GARDENS, WILDLIFE DENSITIES, AND SUBSISTENCE HUNTING BY MAYA INDIANS IN QUINTANA ROO. MEXICO By JEFFREY PAUL JORGENSON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1993 My Best Friend, My Wife ACKNOWLEDGMENTS This project could not have been carried out without the help of numerous people. Most importantly, I thank Kent H. Redford and John G. Robinson for suggesting that a study of Maya subsistence hunters in Quintana Roo, Mexico, would be a wonderful opportunity to mesh my interests in indigenous peoples and the natural history of Neotropical birds and mammals. For their suggestions during the planning stage and for helping me to interpret the data, I would like to thank the members of my committee: Allan F. Bums, John F. Eisenberg, Kent H. Redford (Chairman), Allyn MacLean Stearman, and Melvin E. Sunquist. Allyn MacLean Stearman and Allan F. Bums were especially instrumental in helping me see the human side of this project and ir appreciating the role indigenous people play in natural resource use and management. I thank John Smallwood and Jaime E. Jimenez for their guidance on statistical analyses and computer programming. Funding for this project was provided by World Wildlife Fund-U.S. (Project 6470), World Nature Association, Organization of American States. Roger and Bemita Jorgenson (my parents), Centro de Investigaciones de Quintana Roo, and Program for Studies in Tropical Conservation/Tropical Conservation and Development Program (University of Florida). The Secretaria de DesarroUo Urbano y Ecologia kindly granted a research permit to work in Mexico (# 301878, dated 17 October 1990). My host agency in Mexico was the Centro de Investigaciones de Quintana Roo (CIQRO). I thank Eduardo Suarez Morales, CIQRO Director, and his staff for their patience and assistance, especially with respect to purchasing a truck, providing several field assistants, obtaining research permits, coordinating the identification of scientific specimens, and organizing an airplane reconnaissance flight over the study area. I especially thank Alvaro del Campo Parra Lara, Jesus Pahna Gutierrez, Laura Perez del Valle-Cunille, Enrique Montes de Oca, Julio Rafael Castillo Espadas, Teresa Jimenez A., Julio Juarez G., Miguel A. Marmolejo, Mariano Ceballos. and Daniel Navarro L., lU former CIQRO biologists, for introducing me to the Maya at X-Hazil Sur and sharing their experiences about how to conduct fieldwork in Mexico. Their friendship was a source of inspiration during many trying times. Several Mexican counterparts assisted in the identification of scientific specimens. For the classification of soil samples, I wish to thank Pedro Macario Mendoza. For the identification of plants, I wish to thank Edgar Cabrera Cano and Javier Chavelas Polito. For the identification of animal specimens, I wish to thank Hector Arita Watanabe, J. Enrique Escobedo Cabrera, Elsa M. Figueroa, Hector Flores, Livia Leon Paniagua, Adolfo G. Navarro Sigiienza, and Carmen Pozo de la Tijera. For assistance with Maya soil and vegetation type terminology, I wish to thank Francisco J. Rosado May. Several Mexican collaborators provided valuable information about the study area and state of Quintana Roo. For sharing their vegetation maps and information about forest inventories, I thank Marcelo Carre6n Mundo and Victoria Santos Jimenez. I thank Juan Bezaury Creel and Pedro Ramirez Guillen for numerous acts of kindness and for obtaining for me historical information about the area after I returned to the United States. My introduction to the Maya of Quintana Roo was greatly facilitated by having met Don Margar Tuz, his wife Dona Mila Novelo, and two of their sons, Alejandro and Margarito. They welcomed me into their home at Rancho Las Palmas, shared with me their knowledge of the forest, and made my stay in Quintana Roo much more enjoyable. I will treasure their friendship always. My stay at X-Hazil Sur was greatly enhanced by the cooperation and generosity of many people. I would especially like to thank Modesto Cab Cohuo and his wife, Martha Tuz Novelo, for receiving me in their home, introducing me to the other village residents, and explaining many important aspects of Maya culture to me. My stay also was enriched by the special attention afforded my wife and myself by several families. I especially thank Mauro Cab Cohuo and Rosa Pool Tun, Julio Poot Ake and Gudelia Uc Puc, Graciliano Gomez Puc, Fernando Balam Can, Joaquin Humberto Balam Xiu, Celestino Chan Cante, their families for their hospitality. Fieldwork at X-Hazil Sur was greatly improved by the cooperation and insight offered by iv many local residents. For their assistance during the censuses and the identification of stomach contents, I thank Jacinto Foot Cruz, Julio Foot Ake, Graciliano Gomez Fuc, Joaquin Humberto Balam Xiu, Jesus Cab Can, and Deltino Santos Chan. For allowing me to study their gardens and construct animal census transects in their work areas, I thank Simon Yeh Ake, Sotero Yeh Can, Lorenzo Tun Can, Lorenzo Chac Canul, Vicente Chan Che, Daniel Cocom Ucan, Domingo Guzman Foot Cruz, and Martin Cab Ake. For sharing their perspecUves on hunting, I thank Marcos Aban Noh, Benito Can Chi, Jacinto Foot Cruz, and Gilberto Yeh Foot. For permission to conduct research at Ejido X-Hazil y Anexos, I thank Delfino Ake Ayala, Comisariado of Ejido X-Hazil y Anexos, Juan Yeh Foot, Siib- delegado of X-Hazil Sur, and the residents of X-Hazil Sur, Uh-May, and Chancah Veracruz. Without their patience and understanding, this study would never have been possible. I hope that I have not betrayed their trust. I am especially indebted to my two Maya field assistants, Armando Balam Xiu and Rufino Ucan Chan. With great patience, humor, and interest, they processed game kills, conducted hunter interviews, cleared census transects, and identified game tracks. Their reliability, honesty, and ability to work with the various hunters at X-Hazil Sur were critically important to the success of this project. Armando and Rufino also became two of my best friends and provided important insights into village life. I hope that they also benefitted from their association with me. I would like to thank my friends for their ideas, comments, and technical support during this project, especially Jackie Belwood, Marjorie Bingham, Richard Bodmer, Kathy Campbell, Peter G. Crawshaw Junior, Sally and Josh Dickinson, Eduardo Ifligo Elias, Jaime E. Jimenez, Cynthia Lagueux, Julia Murphy, Patricia Negreros, Laura Snook, Jody Stallings, Wendy Townsend, and Laurie Wilkins. I especially would like to thank my parents, Roger and Bemita Jorgenson, and my mother-in- law, Alicia Rodriguez de Barrera, for their love and support during this endeavor. While they may not have understood the technical aspects of the study and certainly would have enjoyed a greater frequency of calls and letters from me during the research and writing, their patience, encouragement, and understanding were essential to the successful completion of this project. I would like to extend my sincerest ^preciation to Sally and Josh Dickinson for making available to me the resources of Tropical Research & Development, Inc., to produce this dissertation. By permitting access to their computers, printers, copiers, and fax machine, Sally and Josh greatly reduced the stress associated with the technical aspects of producing a dissertation and greatly enhanced the quality of the final document. To my wife and best friend, Amanda, I offer my deepest gratitude for her insight and patience during these trying times. Her cheerfulness and ability to help out in periods of great stress were essential to the completion of this study. I hope that in the future we both will be able to recall fondly these times and appreciate better the experiences that we shared. VI TABLE OF CONTENTS ACKNOWLEDGMENTS i" ABSTRACT ^^i RESUMEN ^v"> CHAPTERS 1. OVERVIEW 1 2. NATURAL. CULTURAL, AND SOCIOECONOMIC ENVIRONMENT OF THE MAYA INDIANS IN QUINTANA ROO. MEXICO 13 Introduciion 13 Natural Environment 1'^ Study Area ^^ Cultural and Socioeconomic Environments 30 Closing Comments 52 3. GAME HARVEST BY M.\YA HUNTERS IN QUINTANA ROO, MEXICO 53 Introduction ^- Methods ^^ Results 6^ Discussion . ^^ 4. WILDLIFE DENSITIES IN SUCCESSIONAL FORESTS AROUND A MAYA VILLAGE IN QUINTANA ROO, MEXICO 120 Methods P^ Results ^32 Discussion 13' 5. CONSUMPTION OF GARDEN CROPS BY GAME ANIMALS IN QUINT.\NA ROO, MEXICO 1'^'^ Introduction 1^' Methods 149 Results 155 Discussion 168 6. CHARACTERISTICS OF MAYA GARDENS AND THEIR USE BY \VTLDLIFE IN QUINTANA ROO, MEXICO 179 Introduction 179 Methods ^83 Results 191 Discussion 209 Vll 7. SYNTHESIS AND CONCLUSIONS 223 APPENDICES A TEMPERATURE (°C) AND PRECIPITATION (MM) AT X-HAZIL SUR, QUINTANA ROO, MEXICO, DURING JULY 1989-DECEMBER 1990 231 B MEAN BODY MASS (G), DIET CLASSIFICATION, MEAN POPULATION DENSITY (NUMBER OF INDIVIDUALS/KM^), AND GAME STATUS FOR SELECTED MAMMALS IN QUINTANA ROO 238 C MEAN BODY MASS (G), DIET CLASSIFICATION, MEAN POPULATION DENSITY (NUMBER OF INDIVIDUALS/KM-), AND GAME STATUS FOR SELECTED BIRDS IN QUINTANA ROO 241 D DATA FORMS USED DURING HUNTER INTERVIEWS. FORM A INCLUDES SPECIFIC QUESTIONS ABOUT THE OUTING (ONE FORM PER OUTING), WHEREAS FORM B INCLUDES QUESTIONS ABOUT THE GAME SPECIES TAKEN (ONE FORM PER PREY ITEM). BOTH FORMS ORIGINALLY APPEARED IN SPANISH 243 E GAME HARVEST INVENTORIES FOR THE SEVEN MAIN HUNTERS (315 PREY WITH WEIGHTS) AT X-HAZIL SUR COMPARED WITH THE GAME TAKEN BY THE REMAINING 79 HUNTERS (261 PREY WITH WEIGHTS) THAT REPORTED TAKING GAME (ARRANGED BY GRAND TOTAL NUMBER OF PREY ITEMS) 249 F CATALOGUE OF GAME AND NONG.\ME ANIMALS COLLECTED AT EJIDO X-HAZlL Y ANEXOS DURING 1989-1990, INCLUDING DONATIONS BY LOCAL RESIDENTS, ANIMALS FOUND ALONG THE ROAD, AND GAME TAKEN BY HUNTERS 251 G PERCENT FREQUENCY OF THE GAME KILL SITES IN THE FOUR MAIN VEGETATION TYPES FOR THE GAME TAKEN BY HUNTERS AT X- HAZIL SUR 254 H SUMMARY OF CENSUS RESULTS AND CHARACTERISTICS OF 12 TRANSECTS IN THREE SUCCESSIONAL STAGES OF FOREST 256 I CENSUS DATA FORM 258 J MEAN SIGHTING DISTANCES OF WILDLIFE (M), SUMMARIZED BY SPECIES, GROUPS, AND GAME AND NONGAME BIRDS AND MAMMALS CENSUSED ALONG 12 TRANSECTS IN THREE SUCCESSIONAL STAGES OF FOREST AT EJIDO X-HAZIL Y ANEXOS, QUINTANA ROO, MEXICO, DURING 1990 (D.F. = 2 THROUGHOUT) 261 K NUMBER OF SIGHTINGS/ 100 KM AND TOTAL NUMBER OF SIGHTINGS (N), SUMM.\RIZED BY SPECIES, GROUPS, AND GAME AND NONGAME BIRDS AND MAMMALS 263 viu L COMMON AND SCIENTIFIC NAMES OF BIRDS AND MAMMALS SIGHTED AT EJIDO X-HAZIL Y ANEXOS DURING 121 CENSUSES IN 1990 266 M DATA FORM; STOMACH CONTENTS ANALYSIS 268 N PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF PACA FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 36 STOMACHS (TOTAL VOLUME ANALYZED = 1886.5 ML) COLLECTED FROM PACAS TAKEN BY MAYA HUNTERS AT X- HAZIL SUR, QUINTANA ROO, MEXICO, DURING JUNE 1989- OCTOBER 1990 270 O PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF AGOUTI FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 26 STOMACHS (TOTAL VOLUME ANALYZED = 1167.5 ML) 273 P PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF COATI FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 129 STOMACHS (TOTAL VOLUME ANALYZED = 4678.5 ML) 276 Q PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF WHITE- LIPPED PECCARY FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF THREE STOMACHS (TOTAL VOLUME ANALYZED = 208.0 ML) 280 R PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF COLLARED PECCARY FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 29 STOMACHS (TOTAL VOLUME ANALYZED = 1829.0 ML) 282 S PERCENT OCCURRENCE {% OCC.) AND VOLUME (% VOL.) OF BROCKET DEER FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF EIGHT STOMACHS (TOTAL VOLUME ANALYZED = 446.0 ML) 286 T PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF WHITE- TAILED DEER FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 11 STOMACHS (TOTAL VOLUME ANALYZED = 601.0 ML) 289 U PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF THICKET TINAMOU FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF THREE STOMACHS (ESOPHAGUS, CROP, PROVENTRICULUS, AND GIZZARD; TOTAL VOLUME ANALYZED = 13.0 ML) 293 V PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF GREAT CUR.