Oxford: British Archaeological Reports. Bender, Barbara 2002 Landscape. Encyclopedia of Social and Cultural Anthropology edited by A. Barnard and J. Spencer, pp. 323-324. New York: Routledge. Bill, Cassandra R., Hernndez, Chris tine L. and Victoria R. Bricker 2000 The Relationship Between Early Colonial Maya New Years Ceremonies and Some Almanacs in the Madrid Codex. Ancient Mesoamerica 11:149. Binford, Lewis R. 1962 Archaeology is Anthropology. American Antiquity 28(2):217-225. Bonor Villarejo, Juan Luis 1998 Caves Branch Caves: Archaeological Field Report. FAMSI website. 8 December 2008 < http://www.famsi.org/reports/96044/>. 2003 Caves Branch Caves, Distrito de El Ca yo, Belice. Reporte Arqueolgico de Campo. FAMSI website. 8 December 2008 < http://www.famsi.org/reports/96044es/ >. Brady, James E. 1989 Investigation of Maya Ritual Cave Use with Special Reference to Naj Tunich, Peten, Guatemala PhD Dissertation, Department of Anthropology. Los Angeles, University of California. 1990 Investigaciones en la Cueva de la Sangre y Otras Cuevas de la Regin de Petexbatun. In: Demarest AA, Houston SD, editors. Pr oyecto Arqueolgico Regional Petexbatun. Informe Preliminar #2, Segunda Temporada 1990. Nashville, TN: Vanderbilt University Press. p 438-490. 1991 Caves and Cosmovision at Utatlan. California Anthropologist 18(1):1. 1997 Settlement Configuration and Cosmology: The Role of Caves at Dos Pilas. American Anthropologist New Series, 99(3):602-618. 2000 Early Political Appropriation of the Sacred Landscape. In The Sacred and the Profane: Architecture and Identity in the Maya Lowlands edited by Pierre Robert Colas, Kai Delvendahl, Marcus Kuhnert, Annette Schubart, pp. 129-136. Acta Mesoamericana 10. 2003 In My Hill, In My Valley: The Importance of Place in Ancient Maya Ritual. In Mesas and Cosmology in Mesoamerica, edited by Douglas Sharon, pp. 83-91. San Diego Museum Papers 42. San Diego Museum of Man, San Diego. 2005 The Impact of Ritual on Ancient Maya Economy. In Stone Houses and Earth Lords: Maya Religion in the Cave Context K. M. Prufer and J. E. Brady, pp. 115-134. Boulder, University Press of Colorado. 209

PAGE 210

Brady, James E. and Wendy Ashmore 1999 Mountains, Caves, Water: Ancien t Maya ideational landscapes. In Archaeologies of Landscapes: Contemporary Perspectives edited by Ashmore,W., and Knapp, A. B., pp. 124. Malden, Mass.: Black well Publishers. Brady, James E., Luin, Fernando, Wright, Lori, Foncea, Carol, and Sandra Villagran De Brady 1992 Descubrimientos recientes en la Cueva de Sangre de Dos Pilas. In Simposio de Investigaciones Arqueolgicas en Guatemala (4 session, 1990) pp. 153-168. Ministerio de Cultura y Deportes; Instituto de Antr opologa y Etnologa e Historia., Guatemala Brady, James E., Luin, Fernando, Foncea, Carol, Wright, Lori, and Sandra Villagran De Brady 1990 Investigaciones en la Cueva de Sangre y otras cuevas de la regin de Petexbatn. In El Proyecto Arqueolgico Regional Pet exbatn: Informe Preliminar 2 Edited by Arthur A. Demarest and Stephen D. Houston, eds. pp. 438-566. Instituto Nacional de Antropologia e Historia de Guatemala, Guatemala Brady James E. and Keith M. Prufer 2005a In The Maw of the Earth Monster: Mesoamerican Ritual Cave Use. The Linda Schele Series in Maya and Pre-Columbian Studies University of Texas Press, Austin. 2005b Introduction: A History of Mesoam erican Cave Interpretations. In In the Maw of the Earth Monster: Mesoamerican Ritual Cave Use edited by J. E. Brady and K. M. Prufer, pp. 1-17. Austin, University of Texas Press. Brady, James E. and Irma Rodas 1992 Hallazgos recientes y nuevas interpretaciones de la Cueva de El Duende, Petn. En V Simposio de Investigaciones Ar queolgicas en Guatemala, 1991 (editado por J.P. Laporte, H. Escobedo y S. Brady), pp.169176. Museo Nacional de Arqueologa y Etnologa, Guatemala. Brady, James E., Rodas, Irma, Wright, Lori, Em ery, Katherine, Lopez, Nora Maria, Stiver, Laura, and Robert Chatham 1991 Proyecto Arqueologico regiona l de Cueva Petetabatun. In El Proyecto Arqueolgico Regional Petexbatn: Informe Preliminar 3 Edited by Arthur A. Demarest, Takeshi Inomata, Hector Escobedo, and Joel Pa lka, pp. 652-748. Instituto Nacional de Antropologia e Historia de Guatemala, Guatemala Brady, James E. and Ann Scott 1997 Excavation of Buried Cave Deposits: Implications for Interpretations. Journal of Cave and Karst Studies. 59(1): 15-21. Brady, James E., Scott, Ann Cobb, Allan Rodas, lrma Fogarty, John, and Monica Urquizu Sanchez 1997 Glimpses of the Dark Side of the Pe texbatun Project: The Petexbatun Regional Cave Survey. Ancient Mesoamerica 8:353-364. Bricker, Victoria R. 210

PAGE 211

1983 Directional Glyphs in Maya Inscriptions and Codices. American Antiquity 48(2):347353. 1991 Faunal Offerings in the Dresden Codex. In Sixth Palenque Round Table, 1986 edited by Merle Greene Robertson and Vi rginia M. Fields, pp. 285-292. University of Oklahoma Press, Norman. 1997 The Structure of Almanacs in the Madrid Codex. In Papers on the Madrid Codex edited by Victoria R. Bricker and Ga brielle Vail, pp. 1-25. Middle American Research Institute, Pub. 64. Tulane University, New Orleans. Brown, Linda A. 2000 From Discarded to Divina tion: Demarcating the Sacred through the Collection and Curation of Discarded Objects. Latin American Antiquity 11(4): 319-334. 2004 Dangerous Places and Wild Spaces: Creating Meaning with Materials and Space at Contemporary Maya Shrines on El Duende Mountain. Journal of Archaeological Method and Theory. 11(1):31-58. 2005 Planting the Bones: Hunting Ceremonialis m at Contemporary and Nineteenth-Century Shrines in the Guatemalan Highlands. Latin American Antiquity 16(2):131-146. Brown, Linda A. and Kitty Emery 2008 Negotiations with the Animate Forest: Hun ting shrines in the Guatemalan Highlands. Journal of Archaeological Method and Theory 15:300-337. Bruchez, Margaret Sabom 2007 Artifacts that speak for themselves : Sounds under afoot in Mesoamerica. Journal of Anthropological Archaeology 26:47-64. Chase, Diane Z., and Arlen F. Chase 1998 The Architectural Context of Caches, Burials, and Other Ritual Activ ities for the Classic Period Maya (as Reflected at Caracol, Belize). In Function and meaning in classic Maya architecture: a symposium at Dumb arton Oaks, 7th and 8th October 1994 edited by Stephen D. Houston, pp. 299-332. Washington, DC: Dumbarton Oaks Research Library and Collection Coggins, Clemency C. 1980 The Shape of Time: Some Political Implications of a Four-Part Figure. American Antiquity 45(4):727. Ebert, David 2004 Applications in archeological GIS, Canadian Journal of Archaeology, 28, 219. Emery, Kitty F. 1997 The Maya Collapse: A Zooarchaeological Investigation PhD dissertation, Anthropology, Cornell University. 656 pp. 2003 The Noble Beast: Status and Differentia l Access to Animals in the Maya World. World Archaeology 34(3): 498-515. 211