\SSOW FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 13 STOMACHS (ESOPHAGUS, CROP, PROVENTRICULUS, AND GIZZARD; TOTAL VOLUME ANALYZED = 694.0 ML) 295 IX W PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF PLAIN CHACHALACA FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 21 STOMACHS (ESOPHAGUS, CROP, PROVENTRICULUS, AND GIZZARD; TOTAL VOLUME ANALYZED = 329.5 ML) 298 X PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF OCELLATED TURKEY FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF FIVE STOMACHS (ESOPHAGUS, CROP, PROVENTRICULUS, AND GIZZARD; TOTAL VOLUME ANALYZED = 211.5 ML) 301 Y GARDEN INTERVIEW FORM • 303 Z GAME SPECIES AND MAIN CROP SPECIES AT EJIDO X-HAZIL Y ANEXOS, QUINTANA ROO, MEXICO, DURING 1989-1990 306 LITERATURE CITED 309 BICXjRAPHICAL SKETCH 335 LIST OF TABLES 2-L Land uses and vegetation types at Ejido X-Hazil y Anexos, Quintana Roo, Mexico 28 3-1. Reported number of individuals taken, mean weight (kg), and total weight (kg) of game taken by Maya hunters at X-Hazil Sur, Quintana Roo, Mexico, during June 1989- October 1990 71 3-2. Sex and age class of game taken by hunters at X-Hazil Sur 72 3-3. Hunter group size and characteristics of successful outings by hunters at X-Hazil Sur 87 3-4. Mean prey body weight (kg) by type of weapon for game species taken at X-Hazil Sur .... 88 3-5. Type of hunt and characteristics of successful outings by hunters at X-Hazil Sur 89 3-6. Combinations of game mammals and birds taken per outing by hunters at X-Hazil Sur .... 90 3-7. Percent frequency of game kill sites in Late Secondary Forest versus all Combined/Early Secondary Forest and x^ test results for the game taken by hunters at X-Hazil Sur 92 3-8. Minimum harvest area (km-) and mean distance (km) between X-Hazil Sur (water tank) and kill sites of primary game species taken by hunters at X-Hazil Sur 93 3-9. Mmimum catchment area (km^) and mean kill site distance (km) from X-Hazil Sur (water tank), by hunter, for 315 game animals taken by the seven main hunters versus all other hunters combined at X-Hazil Sur 94 3-10. Percent frequency of game kill sites in forests vs. fallows/gardens in Brazil (Ka'apor Indians; Balee, 1985) and Ecuador (Runa Indians; Irvine, 1987) 114 3-11. Catchment areas for various mestizo and indigenous groups in the Neotropics (arranged by size of catchment area) 115 3-12. Comparison of the biomasses of the Cerro Brujo mammals with the biomasses for the same species at Ejido X-Hazil y Anexos 117 4-1. Average sighting frequency (mean number of sightings per 10,000 km) of wildlife, summarized by species, groups, and game and nongame birds and mammals censused along 12 transects in three successional stages of forest at Ejido X-Hazil y Anexos, Quintana Roo, Mexico, during 1990 (d.f. = 2 throughout) 136 4-2. Population density estimates (individuals/km^; total number of sightings [n]) determined by King's Method and summarized by species, taxonomic groups, and game and nongame birds and mammals 138 XI 4-3. Reported number of individuals taken by Maya hunters at X-Hazil Sur versus number of sightings during animal censuses (game species only) 144 4-4. Estimated population densities of mammals at Ejido X-Hazil y Anexos compared with those at six other study areas in Neotropical forests (summarized from Glanz [1990]) ... 145 5-1. Game species and main crop species at Ejido X-Hazil y Anexos, Quintana Roo, Mexico, during 1989-1990 152 5-2. Percent occurrence of major crop species in the stomach contents of game species (n = 284 stomach contents analyzed; total volume analyzed = 12,064. 1 ml) during June 1989-October 1990, excluding the pocket gopher 162 5-3. Seasonal occurrence of major CTop species in the stomach contents of game species (n = 284 stomach contents analyzed, total volume analyzed = 12,064.1 ml) during June 1989-October 1990, excluding the pocket gopher 164 5-4. Seasonal variation of major aop species in (a) paca stomach contents (n = 36 stomach samples, total volume analyzed = 1886.5 ml) and (b) agouti stomach contents (n = 26 stomach samples, total volume analyzed = 1167.5 ml), expressed as percent occurrence (% Occ.) and volume (% Vol.) 166 5-5. Seasonal variation of major crop species in (a) coati stomach contents (n = 129 stomach samples, total volume analyzed = 4678.5 ml) and (b) collared peccary stomach contents (n = 29 stomach samples, total volume analyzed = 1829.0 ml), expressed as percent occurrence (% Occ.) and volume (% Vol.) 167 6-1. Density of crops (mean number of stalks/m^) planted in the seven selected gardens in 1990. For crop density, the plots were surveyed during 7/90-11/90, beginning before planting and ending when the com harvest began in the garden. Additional surveys were conducted through 1/91 to monitor crop phenology 203 6-2. Frequency of occurrence of wildlife track sets in 18 track slicks (1 m by 20 m) along gardens and transects in adjacent forest during 1990 205 xu LIST OF FIGURES 2-1. Location of Ejido X-Hazil y Anexos (study area), Quintana Roo, Mexico 15 2-2. Distribution of X-Hazil Sur residents by age class and sex. There were 539 males and 501 females during the census of 1992 26 2-3. Mean monthly temperature and precipitation at X-Hazil Sur, Quintana Roo, Mexico, durmg July 1989-December 1990 27 2-4. Land uses and vegetation types at Ejido X-Hazil y Anexos 29 2-5. Monthly occurrence of subsistence and economic activities in relation to temperature and rainfall (Max = month with highest average temperatures and Min = month with lowest average temperatures) 47 3-1. Reproductive status of female birds and mammals. Animals were categorized as gravid, lactating/brooding, not gravid/lactating/brooding, or unknown (abbreviations correspond with scientific name; Ohi = pocket gopher, Apa = paca, Dpu = agouti, Nna = coati, Tpe = white-lipped peccary, Tta = collared peccary, Mam = brocket deer, Ovi = white-tailed deer, Cci = thicket tinamou. Cru = great curassow, Ove = plain chachalaca, and Aoc = ocellated turkey,) 74 3-2. Reported monthly harvest levels for a) pocket gopher, b) paca, c) agouti, and d) coati 75 3-3. Reported monthly harvest levels for a) white-lipped peccary, b) collared peccary, c) brocket deer, and d) white-tailed deer 76 3-4. Reported monthly harvest levels for a) thicket tinamou, b) great curassow, c) plain chachalaca, and d) ocellated turkey 77 3-5. Monthly occurrence of gravid or lactating female mammals 79 3-6. Monthly occurrence of gravid or brooding female birds 80 3-7. Number of game birds and mammals taken per hour 81 3-8. Age class (years) distribution of male residents (539 residents based on 1992 census) and male hunters (84 hunters during 1989-1990 study) 82 3-9. Number of kills per hunter. A total of 47 hunters had only one or two kills each and are not shown here 83 3-10. Take of mammals and birds according to type of weapon used. Species arranged by mean body mass determined during this study (see Figure 3-1 for key to abbreviations) 85 4-1. Distribution of 12 transects (ca. 2 km long each) used to census wildlife populations at Ejido X-Hazil y Anexos, Quintana Roo, Mexico, during 1990 127 xiu 4-2. Cumulative number of avian (7) and mammalian (16) taxa sighted during 121 censuses in 1990 133 5-1. Proportion (percent volume) of food items identified from stomach contents of game species taken by Maya hunters at X-Hazil Sur, Quintana Roo, Mexico, during June 1989- October 1990. (Food items: FRUITS/SEEDS = fruits or seeds, LEAVES = leaves, ANIMAL MAT = animal material, and OTHER = other [see Methods for further information]. See Table 5-2 for key to species abbreviations.) 157 5-2. Percent volume of crops in game stomachs (n = 284 stomachs analyzed) and percent occurrence of kill sites in areas categorized as Combined/Early Secondary Forest (n = 584 game kills) 176 6-1. Annual horticultural cycle at Ejido X-Hazil y Anexos during 1989-1990 (Max = month with highest mean temperature and Min = month with lowest mean temperature) 193 6-2. Distribution of gardens by size during 1989 (n = 150 gardens, average size = 2.1 ha) and 1990 (n = 40 gardens, average size = 2.7 ha) 195 6-3. Distribution of gardens by distance from X-Hazil Sur during 1989 (n = 150 gardens, average distance = 4.0 km) and 1990 (n = 40 gardens, average distance = 3.8 km) 197 6-4. Frequency with which crops were planted in gardens during 1989 and 1990 (C = com, S = squash, KB = kidney beans, 1 = ibes, M = yam, SP = sweet potato, CHI = chile, F = fruits, Y = yucca, J = jicama, CHA = chaya, L = lentils, and O = other) .... 198 6-5. Number of crops per garden during 1989 (n = 150 gardens, average = 5.9 crops per garden) and 1990 (n = 40 gardens, average = 3.9 crops per garden) 199 6-6. Monthly phenology of com plants in seven selected gardens during 1990 (n = 7 gardens). Plants were categorized as seedlings, with flowers, with immature fruits, with mature fruits, and harvested 200 6-7. Monthly phenology of kidney beans at seven selected gardens during 1990 (n = 7). See Figure 6-6 for key to abbreviations 201 6-8. Monthly phenology of squash as seven selected gardens during 1990 (n = 7 gardens). See Figure 6-6 for key to abbreviations 202 6-9. Frequency with which wild animals were identified by gardeners as the main crop predators during 1989 and 1990 (none = no crop predators, c peccary = collared peccary, wt deer = white-tailed deer, gusanos = various taxa of Gastropods) 204 6-10. Monthly frequency of wildlife track sets in a) Late Secondary Forest without Gardens (LSF W/0 G; 16 track sets during 48 readings), b) Late Secondary Forest with Gardens XIV (LSF W/ G; 40 track sets during 38 readings), and c) Early Secondary Forest (ESP; 40 track sets during 49 readings) 208 XV Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfilhnent of the Requirements for the Degree of Doctor of Philosophy GARDENS, WILDLIFE DENSITIES. AND SUBSISTENCE HUNTING BY MAYA INDIANS IN QUINT ANA ROO, MEXICO By Jeffrey Paul Jorgenson December 1993 Chairman: Dr. Kent H. Redford Major Department: Forest Resources and Conservation Subsistence hunting by Maya Indians was studied at Ejido X-Hazil y Anexos, Quintana Roo, Mexico, during 1989-1990. These highly acculturated Indians have hunted and planted gardens in the Yucatan Peninsula for several thousand years. The goal of tliis study was to evaluate the premises of "garden hunting." This kind of hunting is a special mammal harvesting pattern based on interactions between hunters, who plant gardens and harvest game, and game species Uiat eat crops and have greater population densities in the vicinity of gardens than in forest areas without gardens. The specific objectives included the following: (1) determine game harvest patterns by Maya subsistence hunters; (2) compare wildlife densities in three forest successional stages; (3) determine the kinds and amounts of crops consumed by game species; and (4) characterize Maya gardens and tneir use by wildlife. Maya hunters harvested eight species of mammals and four species of birds as game. A total of 584 animals were harvested by 88 hunters during 17 months. The coati [Nasua nasua) and plain chachalaca (Ortalis vetitla) were the most frequently harvested taxa (167 individuals each). Game was harvested throughout the year and frequently was taken in areas associated with shifting cultivation. Wildlife densities were low compared with those at other Neotropical forest sites, and significant differences in XVI wildlife density among three forest successional stages by species, taxonomic group, and game or nongame status generally were not observed. The average garden size was 2.1 ha in 1989 and 2.7 ha in 1990 (maximum size was 18.0 ha). Sixteen crops were planted by Maya gardeners, of which six crops were consumed by game species. Com, kidney beans, and squash were most frequently planted. Com was the most frequently eaten aop (25.4% of 284 stomach samples). Four game species were the primary consumers of crops: paca (Agouti paca), agouti (Dasyprocta punctata), coati, and collared peccary {Tayassu tajacu). These game species were among the most frequently taken taxa by number of individuals and total body weight. Given that wildlife populations generally were not greater in the vicinity of gardens, these results generally do not support the premises of garden hunting. These results also suggest that an effective wildlife management program should be implemented in order to ensure the survival of these wildlife species and the cultural traditions of the Maya hunters. xvu Resumen de Disertaci6n para la Escuela de Postgrado de la Universidad de Florida en Cumplimiento Parcial de los Requisites para el Grado de Doctor de Filosofia MILPAS, DENSIDADES DE VIDA SILVESTRE Y CACERIA DE SUBSISTENCIA POR LOS INDIGENAS MAYAS EN QUINT ANA ROO, MEXICO For Jeffrey Paul Jorgenson Diciembre de 1993 Director: Dr. Kent H. Redford Departamento: Recursos Forestales y Conservaci6n La caceria de subsistencia por los indi'genas mayas fue estudiada en el Ejido de X-Hazil y Anexos, Quintana Roo, Mexico, durante 1989-1990. Esta gente, bastante modemizada. ha cazado y cultivado milpas en la Peninsula de Yucatan per miles de anos. El objetivo de este estudio fue