PAGE 212

2004a Maya Zooarchaeology: Historical Perspe ctives on Current Research Directions. In Maya Zooarchaeology: New Direct ions in Method and Theory edited by F. Emery, pp. 1-12. Los Angeles, Cotsen Institute of Ar chaeology, University of Los Angeles. 2004b Maya Zooarchaeology: In Pursuit of Social Variability and Environmental Heterogeneity. In Continuities and Changes in Maya Archae ology: Perspective at the millennium edited by C.W. Golden and G. Borgstede, pp. 309-333. New York: Routledge. 2004c Animals from the Maya Underworld: Recons tructing Elite Maya Ritu al at the Cueva de los Quetzales, Guatemala. Behaviour Behind Bones: The Zooarchaeology of Ritual, Religion, Status, and Identity edited by S. J. O'Day, W. Van Neer and A. Ervynck, pp. 101-113. Oxford, Oxbow Books. 2004d Tropical Zooarchaeology: An Introd uction to the Themes and Issues. Archaeofauna .13: 7-10. 2004e Introduction to Maya Zooarchaeology, In Maya Zooarchaeology: New Directions in Method and Theory edited by K.F. Emery, pp. xi-xiii. Los Angeles, CA: Institute of Archaeology, UCLA Press 2004f Maya Zooarchaeology: New Directions in Method and Theory Edited volume. Cotsen Institute of Archaeology, Monograph 51. Los Angeles, CA. Flannery, Kent V. and Joyce Marcus 1993 Cognitive Archaeology. Cambridge Archaeological Journal. 3(2):260-270. 1996 Cognitive archaeology. In Contemporary Archaeology in Theory: A Reader edited by R.W. Preucel, I. Hodder, pp. 350-63. Oxford: Blackwell. 1998 Cognitive archaeology. In Reader in Archaeological Theory: Post-processual and cognitive approaches edited by D.S. Whitley, pp. 35-48. New York: Routledge. Fogelin, Lars 2007 The Archaeology of Religious Ritual. Annual Review of Anthropology. 36:55-71. Garcia-Zambrano, Angel J. 1994 Early Colonial Evidence of Pre-Columbian Rituals of Foundation. In Seventh Palenque Round Table, 1989 edited by Merle Greene Robertson and Virginia M. Fields, pp. 217227. The Pre-Columbian Art Research Institute, San Francisco. Glassman, David M. and Juan Luis Bonor Villarejo 2005 Mortuary Practices of the Prehistoric Maya from Caves Branch Rock Shelter, Belize. Stone Houses and Earth Lords: Ma ya Religion in the Cave Context edited by Keith M. Prufer and James E. Brady, pp. 285-296. Univ ersity Press of Colorado, Boulder. Grayson, Donald K. 1984 Quantitative Zooarchaeology. New York: Academic Press. Griffith, Cameron and Chis tophe G. B. Helmke 2000 Preliminary Inverstigation and GIS Spatia l Analysis in Actun Halal, Belize. In The Western Belize Regional Cave Projec t: A Report of the 1999 Field Season edited by C. Griffith, R. Ishihara and J. J. Awe, pp. 99-114. Department of Anthropology, Occasional Paper No.3, Durham, New Hampshire, University of New Hampshire. 212

PAGE 213

Halperin, Christina T. 2005 Social Power and Sacred Space at Actun Nak Beh, Belize. In Stone Houses and Earth Lords: Maya Religion in the Cave Context edited by Keith M. Prufer and James E. Brady, pp. 71-90. University Pr ess of Colorado, Boulder. Hanks, William F. 1990 Referential Practice: Language and Lived Space among the Maya University of Chicago Press: Chicago. Healey, Paul F. 2007 The Anthropology of Mesoamerican Caves. Reviews in Anthropology 36(3): 245-278. Helmke, Christophe G.B., and Jaime J. Awe 1998 Preliminary Analysis of the Pictographs, Petroglyphs and Sculptures of Actun Uayazba Kab, Cayo District, Belize. In The Western Belize Regional Ca ve Project: A Report of the 1997 Field Season pp.145-204, Department of Anthropology, Occasional Paper No.1, University of New Hampshire, Durham. Heyden, Doris 1975 An Interpretation of the Cave underneath the Pyramid of the Sun in Teotihuacan, Mexico. American Antiquity 40(2):131-147. Hickman, Cleveland P. 1967 Biology of the Invertebrates C.V. Mosby Company: St. Louis. Ishihara, Reiko 2007 Bridging the chasm between reli gion and politics: archaeologic al investigations of the grietas at the Late Classic Maya site of Aguateca, Peten, Guatemala Doctoral dissertation, Department of Anthropology, University of California, Riverside. 2008 Rising clouds, blowing winds: Late Classic Maya rain ri tuals in the Main Chasm, Aguateca, Guatemala. World Archaeology. 40:2:169-189. Ishihara, Reiko and Cameron Griffith 2004 Archaeological Investigations and the Evalua tion of Sacred Spaces Construction in Stela Cave, Cayo District, Belize. The Belize Valley Archaeological Reconnaissance Project: A Report from the 2003 Field Season (eds C. M. Audet and J. J. Awe) pp. 56-75. Knapp, A. Bernard and Wendy Ashmore (eds) 1999 Archaeological Landscapes: Constructed, Conceptualized, Ideational. In Archaeologies of Landscape: Contemporary Perspectives edited by W. Ashmore and A.B. Knapp, pp. 130. Malden, Massachusettes: Blackwell Publishers. Kozelsky, Kristin Linda 2005 Identifying social drama in the Maya re gion: Fauna from the Lagartero Basurero, Chiapas, Mexico unpublished MA Thesis (Anthropol ogy), Florida State University, Tallahassee. 213