evaluar las teorias acerca de la caceria de "parcela agricola" de subsistencia. Esta clase de caceria de mamiferos es un tipo especial de explotaci6n basado en las interacciones entre los cazadores, que siembran las parcelas, y los animales y especies cinegeticas, que se alimentan de los cultivos y tienen densidades poblacionales grandes alrededor de las parcelas en comparacidn a las areas forestales sin parcelas. Los objetivos especiTicos incluyeron: (1) detenninar los patrones de la caza de animales; (2) comparar las densidades de los animales en tres estados sucesionales del bosque; (3) determinar las clases y las cantidades de cosechas consumidas por los animales cinegeticos; y (4) caracterizar las parcelas agricolas y su uso por los animales. Los cazadores mayas cazaron ocho especies de mamiferos y cuatro especies de aves. Un total de 584 animales fueron cazados por 88 cazadores durante 17 meses. El tej6n (Nasua nasua) y la chachalaca (Ortalis vetula) fueron las especies cazadas con mas frecuencia (con 167 individuos cada una). Hubo caceria a lo largo del ano y con frecuencia los animales fueron cazados en areas de agricultura de roza, tumba y quema. Las densidades de XVIU animales fueron bajas, comparadas con otras areas forestales neotropicales, y no se observaron diferencias significativas en la densidad de los animales por especie, grupo taxon6mico, o estado cinegetico o no cinegetico en tres areas forestales en estado sucesional diferente. El tamafio promedio de las milpas fue de 2.1 ha en 1989 y 2.7 ha en 1990 (el tamafio maximo fue 18.0 ha). Dieciseis especies de plantas fueron cultivadas en las milpas, de las cuales seis fueron consumidas por las especies de caceria. El mafz, frijol, y calabaza fueron los cultivos mas comunes. El maiz fue el cultivo mas frecuentemente utilizado por los animales (25.4% de 284 muestras estcmacales). Cuatro especies cinegeticas fueron los consumidores principales de las cosechas: el lepezcuintle [Agouti paca), sereke (Dasyprocta punctata), tej6n y jabali de collar (Tayassu tajacu). Estas especies fueron las mas cazadas tanto por niimero de animales como por peso total del cuerpo. Dado que las poblaciones de fauna silvestre no fueron mayores alrededor de las milpas, estos resultados en general no apoyan las teorias acerca de las caceria en parcelas agricolas. Ademas, estos resultados sugieren que un programa efectivo de manejo de vida silvestre debe ser ejecutado para asegurar la supervivencia de estas poblaciones de vida silvestre y las tradiciones culturales de los cazadores mayas. XIX CHAPTER 1 OVERVIEW This dissertation proposes to examine the interrelation between gardens, wildlife densities, and subsistence hunting in a Maya Indian community in southeastern Mexico. The Maya are a rural people and have practiced shifting cultivation for several thousand years. In many regards, they subsist today as they have done for thousands of years by hunting, caring for wild and domestic animals, and by planting crops in their gardens. The Maya, however, have undergone rapid cultural change due to tourism and economic development in the Cancun area and slowly are being incorporated into Mexican political and economic activities. Subsistence hunting as practiced today by the Maya is unusual when compared with hunting by other indigenous groups in two regards: One, due to the availability of domestic livestock and canned meat, hunters are not dependent upon the game they harvest as a source of food. Two, the Maya have been able to live in permanent settlements and exploit the local wildlife populations apparently without exhausting local wildlife populations, while many other indigenous groups have depleted local wildlife populations and consequently have had to move their settlements repeatedly or undergo seasonal treks (Stearman, 1990; Vickers, 1988, 1991; Werner, 1983). An understanding of how the Maya have been able to continue subsistence hunting for so long, while indigenous peoples in other areas have not been able to do so, will provide important information to biological and social scientists seeking to balance the issues of conservation and economic development. The research described here was designed to examine the hypothesis that subsistence hunting by Maya Indians may be more than a simple game harvest activity. Rather, Maya subsistence hunting may be part of a complex game-procurement system composed of gardens, wildlife, and hunters. "Garden hunting," desaibed by Linares (1976) for a prehistoric indigenous people in Panama, may actually be the model whereby the Maya have been able to hunt and practice shifting cultivation in the 1 2 Yucatan Peninsula for several thousand years. If it can be shown that garden hunting is compatible with sustainable wildlife use, then biological and social scientists will have an important tool in developing management plans diat balance the needs of both die hunters and the hunted. Uses of Wildlife The importance of wild animals to Uie Maya and other indigenous people in the Neotropics is best appreciated by noting the benefits obtained from their use. While indigenous groups differ in the species of game animals they take and taboo (prohibit taking for cultural reasons), several broad patterns of wildlife use can be identified. Here I will consider three of the principal uses of game; food, nonedible products, and live animals. Bushmeat, meat from edible wildlife, is one of the most important uses of wild animals (Dufour, 1983; Flowers, 1983; Yost and Kelley, 1983). A wide variety of fishes, birds, mammals, reptiles, amphibians, and insects are taken in a multitude of habitats by a broad range of people (Beckerman, 1980, 1983; Chemela, 1985; Redford and Robinson, 1987, 1991; Robinson and Redford, 1991a, 1991b, 1991c; Stocks, 1983). Game taken by indigenous people in die Neotropics usually is for subsistence (Lugo et al., 1987; Robinson and Redford, 1991a, 1991b), but meat and eggs of several taxa frequendy are also taken for commercial purposes and are sold locally or outside the community. Capybara (Hydrochaeris hydrochaeris) meat, which, because of its aquatics habits is actually considered a "fish," is traded commercially in Venezuela during Lent (Ojasti, 1991). while eggs from sea turtles and freshwater turUes (Order Testudines) are taken along die coasts and inland waterways in Brazil and Honduras and sold in local markets (Cornelius et al., 1991; Lagueux, 1991; Smith, 1974). In addition to the nutritional aspects of meat, game is important for social reasons. Investigators have shown that die type and amount of game obtained is important to a hunter's prestige and die social cohesion of die family or village (Balee, 1985, 1989; Paolisso and Sacket, 1985; Siskind, 1973; Stearman, 1989, 1990). Occasions without meat often are perceived by indigenous people as times of hunger even though plenty of food from plants is available (Werner, 1983). 3 A second important use of game is to provide nonedible products such as leather, skins, hides, and feathers. Use of skins and hides initially was limited to that by indigenous people to perform rituals, conduct trade with other indigenous groups, and manufacture clothing and artifacts. Today, the use of nonedible products has greatly increased in volume and includes indigenous groups, mestizos, and local and foreign business people. Much of the trade in leather, skins, hides, and feathers now is for commercial purposes. Most of the recent leather trade for example, has been composed of peccaries (Tayassu tajacu, T. pecari; Broad, 1984; Hvidberg-Hansen, 1970), capybaras (Smith, 1981b), and various species of lizards (Suborder Sauria) and crocodiles (Order Crocodylia; Luxmoore, 1988). Until recendy, die skin trade was concentrated on a few species of carnivores taken to satisfy Uie international demand, including giant otter (Pteronura brasiliensis), river otter (Lutra longicaudis), jaguar {Panthera onca), and several smaller feline species (Felis spp.; McMahan, 1986). This trade in skins, however, generally has ended due to inteiTiational restrictions such as the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). Trade in feadiers has declined markedly, too, but Aztec and Inca Indians formerly used feaUiers in capes and artifacts utilized during religious ceremonies (Haemig, 1978). The feaUier trade today is limited primarily to feadiers from rheas (Pterocnemia pennata and Rhea americana) that are made into dusters (Cajal, 1988). Other nonedible products obtained from wild animals include guano from bats and birds (Haynes, 1987); oil from reptiles, bird eggs, and manatees (Trichechus manatus, and T. inunguis; Pereira, 1944; SmiUi, 1974, 1981a. 1981b); bones, teeth, and claws for tools, handicrafts, and ornamental purposes (March M., 1987; Parra Lara, 1986; Santana et al., 1990); and animal parts used for medicinal and ritual purposes, such as sloth (Bradypus tridactylus and Choloepus hoffmanni; Romanoff, 1984) and little spotted cats (Felis tigrina; Jorgenson and Jorgenson, 1991). A third principal use of wildlife is as live animals. Both indigenous and nonindigenous people keep many species of live animals as pets. Song birds and parrots commonly are used for this purpose (Beissinger and Snyder, 1991; Inigo-Elias and Ramos, 1991; Jorgenson and Thomsen, 1987; Thomsen 4 and Brautigam, 1991). Many wild aninials originally kept as pets subsequently are eaten, for example, guanaco (Lama guanaco; Gilmore, 1950), collared peccary (Tayassu tajacu), paca (Agouti paca), and agouti (Dasyprocta punctata; Jorgenson, pers. obs.). Another use of live animals from the Neotropics is for the biomedical and zoo trade. Between 1961-1965, for example, 139,000 live primates were exported from Iquitos, Peru (Grimwood, 1969), and in 1973, at least 91,662 primates were exported from Iquitos (Castro, Revilla, and Neville, 1975- 76). Reptiles and amphibians also were traded for these purposes. Today, international trade in these species for commercial purposes is strictly regulated. The studies mentioned above show that 1) many taxa of wild animals are used by settlers and indigenous people, 2) these taxa have many different types of uses, and 3) these uses have occurred over millennia. These facts suggest that wildlife can be used sustainably in some cases. The current rate of human population growth and corresponding decrease in natural habitat, however, suggest that many species of plants and animals will become extinct in the near future as the areas where they occur are converted by humans to areas unable to support the original complement of plants and animals (Meffe et al., 1993; Myers, 1987, 1989; Robinson, 1993; World Resources Institute, 1990). The challenge as conservation biologists is to identify and promote ways in which wild animals can be used sustainably. Sustainable Use of Wildlife While studies have shown that some people in the Neotropics appear to use wildlife at sustainable levels (Dufour, 1990; Stearman and Redford, 1992; Vickers, 1991), there also is evidence that other people hunt or capture wildlife at levels that may not be sustainable, especially when these people participate in market economies (Dourojeanni, 1985; Fitzgerald, 1989; Ojeda and Mares, 1984; Redford and Robinson, 1985; Vickers, 1988). Evidence that people can overexploit local wildlife populations presents a dilemma to many organizations. For example, the United Nations Environmental Programme (UNEP), World Wildlife 5 Fund-U.S. (WWF-U.S.), World Wide Fund for Nature (WWF), and the International Union for Conservation of Nature and Natural Resources (lUCN) promote the incorporation of settlers and indigenous people into the management of natural areas and local populations of plants and animals (Chicch6n, 1992; Di Castri et al., 1981; Halffter, 1981; Redford and Stearman, 1989). However, in view of the willingness of some people in the Neotropics to deplete local wildlife populations by overhunting and habitat alteration, it is critically important that studies be conducted that carefully measure wildlife use and habitat alteration. The purpose of these studies should be to identify sustainable uses of the wildlife and the areas where they occur. Given that the factors affecting local wildlife and habitat communities vary and that local settlers and indigenous people have different needs, interests, and abilities, several questions need to be considered in a study of sustained use of wildlife, such as subsistence hunting: What are the pressures to overexploit plants and animals? Could market hunting become a problem? Are there any cultural limitations that need to be considered? The results likely will suggest certain basic principles about the sustainability of subsistence hunting, but that people and wildlife in different areas are subject to distinctive factors that affect this activity. Many anthropological and biological studies about hunting by settlers and indigenous people have been undertaken recently in the Neotropics. These studies are of limited value in evaluating the sustainability of hunting or in generalizing hunting from one site to another because the data were collected for other reasons. Recent anthropological studies in the Amazon Basin, for example, have focused on such topics as wildlife use (Vickers, 1991), optimal foraging (Hames and Vickers, 1982; Hill and Hawkes, 1983), protein consumption (Beckerman, 1979; Gross, 1975), hunting strategies (Paolisso and Sackett, 1985; Saffirio and Scaglion, 1982), and resource availability (Bailey et al., 1989). Recent biological studies have examined such topics as food habits and habitat use of game species (Bodmer, 1989, 1990, 1991), game use by settlers and indigenous people (Ayres et al., 1991; Redford and Robinson, 1987; Vickers, 1991), and the effects of habitat disturbance on wildlife (Johns, 6 1986, 1988). Taken together, these studies provide much useful information about the nature and extent of human hunting, but they have limited use in applications to other issues. In order to overcome the limitations of these anthropological and biological studies about hunting and to examine critically the issue of the sustainability of subsistence hunting, integrated studies are needed. These studies must provide data on what kinds of wildlife are hunted and, simultaneously, what kinds of wildlife are potentially available to hunters (Redford and Robinson, 1990). Garden Hunting The specific methods of subsistence hunting in the Neotropics vary considerably among groups of indigenous people with respect to species taken, technology used, timing and duration of hunts, and where the hunt is undertaken (cf., Hames and Vickers, 1983) One place where hunting frequently occurs is in the gardens planted by local residents. Hunting in gardens has been described for many indigenous groups, including the Achuara-Jivaro of Peru (Ross, 1978); the Yanomama (Smole, 1976) and Ye'kwana of Venezuela (Hames, 1980); the Lacandon Maya of Mexico (Nations and Nigh, 1980); the Yukpa, along the frontier between Venezuela and Colombia (Ruddle, 1974); the Miskito Indians of coastal Nicaragua (Neitschmann, 1972, 1973); the Sirion6 of Bolivia (Holmberg, 1969); and the Kayap6 (Posey, 1982, 1985) and Ka'apor (Balee, 1985) of Brazil. In general, these studies have quantified hunting or have described game use, but they have not demonstrated diat this game harvest is part a system where the wild animals actually are managed by the local residents. Linares (1976) was the first to propose diat subsistence hunting by indigenous people in the Nootropics was something more than hunters taking wild animals where the wild animals were abundant or easy to harvest. Based on archeological evidence, Linares (1976) proposed that, prior to the arrival of Columbus, indigenous people at Cerro Brujo, Panama, practiced a type of wildlife management in which they planted crops in gardens and hunted the wild animals that fed incidentally on these crops. This game-procurement system, according to an analysis of about 6,000 bones from refuse piles at two occupation sites, appeared to be selective in that only a certain group of terrestrial 7 mammals was taken. Although Linares (1976) did not present information about the occurrence and relative abundance of the other taxa present at the area, she concluded that the number and kinds of animals taken by the hunters were affected by 1) shifting cultivation, especially of cultivated root crops, and 2) the behavioral preadaptation of these mammals to become commensals of people. Linares (1976) called this hypothetical game-procurement system "garden hunting" and characterized it (p. 344) from two different perspectives: One, "The most abundant animals present are either smallish animals that live in the underbrush or in burrows, often in the vicinity of encampments or recently cleared fields (the caviomorph rodent[sl and armadillo), or larger forms that are not too shy and live— or can live— in forest-edge conditions (the collared peccary and the white-tailed deer)." Two, "The mammals missing altogether or poorly represented are either those that inhabit the high canopy (monkeys [Cebus, Ateles, Alouatta, etc.], sloths [Bradypus and Choleopus] or those that are fast climbers (coatis [Nasua nasua], squirrels [Sciurus]) or those that are very shy and live in forested conditions away from man (the brocket deer and tapir [Tapirus bairdii]." The supporting evidence for these characterizations was based on 1) fauna! assemblages in Neotropical forests at Suriname and Barro Colorado Island, Panama (Eisenberg and Thorington, 1973) and 2) behavioral and ecological traits of the most important mammals hunted by the prehistoric Cerro Brujo group. Based on a comparison of mammalian biomasses, Linares (1976) determined that none of the dominant species (in terms of biomass) at Surinam or Barro Colorado Island was taken by Cerro Brujo hunters. Linares (1976) hypothesized that this was due to a difference between faunal assemblages among the three areas, and indicated that Cerro Brujo represented a cultural faunal 8 assemblage, whereas Suriname and Barro Colorado Island represented natural or undisturbed assemblages. Based upon a comparison of behavioral and ecological traits for the terrestrial mammals harvested at Cerro Brujo, Linares (1976) determined that there were differences between closely related species that affected the frequency with which they were harvested. Between peccaries, the white- lipped peccary needs a large home range and probably a large forest, whereas the collared peccary is known to live in disturbed conditions and readily eats cultivated crops. Between deer, the brocket deer is shy and dwells in the forest, whereas the white-tailed deer occurs in cleared and cultivated fields. The collared peccary and white-tailed deer were harvested more frequently by Cerro Brujo hunters than were white-lipped peccaries or brocket deer. The other mammals taken at Cerro Brujo also ate crops or used gardens. Garden hunting has been described for various indigenous people, but the intensity and nature of these studies has varied. Perhaps one of the most thorough studies of this practice was conducted by Irvine (1987) at a site in the Ecuadorian Amazon that was populated by Runa Indians. Irvine (1987) quantified the take of game with respect to habitat type (i.e., gardens, fallow, and forest). By rank order of frequency of take, the acouchy (Myoprocta pratii, 25.6% of kills, total number of kills = 833), agouti (14.0%), paca (13.3%), and squirrels (Sciurus spp., 11.1%) were most commonly taken by Runa hunters. By rank order by weight of kills, the paca (25.8%, total weight = 3,439.7 kg), collared peccary (20.4%), agouti (14.1%), and brocket deer (Mazama spp., 12.7%) were more conunon. Among the caviomorph rodents and the large mammals, there were differences in the frequency of game kills by habitat type. For the rodents, the paca (20.8% of kills in gardens, 21.7% of kills in fallows, and 57.5% of kills in forest) and the agouti (15.8%, 32.4%, and 51.8%) readily used gardens and fallows, whereas the acouchy (1.5%, 27.5%, and 70.7% [corrected values from Table 4-5 of Irvine]) readily used fallow and forest, but only infrequently used gardens. For the large mammals, the collared peccary (0.0%, 14.3%, and 85.7% [corrected values from Table 4-5 of Irvine]) and the brocket deer (0.0%, 35.0% and 65.0%) readily used fallow and forest, but did not use gardens. 9 With respect to whether the Runa practice garden hunting, these results showed that the agouti and paca were both numerically important and frequently harvested in gardens. The acouchy, on the other hand, was numerically important, but did not use gardens. Among the large mammals, the collared peccary and brocket deer were neither numerically important nor harvested in gardens. Both species, however, were harvested in fallow areas. Irvine (1987) concluded that the primary factor responsible for the occurrence of game kill sites in gardens and fallows was not the availability of crops, but rather the occurrence of fruiting trees such as chunda (Bactris gasipaes) and sacha abillu (species undetermined. Family Sapotaceae). The present research is the first integrated study of garden hunting to demonstrate the possible interdependence between hunting, gardens, and game abundance. In this study, I examine simultaneously wildlife take, wildlife availability, wildlife food habits, and wildlife use of gardens in a single area. Maya Gardens Maya gardeners today plant two types of gardens: milpa and solar (Caballero, 1992; Forrest, 1991). The milpa is a relatively large plot of land that is located away from the house. Crops such as com and beans are planted in these gardens. Milpas are moved to new sites every 1-2 y. The solar is a relatively small plot that is located around die house or in the immediate vicinity of the house. Fruits, herbs, and medicinal plants are planted in these gardens. Solares are not moved from year to year. In this dissertation, the term garden will be used to mean milpa. Research Objectives The overall research question of this study is the following: "Do Yucatec Maya Indians at Ejido X-Hazil y Anexos, Quintana Roo, Mexico, practice garden hunting?" If Maya subsistence hunting is indeed garden hunting per Linares (1976), I would predict the following: one, wild animals taken as game would use gardens; two, hunters would concentfate their harvest of game on those 10 species of wild animals that used gardens; three, the densities of these wild animals taken as game would be greater in the vicinity of gardens than in forests without gardens; and four, the wild animals taken as game would consume crops from gardens. In order to evaluate the nature and extent of this hunting, questions were formulated following suggestions by Redford and Robinson (1985, 1990). Fieldwork was conducted during 1989-1990. The results form the basis for the remaining chapters of this dissertation. In Chapter 2, I discuss the natural, cultural, and socioeconomic environment of the Maya Indians in Quintana Roo, Mexico. This general discussion of the setting will facilitate comparisons with other sites. In Chapters 3-7, I present and discuss specific data collected during the field study. In Chapter 3, I discuss wildlife harvest by Mayan hunters in Ejido X-Hazil y Anexos. The specific questions answered are: How many people in the village hunted? How many individuals and what kinds of wildlife were taken by hunters? How much wildlife did individual hunters take? When, where, and how was wildlife taken? What was the catchment area of the hunting? Were there temporal or seasonal differences in the frequency at which wildlife was taken? In what kinds of habitat were wildlife taken? This information on wildlife use will be compared with estimates of wildlife population densities presented in Chapter 4. In Chapter 4, 1 discuss wildlife densities in early secondary forest (current and recent agricultural areas) versus late secondary forest (areas not used for agriculture for at least 50 y) in the study area. The specific questions answered are: How many individuals and what kinds of wildlife were sighted? In what kinds of habitat was wildlife observed? 