PAGE 214

Kunen, Julie L., Galindo,Mary Jo and Erin Chase 2002 Pits and Bones: Identifying Maya ritual behavior in the archaeological record. Ancient Mesoamerica 13:197. Kvamme, K. L., 1999 Recent directions and developments in Geographical Information Systems, Journal of Archaeological Research, 7:153. Marciniak, Arkadiusz 1999 Faunal Materials and Interpretive Ar chaeology Epistemology Reconsidered. Journal of Archaeological Method and Theory 6(4): 293-320. Marn Arroyo, Ana Belen 2009 Assessing What Lies beneath the Spatial Di stribution of a Zooarc haeological Record: The use of GIS and spatial correlations at El Miron Cave (Spain). Archaeometry 51(3):506-524. Mathews, Jennifer P. and James F. Garber 2004 Models of Cosmic Order: Physical expression of sacred space among the ancient Maya. Ancient Mesoamerica 15(1):49-59. McBee, Karen and Robert J. Baker 1982 Dasypus novemcinctus. Mammalian Species 162:1-9. McNatt, Logan 1996 Cave Archaeology of Belize. Journal of Cave and Karst Studies 58 (2):81-99. Minjares, Amador, Jr. 2003 Human Skeletal Remains from the Ancient Ma ya in the Caves of Dos Pilas, Guatemala. Masters Thesis. Texas A&M University Moholy-Nagy, Hattula 2004 Vertebrates in Tikal Burials and Caches. In Maya Zooarchaeology: New Directions in Method and Theory edited by F. Emery, pp. 193-205. Los Angeles, Cotsen Institute of Archaeology, University of Los Angeles. Moyes, Holley 2001 The Cave as a Cosmogram: The Use of GIS in an Intrasite Spatial Analysis of the Main Chamber of Actun Tunichil Muknal, A Maya Ceremonial Cave in Western Belize Masters Thesis, Florida Atlantic University. 2002 The Use of GIS in the Spatial Anal ysis of an Archaeological Cave Site. Journal of Cave and Karst Studies 64(1):9-16. 2005 Cluster Concentrations, Boundary Markers, and Ritual Pathways : A GIS Analysis of Artifact Cluster Patterns at Act un Tunichil Muknal, Belzie. In In The Maw of the Earth Monster: Mesoamerican Ritual Cave Use edited by J. E. Brady and K. M. Prufer, pp. 214

PAGE 215

269-300. The Linda Schele Series in Maya a nd Pre-Columbian Studies. University of Texas Press, Austin. Moyes, Holley and Jaime J. Awe 1998 Spatial Analysis of Artifact s in the Main Chamber of Ac tun Tunichil Muknal, Belize: Preliminary Results. In The Western Belize Regional Cave Project: A Report of the 1997 Field Season edited by J. J. Awe., pp. 23-36. Department of Anthropology, Occasional Paper No. 1, University of New Hampshire, Durham. Nardini, Alessandra and Frank Salvadori 2003 A GIS Platform Dedicated to the Pro duction of Models of Distribution of Archaeo(zoo)logical Remains. Archaeofauna 12:127-141. Palka, Joel W. 2002 Left/Right Symbolism and the Body in Ancient Maya Iconography and Culture. Latin American Antiquity 13(4):419-443. Pendergast, David M. 1969 The Prehistory of Actun Balam, British Honduras Art and Archaeology Occasional Paper No. 16. Royal Ontario Museum, Toronto. 1971 Excavations at Eduardo Quiroz Cave, British Honduras (Belize) Art and Archaeology Occasional Paper No. 21. Royal Ontario Museum, Toronto. 1974 Excavations at Actun Polbilche, Belize Royal Ontario Museum Monograph 1. Toronto. Peterson, Polly Ann 2006 Ancient Maya ritual cave us e in the Sibun Valley, Belize Doctoral dissertation, Department of Archaeology, Boston University, Boston, Massachusetts. Pohl, Mary D. 1981 Ritual Continuity and Transformation in Me soamerica: Reconstructing the Ancient Maya Cuch Ritual. American Antiquity 46(3): 513-529. 1983 Maya Ritual Faunas: Vertebrate Remains from Burials, Caches, Caves, and Cenotes in the Maya Lowlands. In Civilization in the ancient Americ as: essays in honor of Gordon R. Willey edited by R. M. Leventhal and A. L. Kolata, pp. 55-103. Albuquerque, N.M., University of New Mexico Press Preucel, Robert W. 2006 Archaeological Semiotics Oxford: Blackwell Publishing. Prufer Keith M. and James E. Brady 2005a Concluding Comments. In In the Maw of the Earth Monster: Mesoamerican Ritual Cave Use edited by J. E. Brady and K. M. Pruf er, pp. 403-411. Austin, University of Texas Press. 2005b Stone Houses and Earth Lords: Ma ya Religion in the Cave Context University Press of Colorado, Boulder. 215