11 Were there temporal or seasonal differences in the frequency at which wildlife was sighted? This information will give a measure of the number and kinds of wildlife potentially available to hunters, and will be compared to wildlife use as described in Chapter 3. In Chapter 5, I discuss wildlife food habits and evaluate the importance of gardens and domestic crops in wildlife diets. The importance of gardens and crops will be analyzed with respect to wildlife use by hunters (Chapter 3) and wildlife availability (Ch^ter 4). The specific questions answered are: WTiat kinds and what parts of plants were eaten by wildlife? Were there seasonal differences in wildlife diets? Were cultivated crops important as food to wildlife? This information will be used to determine if wildlife is attracted to domestic crops or wild plants that occur in gardens as a result of disturbance (e.g., palms). In Chapter 6 I discuss characteristics of Mayan gardens and their uses by hunters and wildlife. The specific questions answered are: How large were gardens? What crops were planted in gardens? In what habitat types and at what distances from the village were gardens located? When were gardens cleared, planted, and harvested? This information will be used to explain wildlife availability (Chapter 4) and wildlife food habits (Chapter 5). In Chapter 7, I summarize the results of this study with respect to the model of garden hunting and offer a series of conclusions. I focus on the main premises of Linares (1976): one, wild animals used gardens; two, hunters had to focus their harvest of game on those species of wild animals that used gardens; three, the densities of these wild animals were greater in the vicinity of gardens than in forests without gardens; and four, these wild animals had to consume crops from gardens. This study 12 was designed to test each of these premises. In closing, I will propose that strong conservation measures be enacted to protect the wild animals at X-Hazil Sur and the cultural traditions of the Maya hunters that use these species. Maya and Spanish Terms One of the challenges that accompany working in a different culture is being able to translate specific terms into your native language. During this study, many Maya and Spanish terms were learned for gardens, vegetation, and wildlife that did not have equal counterparts in English. In order to enhance clarity, maintain the Maya and Spanish sense of the term, and avoid lengthy and ambiguous English translations in this paper, several Maya and Spanish terms were retained and italicized (e.g., huaya). CHAPTER 2 NATURAL, CULTURAL, AND SOCIOECONOMIC ENVIRONMENT OF THE MAYA INDIANS IN QUINTANA ROO, MEXICO Introduction Subsistence hunting in the Neotropics, whether practiced by Maya Indians, peasants, or tribal peoples, varies in response to different natural and cultural factors. Each group, for example, hunts a wide range of wildlife, uses different techniques to locate and kill prey, and observes specific traditions in the use of wildlife and exchange of game. An understanding of these natural and cultural factors will help to explain specific hunting methods and results as well as help to analyze the variation in subsistence hunting practices by these people. A key additional consideration to understanding hunting is the nature and extent to which cultural factors break down when the group becomes involved with market economies. For this dissertation, I will analyze subsistence hunting by Maya Indians in the village of X-Hazil Sur, Ejido X-Hazil y Anexos, State of Quintana Roo, Mexico (the Spanish term ejido is used to desaibe a geographic and administrative unit in Mexico, see below). The following summary describes published accounts as well as personal observations during 1987 and 1989-1990 of the natural, cultural, and socioeconomic environment of the Maya at the local (village), state (Quintana Roo), and regional level (Yucatan Peninsula). While the personal observations were made primarily in the village of X-Hazil Sur, trips also were made throughout the region. The final section of this chapter is an extensive description of the study area. The term subsistence hunting, as contrasted with commercial hunting, will be used here to describe hunting by Maya Indians at X-Hazil Sur as game rarely is sold in markets. However, as many Maya hunters care for domestic animals or can purchase canned meat, this hunting has more of an opportunistic quality than, for example, subsistence hunting practiced by Yuqui Indians in Bolivia (cf., Stearman. 1989) or Siona-Secoya Indians in Ecuador (cf , Vickers, 1988). 13 14 Natural Environment Quintana Roo (50,843 km^ is located along the east coast of the Yucatan Peninsula, between n°49'-2l°3T N lautude and 86°44'-89°24' W longitude (Figure 2-1; Escobar Nava, 1986). To the east lies the Caribbean Sea, to the west the Gulf of Mexico and the Mexican states of Yucatan and Campeche, and to the south lie the countiies of Belize and Guatemala. Ecological characterization of Quintana Roo is incomplete. Studies have been conducted on mammals, birds, reptiles, amphibians, vegetation, and soils, but much of this information has not been updated in 30-50 years. Recent studies, however, include a statewide, botanical survey (Cabrera Cano et al., 1982), an ecological assessment of the southern part of the state (Camarena-Luhrs and Salazar- Vallejo, 1991), and an analysis of the biological diversity and economic aspects of the Sian Ka"an Biosphere Reserve (Dachary and Amaiz Bume, 1989; Navarro L. and Robinson, 1990). The following summary provides information relevant to an understanding of the natural and cultural factors affecting subsistence hunting by Maya Indians in Quintana Roo. Geology and Physiography Quintana Roo is a flat, broad shelf of dolomite and limestone rock jutting north-northeast into the Caribbean Sea (Lee. 1980; Wilson, 1980). The Yucatan Peninsula rises gradually to a maximum of 350 m along the western border with Campeche (Paynter, 1955a, 1955b; West, 1964). Bedrock deposition began in the Eocene age (50 million years ago) and resulted in limestone and dolomite layers that are several hundred meters thick (Bonet and Butterlin, 1962 [not seen, cited in Wilson, 1980]; L6pez Ramos, 1973 [not seen, cited in Wilson, 1980]). Much of the surface limestone is thoroughly karsted; sinkholes and small caves are common. Quintana Roo has three physiographic subdivisions (Wilson, 1980). The Coba District, the northern half of the state, is a karsted plain with linear depressions, several large lakes, and an abundance of small hills and depressions. The Rio Hondo District, the southern half of the state, is 15 er 21- 20* 19"- □ Study Area Ejido X-Hazil y Anexos :;.:::: Slan Ka'an Biosphere Reserve •k Stale Capital • Cities/Towns International Boundaries State Boundaries ^ -Highways i8' 25 50 75 100 Km. ■ ^ 1 1 87' Figure 2-1. Location of Ejido X-Hazil y Anexos (study area), Quintana Roo, Mexico. 16 characterized by low relief and a number of linear fault depressions in which streams, lakes, and lake beds are located. This district also includes offshore islands of fault origin. The Ri'o Bee District, encompassing a small portion of the state along the western boundary, is an area of intermittent lakes; broad, conical hills; and high linear ridges. The drainage patterns of this distria are poorly developed. Hydrology Due to the low relief and limestone surface of Quintana Roo, riyers and lakes are rare, but the water table is high. Water frequently is found in wells today at depths of only a few meters (Wilson, 1980). In pre-Hispanic times, however, the Maya technology was insufficient for digging wells through the limestone so people often constructed cisterns to store water during the dry season (Killion et al., 1989). The Maya located their settlements near sinkholes that contained water during the entire year (Wilson, 1980). Below the rock surface of the Yucatan Peninsula, a layer of freshwater overlies a layer of saline water (Back and Lesser, 1981). Movement of the freshwater is not via underground rivers, however, as is the case in Florida. Rather, due to hydrostatic pressure generated by rainfall, the freshwater moves along underground rock fractures from the interior of the Yucatan Peninsula to each coast. As a result, the underground freshwater is widely available rather than being found only in isolated areas. The wide distribution of underground freshwater also has its drawbacks. Recently, geologists have questioned the wisdom of coastal developments for tourism or agriculture that might pierce the aquifer in the Yucatan Peninsula and allow freshwater and saline water to mix (Perry et al., 1989, 1990; Smart and Whilaker, 1990). Such mixing would be deleterious to wildlife, plants, and humans. There are few sources of surface water in the state for either people or wildlife. The only river in Quintana Roo is the Rio Hondo, which forms the border with Belize and drains into Chetumal Bay. There are about 34 shallow lakes in Quintana Roo (Escobar Nava, 1986). Sinkholes (cenotes) are widely distributed throughout the state and may measure up to 60 m in diameter (e.g., the "Sacred 17 Cenote at Chichen Itza, State of Yucatan; Wilson, 1980), but diameters of 10 m are more common. Many species of small fish occur in the caves, lakes, and sinkholes (Barrera, 1964; Lipske, 1990). Shallow depressions (including natural collapse features called aguadas and small pothole features called sartenejas commonly occur on the surface. These depressions fill with water during Uie rainy season and retain water through much of the dry season, thus providing natural sources of water for wildlife, plants, and humans. In response to the difficulties with obtaining drinking water from natural sources, the Maya of X-Hazil Sur now obtain their drinking water from personal wells near their homes as well as from underground pipes connected to the village water tank and supplied by a community well. These two sources are highly reliable, and water availability no longer is a problem, except during power outages or due to mechanical problems with the pumps. Water potability, however, is a separate issue as chemicals used to purify the water may not always be available or added as necessary to ensure water purity. As a results, village residents often suffered from diarrhea. Climate The climate of Quintana Roo is warm and humid. According to the Koppen terminology, the main climate category is Aw (Tropical Savanna; Garda, 1973). There are two, well-developed seasons; rainy and dry. The rainy season typically lasts from May through October (Sanchez Crispin, 1980; not seen, cited by Careaga Viliesid, 1990). The driest two or three months typically have less than 60 nun precipitation each (Wilson, 1980). Annual precipitation in the state is 900-1,600 mm and mean temperature is 20-25 °C. At the Sian Ka'an Biosphere Reserve (unmediately to die east of the study site), annual precipitation is 1,000-1,200 mm, while about 70 percent of the rain occurs during May-October (Ohnsted and Duran, 1991). The maximum temperatures occur during May-September, while die minimum temperatures occur during December-January. During the dry season, cold fronts (nortes) commonly move soudiward from the United States, causing diunderstorms, overcast skies, and substantially lower (ca. 10-20° change) temperatures (Wilson, 1980). These fluctuauons, however, 18 usually last only a few days at a time. Rainfall and temperature vary along a north-south gradient in the state, with more rainfall and lower temperatures in the south and less rainfall and higher temperatures in the north (Wilson, 1980; Escobar Nava, 1986). Hurricanes frequently occur in the Caribbean Sea and strike the Yucatan Peninsula (Walker et al., 1991). The impact of these storms on the people, plants, and wildlife depends greatly on the force, duration, time of year, and the path the storm takes. During 1952-1978, 46 hurricanes traversed Quintana Roo (about two per year), including Janet (striking Chetumal in 1955) and Carmen (striking Chetumal in 1974; Escobar Nava, 1986). Hurricane Gilbert, the last major hurricane to strike Quintana Roo, raged over the Yucatan Peninsula during 14-15 September 1988 and affected the human residents as well as the wildlife and associated habitats (Anonymous, 1988; Wilder, 1988). Between Canciin and Tulum Ruins, the rains, waves (about 6 m high), and winds (reaching 320 km/h) damaged roads, boats, buildings, beaches, power lines, and water pipes. Throughout the northern half of the state, lowlands were flooded, leaves were stripped from branches, and many trees were uprooted. This destruction was especially critical to Maya farmers throughout the region because their gardens were destroyed, but many species of wildlife that were crop predators also suffered due to the loss of crops. Throughout the northern half of the Yucatin Peninsula, immediate as well as delayed impacts of Hurricane Gilbert became apparent. In the days following the storm, biologists in Canctin noted large numbers of dead and dying bats in the streets. They attributed this to the disappearance of insects, the bats' main food (J. Jucirez G., pers. comm.). During the following dry season, in 1989, forest and grass fires raged across the Yucatan Peninsula and burned about 135,000 ha during four months (L6pez-Portillo et al., 1990). The impact of Hurricane Gilbert on the local people, plants, and wildlife in Quintana Roo has yet to be quantified, but it is considered to be one of the most severe storms ever (Lynch, 1991; Whigham et al., 1991). 19 Soils Soil physical and chemical properties directly affect the vegetation that can grow in an area. These properties indirectly affect people who plant gardens or harvest natural forest products, such as lumber. These properties also affect wildlife species that depend upon this vegetation for food and cover. Unfortunately, the classification of soil types in Mexico is problematic, and a new classification scheme recognizing additional categories, especially in the Yucatan Peninsula, is being prepared (Dunning, 1989). Presently, the soils of Quintana Roo are classified as belonging to the order Rendzinas, suborder Calcomorfics, according to the FAO classification, while according to the USDA classification they are categorized as belonging to the order MoUisols, suborder RendoUs (U.S. Department of Agriculture, 1975). Rendzina soils are shallow; have low fertility levels; can be red or black in color; have good structure and drainage; and have a high content of calcium, carbonates, bicarbonates, and organic matter (Aguilera, 1959). Soil color can vary gready between adjacent sites. Soil physical and chemical changes in gardens in the X-Hazil Sur area are similar to those in other tropical sites subject to shifting cultivation (Noguez-Galvez, 1991). In general, after 20 years of fallow the physical and chemical values of the soil stabilize at levels similar to those of soils in late secondary forest. X-Hazil Sur residents, however, now practice a short fallow cycle, replanting about every 5 y, and increasingly complain about decreasing crop yields. Noguez-Galvez (1991) conducted a study of soil characteristics in the same area as the present study, and reported the following selected soil chemical and physical properties of garden sites for 1 and 70 y of fallow at 0-5 cm of depth: Soil chemical properties: Organic matter (8.08% and 10.49%, 1 and 70 y, respectively). Total nitrogen (0.437% and 0.493%), PO4 (12.16 ppm and 16.31 ppm), Ca^+ (59.48 me/lOOg and 41.91 me/lOOg), Mg"^^ (3.58 me/lOOg and 3.42 me/lOOg), Na^ (1.04 me/lOOg and 0.96 20 me/lOOg), K* (3.11 me/lOOg and 2.37 me/lOOg), pH KCl (7.20 and 6.60), and pH HjO (7.78 and 7.32). Soil physical properties: Field edacity (52.39% and 57.23%), wilting point (21.59% and 22.55%), available water (30.77% and 33.54%), and bulk density (0.86 g/cm^ and 0.76 g/cm'). These results suggested that the soils of the area can support horticulture, but that the present period of only a few years that Maya farmers leave the land to fallow will result in serious soil degradation. A fallow period of 20 y would allow soil physical and chemical properties to return to normal levels. Below the soils of the Yucatan Peninsula there is hard surface limestone (Wilson, 1980). Widespread, especially in the north, this rock is formed by solution and precipitation of calcium carbonate near the ground surface. Plant roots break up the cemented grains and shell fragments into large chunks that litter the surface. A material called sascab occurs below the hard surface layer (Wilson, 1980). Sascab is nearly pure calcium carbonate and forms as the surface limestone weathers. Friable, sascab is commonly mixed with cement by Mexican workers to make roads and building foundations. Species Richness Species richness is the number of species that occur in a specific geographical area without regard to their density or abundance. Among Mexican states, the species richness of Quintana Roo is relatively low, containing only about 20-30% as many species as the states with greatest richness (Flores-Villela and Gerez Fernandez, 1989). About 151 species of vertebrates and 1,257 species of plants occur in Quintana Roo. Vertebrate endemism is low (five species), while plant endemism is 21 estimated to be high. Species richness of Quintana Roo is enhanced by the fact that 79% of the natural habitat is categorized as unaltered and has not been significantly perturbed by recent human activity. Flora Several attempts have been made to characterize the vegetation of Quintana Roo. In describing and classifying the Yucatan Peninsula, Standley (1930), Lundell (1934, 1937), Leopold (1950), Paynter (1955a), Cabrera Cano et al. (1982), and Rzedowski (1986) generally recognized from three to six vegetation types. Tropical semi-evergreen forest is the dominant type in the Quintana Roo, occurring in the south and east, while tropical semi-deciduous forest is less dominant and occurs in the north and west (Rzedowski and Calder6n de Rzedowski, 1989; Figure 2-1). Phytogeographically, the Yucatin Peninsula is more similar to northern Central America than to southern Mexico (Estrada-Loera, 1991). The flora of central Quintana Roo is characterized by a medium-height forest that is 7-25 m tall (Cabrera Cano et al., 1982). The forest floor has boUi woody and herbaceous plants but little soil or organic matter. Lianas and epiphytes are abundant and are supported by a large number of trees. The dominant trees include the following; Brosimum alicastrum, Bursera simaruba, Manilkara zapota, Metopium brownei, Psidium sartorianum, and Vitex gaumeri. The State of Quintana Roo has a long history of logging and shifting cultivation (Edwards, 1986). Although litde information about deforestation rates is available for the state, Flores-Villela and Gerez Fernandez (1989) determined that only 21% of the natural habitat had been significantly perturbed or was in the process of being transformed. Since 1957, commercial logging has been managed through cooperative agreements between the ejidos and the State Forestry Program (see Snook, 1993). Fauna Mexico is among the richest countries in the world in terms of the number of autochthonous animal species (Ramamoorthy et al., 1992). Within Mexico, however, Uie fauna of the Yucatan 22 Peninsula differs from other areas in that species endemism and diversity are lower than in the rest of the country. In addition, species distribution and abundance patterns for mammals and herps suggest that taxa in the northern part of the peninsula should be grouped into the Yucatan Biotic Province, while those in the southern part should be included with the Middle America taxa (Lee, 1980; Dowler and Engstrom, 1988). Characteristics of Selected Mammals and Birds Human cultural traditions and the distribution and abundance of animals determine which wildlife species are considered as game in an area and which are not. Manunals and birds were the main taxa of interest in this study. Excluding bats, insectivores, small rodents, marine mammals, and endemic species on offshore islands, there are about 40 species of mammals and 15 species of birds that are potential game taxa for Maya Indians in Quintana Roo (Chavez Le6n, 1980, 1983; Gaumer, 1917; Leopold, 1972, 1977; Ramirez Pulido et al., 1982; Ramirez Pulido et al., 1983; Appendices A and B). The following summary is based on the literature, my personal observations, and information provided by Maya hunters. Marsupials were not taken as game by mestizos or Maya hunters but were killed as they preyed upon chickens and ate pineapples (Ananas comosus) and other fruits in house gardens. These animals were viewed as pests (M. Cab Cohuo, pers. comm.). Primates were not taken as game by Maya hunters in Quintana Roo, but in the past, adult females occasionally were killed so that their young could be sold as pets, especially spider monkeys {Ateles geoffroyi). Spider monkeys and howler monkeys (Alouatta pigra) are uncommon in the forests and rarely are taken by hunters (Navarro L. et al., 1990; Watts and Rico-Gray, 1987). These primates reportedly were much more abundant and widespread in the past (J. Poot Cruz, pers. comm.). Edentates were not taken as game. Armadillo (Dasypiis novemcinctus) meat, although regularly consumed by several other indigenous groups in the Neotropics (Redford and Robinson, 1987), was not consumed by the Maya because it contained muscle tumors. Hunters reported that their 23 parents and grandparents frequently ate armadillo meat in the past as they knew how to remove the tumors (J. H. Balam Xiu, pers. comm.)- Tamanduas {Tamandua mexicatia) often were killed by motor vehicles at night along Mexico Route 307. The single lagomorph species reported for western Quintana Roo was not present in the study area (Dowler and Engstrom, 1988; Jones et al., 1974b). Three of the six rodents were relatively common in horticultural areas and regularly were taken as game; paca {Agouti paca; ca. 8.2 kg), agouti (Dasyprocta punctata; ca. 3.6 kg), and pocket gopher (Orthogeomys hispidus; ca. 0.4 kg). The biology of these species in the Yucatan Peninsula is poorly known (Jones et al., 1974a). Several species of carnivores occurred in the study area (Bimey et al., 1974; Genoways and Jones, 1975). The coa.ti (Nasua nasua; 3.8 kg) was an important game species as well as a troublesome crop predator. Coatis in large groups of up to 30-40 adult females and their young frequently entered gardens and caused extensive damage to the com. Hunters reported taking pumas {Felis concolor) and jaguars (Panthera onca) for their skins, teeth, and meat. The ocelot (Felis pardalis) and margay (F. wiedii) also were taken for their skins. Only two hunters reported taking kinkajous {Potos flavus) and raccoons (Procyon lotor). Raccoons often were shot as they consumed com seedlings, but the carcass usually was left by hunters to rot. The tapir (Tapirus bairdii) was the largest mammal in the study area at about 275 kg, but hunters did not report taking this species. Hunters indicated that tapirs were hard to kill due to their thick skin and that the meat was not especially tasty. Tapirs occurred in the flooded lowlands in the eastem part of the study area, but were uncommon in Quintana Roo (Sanchez-Herrera et al., 1986). There were two species of peccaries and two species of deer in the study area, and all four taxa of these ungulates were major game species (Bimey et al., 1974; Genoways and Jones, 1975). Both species of peccaries occurred in groups that ranged widely over gardens and forested areas. The collared peccary (Tayassu tajacu; ca. 17.5 kg) was smaller, but more abundant than the white-lipped peccary {T. pecari; ca. 28.6 kg; Aranda Sanchez, 1981; March M., 1987). The brocket deer (Mazama 24 americana; ca. 26.1 kg) and the white-tailed deer (Odocoileus virginianus; ca. 40.0 kg) frequently were observed in or around gardens. The main game bird species taken by mestizos and Maya hunters at X-Hazil Sur include the tinamou (Tinamus spp. and Crypturellus spp.), great curassow (Crax rubra), plain chachalaca (Ortalis vetula), and ocellated turkey (Agriocharis ocellata; Griscom, 1926; Leopold, 1977; Chavez Le6n, 1983; L6pez Omat, 1991; Lopez Omat et al., 1989; MacKinnon Vda. de Monies, 1989). Other species taken for meat, as pets, or because they are crop predators include the red-billed pigeon (Columba flavirostris), scaled pigeon (C. speciosa), parrots (Amazona spp.), and parakeets (Aratinga spp.). The Maya of X-Hazil Sur, like other indigenous people in the Neotropics, take species of wildlife other than birds and mammals. For example, the Maya consume honey produced by wild and cultivated bees (Apidae; Chemas and Rico-Gray, 1991; Roubik et al., 1990). Honey is an important food item for several indigenous groups in the Neotropics (Posey, 1984; Posey and Camargo, 1985). The Maya also consimie modest quantities of fish taken from local sinkholes during the dry season. About 30 species of freshwater fish have been reported from sinkholes in Quintana Roo, but only two species are important for subsistence purposes (Rhamdia guatemalensis [Punelodidae] and Astyanax fasciatus (Characidae; Gamboa-Perez, 1991; Wilkens, 1982 [not seen, cited in Gamboa-Perez, 1991]). Fish also are widely consumed by indigenous groups in the Neotropics (Cameiro, 1970 Chemela, 1982, 1985). The Maya of X-Hazil Sur, unlike many other indigenous groups in Mexico and throughout the Neotropics, do not consume insects or eggs or meat from snakes, terrestrial turtles, or lizards (Ctenosaura similis). This is in contrast to western Mexico, where lizard (C. pectinata) eggs and meat are consumed by local residents (Parra Lara, 1986; Santana et al., 1990). Maya hunters at X-Hazil Sur, however, recall that 10-20 years ago. along the east coast of Quintana Roo in what is now known as the Sian Ka'an Biosphere Reserve, Maya hunters harvested reptile products for commercial purposes. These hunters took hides from crocodiles {Crocodylus moreleti) and eggs and scutes from 25 sea turtles {Chelonia mydas, Eretmochelys imbricata, and Demwchelys coriacea; J. Cab Can, G. G6mez Puc, and J. Foot Ake, pers. conim.). These wildlife products were sold to intermediaries and eventually were purchased by tourists or business people who fabricated articles made from these products. Study Area The study area was located on the Ejido X-Hazil y Anexos, one of the largest ejidos in Quintana Roo (Figure 2-1). The population of Ejido X-Hazil y Anexos during the study was about 1,680 people distributed in the three villages of X-Hazil Sur, Uh-May, and Chancah Veracruz (Dachary and Amaiz Bume, 1989; Dr. Juan Chi, pers. comm.). Ejido residents were primarily Maya and lived in the villages, except for about 10 people who lived on a semipermanent basis at farms or camps {ranches and campamentos) in the forest. The ejido is composed of the direct descendants of the Maya who fought Mexican soldiers for more than 50 y during the Caste War (Reed, 1964; see below). While Dachary and Amaiz Bume (1989) suggested that the population of the ejido had declined, local officials (D. Ake Ayala, pers. comm.) indicated that the ejido was growing in population. At about 3.0 persons/kmS the human population density on the ejido was greater than the density on many other nearby ejidos. For example, Ejido X-Maben had a density of only 0.8 persons/km^ (Murphy, 1990). The three villages that compose the ejidos were founded about 1915 after a smallpox epidemic. Deaths due to this epidemic occurred throughout the region as well and forced many residents to abandon their homes to seek new places to live (Villa Rojas, 1987:149). The presence of a few small stone structures on the ejido suggests that the area was populated by the early Maya, but these structures have not been studied by archeologists. Local residents suggested, however, that the area had a relatively recent occupation dating from the early 1900s when four or five families founded a small village near the site of Rancho Las Palmas (Kilometer 95, Route 307; and S. Yeh Ake, pers. conmi.). These same families eventually founded the ejido. 