PAGE 216

Pugh, Timothy W. 2001 Flood Reptiles, Serpents Temples, and the Quadripartite Universe : The Imago Mundi of Late Postclassic Mayapan. Ancient Mesoamerica 12:247. 2005 Caves and Artificial Caves in Late Postclassic Maya Ceremonial Groups. In Stone Houses and Earth Lords: Maya Religion in the Cave Context edited by Keith M. Prufer and James E. Brady, pp. 47-69. Universi ty Press of Colorado, Boulder. Reitz Elizabeth J. and Elizabeth S. Wing 1999 [2008] Zooarchaeology Cambridge: Cambridge University Press. Renfrew, Colin 1993 Cognitive Archaeology: Some Thoughts on the Archaeology of Thought. Cambridge Archaeological Journal 3(2):248-253. 1994 Towards a Cognitive Archaeology. In The Ancient Mind: Elements of Cognitive Archaeology edited by C. Renfrew and E. B.W. Zubrow pp. 3. Cambridge: Cambridge University Press. Renfrew, C. and Ezra B.W. Zubrow (eds) 1994 The Ancient Mind: Elements of Cognitive Archaeology Cambridge: Cambridge University Press. Saturno, William A., Taube, Karl A., and David Stuart 2005 The Murals of San Bartolo, El Peten, Guatemala, Part 1: The North Wall. Ancient America 7:1-56. Scott, Ann 1995 1993 Investigations in the Cueva de Sangre. In El Proyecto Arqueolgico Regional Petexbatn: Informe Preliminar 6 Edited by Juan A. Valds, Arthur A. Demarest and Hector Escobedo, eds. pp. 537-541. Instituto Nacional de Antropolog ia e Historia de Guatemala, Guatemala. Spenard, Jon 2006 The Gift in the Cave for the Gift of the Wo rld: An Economic Approach to Ancient Maya Cave Ritual in the San Francisco Hill -Caves, Cancuen Region, Guatemala Unpublished Masters Thesis. Department of Anthropol ogy, Florida State University, Tallahassee, Florida. Stanchly, Norbert 2004 Picks and Stones May Break My Bone s: Taphonomy and Maya Zooarchaeology In Maya Zooarchaeology: New Directions in Method and Theory edited by K. F. Emery, pp.35-43. Institute of Archaeology, UCLA Press, Los Angeles. Stone, Andrea 1989 The Painted Walls of Xibalba: Maya Cave Painting as Evidence of Cave Ritual. In Word and Image in Maya Culture: Explorati ons in Language, Writing and Representation 216

PAGE 217

edited by Hanks, W. F., and Rice, D. S., pp. 319335. University of Utah Press, Salt Lake City. 1995 Images from the Underworld: Naj Tunich and the tradition of Maya cave painting Austin: University of Texas Press. 1997 Regional Variation in Maya Cave Art. Journal of Cave and Karst Studies 59(1):33-42. 2005 A Cognitive Approach to Artifact Distri bution in Caves of the Maya Area. In In The Maw of the Earth Monster: Mes oamerican Ritual Cave Use, edited by J. E. Brady and K. M. Prufer, pp. 249-268. The Linda Schele Series in Maya and Pre-Columbian Studies. University of Texas Press, Austin. Taack, George H. 1973 An Iconographic Study of Deer Hunting Scen es in Maya Painting: Codex Madrid and Vases from Calcehtok and Altun Ha Masters thesis, University of New Mexico, Albuquerque. Taube, Karl 1988 The Ancient Yucatec New Year Festival: The Liminal Period in Maya Ritual and Cosmology. Ph.D. diss., Yale University. Teeter, Wendy G. 2004 Conclusions: Suggesting Solutions to the Issues of Tropical Zooarchaeology. Archaeofauna 13: 199-202. Teeter, Wendy G. and Arlen F. Chase 2004 Adding Flesh to Bones: Using Zooarchaeo logy Research to Answer the Big-Picture Questions. Archaeofauna 13: 155-172. Thompson, J. Eric S. 1972 A Commentary on the Dresden Codex: A Maya Hieroglyphic Book Memoirs of the American Philosophical Society, Vol. 93. Philadelphia. Tozzer, Alfred M. 1941 Landa's Relacin de las cosas de Yucatan Papers of the Peabody Museum of American Archaeology and Ethnology, Vol. 18. Harvard University, Cambridge. Trigger, Bruce G. 2006 A History of Archaeological Thought 2nd edition. Cambridge: Cambridge University Press. Vail, Gabrielle 1997 The Deer-Trapping Almanacs in the Madrid Codex. In Papers on the Madrid Codex, edited by Victoria R. Bricker and Gabrielle Vail, pp. 73-110. Middle American Research Institute, Pub. 64. Tulane University, New Orleans. von Nagy, Christopher 1997 Some Comments on the Madrid Deer-Hunting Almanacs. In Papers on the Madrid Codex edited by Victoria R. Bricker and Ga brielle Vail, pp. 27-71. Middle American Research Institute, Pub. 64. Tula ne University, New Orleans. 217

PAGE 218

218 Vogt, E. Z. and Stuart, D. 2005 Some notes on ritual caves am ong the ancient and modern Maya In The Maw of the Earth Monster: Mesoamerican Ritual Cave Use edited by J. E. Brady and K. M. Prufer, pp. 155-85. The Linda Schele Series in Maya and Pre-Columbian Studies. University of Texas Press, Austin. Wrobel, Gabriel D. 2008 Temporal Changes in the Mortuary Ritual Us e of the Caves Branch Rockshelter, Belize.. FAMSI website. 8 December 2008 . Wrobel, Gabriel D. and James C. Tyler 2006 Revisiting Caves Branch Rockshe lter: Results of the 2005 Excavations. The Belize Valley Archaeological Reconnaissance Project : A Report of the 2005 Field Season edited by Christophe G.B. Helmke & Jaime J. Awe, pp. 1-10. Institute of Archaeology, National Institute of Culture and History, Belmopan. Zender, Marc 2006 Teasing the Turtle from its Shell: AHK and MAHK in Maya Writing. The PARI Journal 6(3):1-14. Zubrow, Ezra B.W. 1994 Knowledge Representation and Archaeol ogy: A Cognitive Example Using GIS. In The Ancient Mind: Elements of Cognitive Archaeology, edited by C. Renfrew and E.B.W. Zubrow, pp. 107. Cambridge: Cambridge University Press.

PAGE 219

BIOGRAPHICAL SKETCH Erol Kavountzis is a native New Jerseyan who migrated south to attend the University of Florida to pursue a masters degree in anth ropology. As a child visiting and living in Greece during his summers, Erol built up an appreciation for the an cient Greeks and the ocean. Although his appreciation for the ocean never changed, hi s intereste in archaeological exploration did. Erol graduated from Boston University (BU) in 2002 with a dual major in Archaeology and Biology with a Specialization in the Marine Sciences. During the spring of 2000, Erol went on his first archaeological excavation as a student in the Xibun Archaeological Research Project (XARP) through BU headed by Patricia McAnany a nd a great team of graduate students. After graduating from BU he decided to pursue a ca reer in archaeology by working multiple jobs in Greece and Ukraine, and also within the United States, mainly in New Jersey, New York, and Pennsylvania. His interest in animal bone analys is allowed him many opportunities to work with domesticated animals within the Old World. After a few years in the Cultural Resource Mana gement field, Erol decided to return to graduate school to increase his knowledge in archaeology. He has spent the last few years working with Prof. Emery on multiple faunal collect ions from the Maya region. Erol also had a chance to work on excavations in Belize during the summer of 2007. The analysis of animal remains is an important part of his life a nd he hopes to one day continue his pursuit in understanding the deeper role that animals a nd humans both shape in th e modern and ancient world. 219