26 Due to the large size of the ejido (55,295 ha) it was decided to focus the study on the community of X-Hazil Sur (19°23'30"N, 88°05'00"W), the largest of the three villages. In 1992 the population of X-Hazil Sur was 1,040 (Dr. Norbierto Ramirez Morales, in litt.; 539 men and 501 women; Figure 2-2). About 52% of the men and 47% of the women were between the ages of 15 and 59, while only about 5% of the total population was > 60 years old. During 1989-1992 the population of X-Hazil Sur grew from 950 to 1,040, but this reflected emigration and inmiigration, as well as births and deaths. Temperature and precipitation were measured at X-Hazil Sur during July 1989-December 1990 (Figure 2-3; Appendix C). The highest temperature recorded was 39°C on 28 and 29 May 55 SO ■45 <40 35 BO 25 20 15 lO O =- «0 y - 59 y - s-* y - A9 y - -*-* y - 39 y - 3A y - 29 y - 2-4 y - 19 y - 14 y - 9 y + -- -- + -- -- •40 30 20 lO 9b Bi<««le» lO 20 '9E> Female* 30 •40 Figure 2-2. Distribution of X-Hazil Sur residents by age class and sex. There were 539 males and 501 females during the census of 1992. 27 TEMPERATURE (C) PRECIPITATION (MM) 300 1 r J A S O MONTH 1— MEAN MINIMUM TEMP PRECIPITATION -B- MEAN MAXIMUM TEMP • - PARTIAL MONTH Figure 2-3. Mean monthly temperature and precipitation at X-Hazil Sur, Quintana Roo, Mexico, during July 1989-December 1990. 1990, while the lowest temperature recorded was 9°C on 27 December 1989 (temperature was measured by a thermometer ±1C° and read daily at 0700 h to obtain the minimum and maximum temperatures for the preceding 24 h). Rainfall during 1 January-12 December 1990 was 1,277.3 mm, slightly above average (rainfall was measured by a rain gauge and read daily at 0700 h). Local residents reported that during the past several years it had been drier than normal and that the rainy season had begun 1-2 months later than normal. The ejido was composed of four major land-use and vegetation types: Plots & Gardens (6.07% of ejido). Early Secondary Forest (5.18%), Late Secondary Forest (88.52%), and Other (0.23%; Table 2-1). Sites categorized as Plots & Gardens mainly were located near the three villages or along roads, while sites categorized as forests usually were located relatively far from the roads (Figure 2-4). The ejido was accessible through many types of thoroughfares. These roads and trails greatly influenced subsistence and economic activities in the ejido. The main road was Mexico Route 307, a hard surface 28 Table 2-1. Land uses and vegetation types at Ejido X-Hazil y Anexos, Quintana Roo, Mexico.' Land uses and vegetation types Area (km-) % Total Plots & Gardens 33.56 Subtotal 33.56 6.07 Early Secondary Forest Former gardens 26.44 Cattle corrals & pastures 2.18 Subtotal 28.62 5.18 Late Secondary Forest & Low Moist Forest Medium height, semi-evergreen forest 418.74 Low height, semi-evergreen forest 63.18 Seasonal wetlands 7.21 Ponds & Sinkholes 0.35 Subtotal 489.48 88.52 Other" 1.29 Subtotal 1.29 0.23 Total 552.95 100.00 ' The area was determined by using a compensating polar planimeter to measure land uses and vegetation types on a 1987 map (1:25,000 scale) produced by the Plan Estatal Forestal. The map was based on aerial photographs (Institute Nacional de Estadistica Geografica e Informatica, February 1985, 1:37,000 scale), topographic maps (Institute Nacional de Estadistica Geogralica e Informatica, 1987, 1:50,000 scale). *" The category "Other" was composed of village sites and previously forested areas (rodales) where logs were sorted, trimmed, and loaded. road that extended from north to south. Secondary roads connected X-Hazil Sur and Chancah Veracruz to Route 307. Most gardens were located within about 2 km of these roads. The ejido was further 29 VF^FTATIOfJ TYPFS \ ^ Plote and Gardens \ TOWNS A Chancat) Veracniz Eaity Secondaiy Forest I • Uh-May [•'•J Fonner Gardens I ■ X-HaiilSur ^ Cattle Corrals and Pastures \ Late Secondaiy Forest 1 1 Medium Height, Semi-Evergreen Forest [\\i Low Height. Semi-Evergreen Forest^^^J^ '^p^i^oo^ 0 Seasonal Wetlands fi^^^^i u ^ 1 Ponds and Sinkholes ^'^"WwS^i^ ^^''^ M °^' , -' " «<~ ^.^aher Roads j pv /f??^*:^,^^^ / A y ^Msfi )p^^ / e> ^i-/K^Siu:-V/ y v/-::;:;:i-^;N'^:r' •'■W-. .£> l 2 cm dbh). The most abundant species in Early Secondary Forest were nees {Gymnanthes lucida, n = 94 trees, dbh range = 2-10 cm, x dbh = 5.3 cm), wild grape {Coccoloba cozumenlensis, n = 59, dbh range = 2-5 cm, x dbh = 5.11 cm), and white gombolimbo (Dendropanax arboreus, n = 55, dbh range = 2-16 cm, x dbh = 4.95 cm). Cultural and Socioeconomic Environments The Maya of X-Hazil Sur are a highly acculturated people, having changed much over the past two-three generations, but they maintain many traditional customs. For example, most residents of this village speak Spanish in addition to Maya, wear Western clothing, own a radio or television, and 31 regularly travel by bicycle or motor vehicle to adjacent towns and villages. However, many of these same individuals also observe special rituals as they butcher game, construct rudimentary dwellings, and plant and harvest their gardens. They also believe in spirits (e.g., San Juan and San Miguel), and commemorate their dead, as they have done for hundreds of years. While some of these practices may appear to follow the Catholic religion, they have their basis in traditional Maya beliefs. The result is that the current cultural environment is a hybrid of traditional and modem beliefs (Careaga Viliesid, 1990). Information about Maya Indians and the Yucatan Peninsula has been compiled since the early 1500s. Among the first European visitors who published descriptions about the area were priests, explorers, and government officials (cf.. Cook, 1769 [cited in Villa Rojas, 1987]; Davila, 1870 [cited in Villa Rojas, 1987]; Landa, 1978; Stephens, 1963, 1969). While these authors were amazed by the size and complexity of die large ceremonial centers encountered throughout the region, Uiey considered the local Maya residents to be backward and ignorant. With respect to the area itself, these writers described the Yucatan Peninsula as an extremely harsh environment. Accounts about Maya culture published since die early 1900s have described a sophisticated society. Linguists and chroniclers, for example, have documented the important role Uiat Maya cosmology has in Maya oral history and daily conversations, including those between Maya and foreigners (Bums, 1973, 1977, 1980, 1983; Hanks, 1990; Sullivan, 1987, 1989). For example, a series of stories may include references to animal spirits, Uie creation of the world, and an epoch of miracles when half a kernel of com was sufficient to make tortilla dough for a meal for 6 or 8 people. The conversation then could switch to comments about the Queen of England, accounts of slavery in the United States, and analyses of recent world wars. From Uiese accounts it is clear that Maya oral histories and conversations contain a rich mixture of references to past, present, and future events. History and cosmology also mix references to actual peoples with Uiose to spirits and deities. Archeologists also have documented the complexity of Maya subsistence practices, settlement patterns, and the chronology of large ceremonial centers (Andrews, 1942, 1960, 1965, 1973; Lizardi 32 Ramos, 1939 [cited in Villa Rojas, 1987]; Lothrop, 1924; Pollock, 1940). Recent archeological studies, however, have challenged traditional views of a cultural elite supported by simple fanners and concluded that the Maya had well-populated cities supported in part by intensive horticulture (Sabloff, 1991). Anthropologists have desaibed the dynamic nature of Maya daily life. Maya grammars and ethnographies, for example, have described a culturally rich people with a written language and extensive knowledge of useful plants and animals around their villages (Gann, 1918, 1926, 1935; Morley, 1927, 1938, 1956; Redfield and Villa Rojas, 1962; Steggerda, 1941; Thompson, 1930; Tozzer, 1907, 1921; Villa Rojas, 1969, 1987). In addition, the Maya have adapted traditional horticultural practices to new environmental conditions as they occupy new areas (e.g., Guatemala; Carter, 1969). Studies also have shown that Maya women are taking a more active role in the political and economic aspects of daily life (Elmendorf, 1976). These examples clearly show that the vibrant nature of contemporary, Maya daily life. In spite of the numerous accounts over the past 150 years, research results published during the past 5 years are forcing investigators to rethink many of their ideas about the historical Maya described from archeological sites. The traditional image of Maya culture, as indicated above, is a series of elaborate ceremonial centers with a peaceful, ruling elite supported by a peasant class practicing hunting, gathering, and shifting cultivation. New interpretations of hieroglyphic inscriptions suggest that the Maya engaged in extensive armed conflict (Marcus, 1991; Scheie, 1991). Other studies have confirmed the presence of complex cities, intensive horticulture, and sophisticated religious symbolism (Hansen, 1991; Rice, 1991; Tolstoy, 1991). This has lead to an expanded view of the historical Maya culture that now appreciates the role and activities of the elite, as well as the nonelite, in daily life and recognizes that the historical Maya were able to manipulate their environment in order to sustain intensive horticulture and avoid environmental degradation (Andrews, 1991; Harlow, 1991; Sabloff, 1991; Turner, 1991). 33 Historical Setting There is some debate about when people first inhabitated Mesoamerica. While some would argue that the earliest settlers in Mesoamerica arrived about 25,000 B.C. (Adams, 1991), others scientists would contend that about 7,000 B.C. is more probable (Nesbitt, 1980). The Maya Indian civilization developed during 1500 B.C.- 150 A.D. (Formative Period). By about 800 B.C., the Maya had founded the city of Dzibilchaltun, near Merida, and several other smaller villages along the coasts of the Yucatan Peninsula and south to Belize, Guatemala, Honduras, and El Salvador (Andrews, 1960, 1965; Deeveyetal., 1979). During 150/300-650/900 A.D. (Classic Period), the Maya increased in number and developed an advanced culture that included a calendar, number system, network of roads, written and spoken language, trade routes that reached from central Mexico to Panama, and a system of horticulture that used raised gardens irrigated by canals (Denevan, 1970; Flannery, 1982; Pohl, 1985; Sabloff, 1991; Turner and Harrison, 1983). At the height of their power during 1250-1519 A.D. (Late Postclassic Period), die Maya in die northern part of the Yucatan Peninsula numbered about 800,000 inhabitants (Clendinnen, 1987; Cook and Borah, 1974). Perhaps the Maya are best known for their large ceremonial centers, including Tikal (Guatemala), Coba and Tulum (Quintana Roo, Mexico), and Uxmal and Chich6n-Itza (Yucatin, Mexico; Lothrop, 1924; Stephens, 1963; Thompson, 1966; Thompson et al., 1932, 1940). During 650/900-1250 A.D., for reasons Uiat still are being debated, the Maya civilization declined in size and power (cf., Adams, 1977, 1991; Cowgill, 1962; Culbert and Rice, 1990; Morley, 1956; Thompson, 1966; Willey and Shimkin, 1973). In 1511, the Maya were first contacted by Europeans— albeit accidentally— when some Spanish sailors were shipwrecked at a site called Las Viboras along die norUiem coast of the Yucatan Peninsula (Landa, 1978). Several skirmishes ensued in die following years as odier Spanish sailors attempted to rescue Uieir comrades and explore die coast. In 1527, die Spanish, who were searching for slaves and wealth, began their conquest of die Yucatan Peninsula, and by 1542 had founded die city of Merida. 34 During the following 300 years, the Spanish expanded their power along the northern coast, but various Maya chiefs retained control in the interior of the Yucatan Peninsula (Reed, 1964; Villa Rojas, 1987). Due to harsh treatment of the Maya as the Spanish expanded their power and presence in the Yucatan Peninsula, relations between the two groups were poor, and in 1847 the Maya began a rebellion, subsequently called "The War of the Castes" (Reed, 1964). The severest fighting lasted from 1847 to 1855, during which about 57,0(X) people perished (about 17% of the total population of the Yucatan Peninsula; Cook and Borah, 1974). Maya farmers armed with guns and machetes eventually held 80% of the Yucatan Peninsula and almost took Merida, by that time the prosperous Spanish capital of the Yucatan; but these men stopped fighting "temporarily" in order to return to their homes and plant their gardens (Reed, 1964). This decision by the Maya to postpone fighting was a major error and allowed the Spanish to regroup. Eventually the Spanish prevailed, but isolated cases of resistance by the Maya, primarily in the area of Felipe Carrillo Puerto (Quintana Roo), continued until the early 1900s (Escobar Nava, 1986; Reed, 1964; Sullivan, 1989). Quintana Roo was declared a territory of Mexico in 1902 and state in 1974 (Escobar Nava, 1986). According to a 1902 estimate, the population of the Quintana Roo was about 5,000 people (Dachary and Amaiz Bume, 1983). By 1970, just prior to the modem period of growth and development, the population of Quintana Roo was 88,150 inhabitants, with about 55.4% of the population bom in-state (mostly Maya Indians) and 43.4% bom out-of-state (mostly mestizos; Dachary and Amaiz Bume, 1984). Largely ignored by the Government of Mexico until the mid 1950s, the Maya and mestizos of the state subsisted mainly by hunting, gardening, extracting lumber from the forest, and harvesting chicle latex from sapodilla (Manilkara zapota) trees. The cultural and physical isolation experienced by residents of Quintana Roo lessened in 1955, after Hurricane Janet struck the Yucatan Peninsula and caused major damage to the Chetumal area in southem Quintana Roo. After receiving the damage reports, government officials and private citizens from throughout Mexico became aware of the poor social and economic conditions of the state. 35 Eventually several social and economic programs, including the construction of roads, schools, and hospitals, were undertaken to improve the quality of life in Quintana Roo (Escobar Nava, 1986). By 1983, the population of Quintana Roo had increased to 330,813 inhabitants and reflected much recent growth and development in the state. Chetumal (75,113 inhabitants), in the southern part of the state, has prospered as the state capital and as a thriving commercial center for trade with Belize. In the north, the cities of Cancun (about 81,000 inhabitants) and Cozumel (23,224 inhabitants) have flourished since the mid 1970s as resorts for national and international tourists (Dachary and Amaiz Burne, 1984). Today, less than 40% of the population of Quintana Roo lives in rural areas, in administrative units called ejidos (described below), and still practices traditional subsistence activities (Dachary and Amaiz Bume, 1984). Ejido System in Mexico The village of X-Hazil Sur is a part of Ejido X-Hazil y Anexos and is located about 25 km S of Felipe Carrillo Puerto (Figure 2-1). The ejido system was instituted in Quintana Roo in 1928 to allow Maya Indians and other rural people to have title to specific tracts of land around their villages in areas that previously were considered govermnent lands. These tracts were held in communal ownership where the residents could live and practice their subsistence activities. Previously, government officials sold concessions on these lands to outside business people without compensating local residents; a great deal of resentment among local residents resulted. The objective of the new system was to improve relations between business people, local residents, and government officials by giving local people control over their lands. Ejidos are properties, either owned by the government or expropriated from private owners, where landless campesinos (subsistence fanners) have usufruct to the area (Gordillo, 1988). In this system, the residents do not have title to individual plots of land within the ejido; rather, they enjoy control over a plot as long as they live on it or continue to work it. Certified residents are called ejidatarios (adult men who are certified legal residents of the area and widows voting on behalf of their 36 deceased husbands). If an ejidatario ceases to live on a plot or to work it, other residents are free to use that site without obtaining permission from any local official or the previous user. The creation of ejidos resulted in a mechanism that allows residents to manage their timber resources and distribute among themselves the profits from any contracts with outside business people. While timber contracts are based on tree inventories conducted by government foresters and approved by the Ministry of Agriculture (Secretaria de Agricultura y Recursos Hidraulicos, SARH), ejido residents conduct tree replanting programs and are free to negotiate prices with potential buyers (M. Carre6n Mundo, pers. comm.). Ejido X-Hazil y Anexos is governed by three officials; mayor (comisariado), secretary/treasurer (secretario), and enforcement officer (inspector). These officials handle the formal affairs of the ejido and are selected by villagers at elections every 5 years. None of these officials can be re-elected. An additional official, the village delegate (sub-delegado), handles local legal matters (e.g., the issuance of birth and death certificates) and is selected by state government officials located in the town of Felipe Carrillo Puerto. Other local individuals supervise work projects on the ejido, such as the collection of chicle and the extraction of timber. These officials also are elected by ejidatarios and an effort is made to rotate these positions among qualified candidates as the coordinators organize work crews, oversee projects, and receive a small salary. An important aspect of the ejido system is that nonresidents are not allowed to exploit ejido resources, such as timber or game, without the permission of ejido residents. Permission depends upon a majority decision and is rarely given. In this way ejido residents maintain control over their local resources. Ejido X-Hazil y Anexos was created by a decree from the state governor in 1941 and enlarged by presidential decree in 1942 (Dachary and Amaiz Bume, 1983). Over the years, areas have been added or removed from the ejido to accommodate the formation of new ejidos. The present size of the ejido is 55,295 ha (Dachary and Amaiz Bume, 1989). 37 Ejido members must be Maya in order to be ejidatarios. Previously ejido membership at X- Hazil y Anexos was granted automatically to all young men (about 16-18 years old) upon application to the comisariado and after fulfilling the requirements of planting a garden, maintaining a residence at the ejido, and collaborating on community work projects (faenas). These members then shared in the profits of timber sales by the ejido. Presently, the ejido has about 400 members. New members no longer are being accepted because current ejidatarios voted to restrict membership after per c^ita profit shares became unacceptably small. Today, as young men come of age, they still may live and work on the ejido but cannot share in any distribution of timber sale profits. Social Organization The social organization of indigenous groups is affected by many factors. One important factor is the ability to obtain food for personal consumption (Harris, 1974). Many indigenous people also share or exchange game for social and nutritional reasons (Stearman, 1989). In some cases, these exchanges extend to all group or village members, while in other instances, the sharing is limited to the immediate family of the hunter. Maya hunters are known to share or exchange game, but the importance of game to Maya social organization has not been studied. Maya households at X-Hazil Sur today vary in size and composition. Nuclear family households (husband, wife, and children) are the most common type, composing about 75% of the households, and are similar to those described by Redfield and Villa Rojas (1962), Webber (1980), and Villa Rojas (1987). Multiple family households (each family a separate economic unit; e.g., an adult man and his immediate family living adjacent to his parents with father and adult son working together, but dividing the crops) and extended domestic families (more than one family constituting a single economic and social unit; e.g., an adult man and his immediate family living adjacent to his parents with father and adult son working together, and sharing the crops) also are common. The typical household at X-Hazil Sur is composed of 8-10 people. Household size and composition may vary over time as children become independent and elderly adults become dependent as they no longer can care 38 for themselves. The size and composition of Maya households is important as members engage in the various subsistence activities necessary for survival. Family members, regardless of age or where they live at X-Hazil Sur, frequently share game and exchange food on special ceremonial days, but are not bound by tradition to share with nonfamily members. The close ties between family members affected the layout of the village through the location of houses. In X-Hazil Sur, family groups frequently formed neighborhoods composed of adjacent, but completely separate solares and houses that contained the grandparents, parents, and adult children. This close physical proximity of houses to each other facilitated inter-family closeness, but also increased friction between relatives during family feuds. Given the cultural traditions of the Maya, with the family and kinship being highly important, and the great difficulties associated with subsisting in Quintana Roo, it is not unusual for there to be few uimiarried adults or abandoned children at X-Hazil Sur. During 1989-1990, there were several single adult men (primarily young men or widowers), while there was only one adult woman (about 25 years old) who was single and unmarried by choice. There were, however, several adult men and women who previously had been married, but subsequently had separated or divorced. Some of these adults continued to live alone in their houses (usually the men), while others returned to the household of their parents (usually the women). In households where a divorce, remarriage, or death of a parent had occurred, the children often were sent to live with their aunts or uncles, but never abandoned. Some widows and widowers continued to live with their children, while others returned to hve with their parents or other family members. Divorced adults frequently found new partners. Thus, by being flexible, both children and adults were able to obtain support from friends or relatives. As young adults mature at X-Hazil Sur, they may continue to live with their parents or leave and establish new households. Unmarried adults usually remain with their parents. After getting married, an adult male frequently builds a new house for himself. This house generally is located in a lot near the house belonging to his parents or other male relatives, such as an uncle or brother. Young men usually marry at about 20 years of age. Men traditionally seek younger spouses in the same 39 village. A young man, recently married, typically will hunt and work with his father, father-in-law, or other adult male relatives for 1-2 years until he accumulates enough resources and experience to become economically independent. Maya men and women at X-Hazil Sur and throughout Quintana Roo have separate roles in the household. Men hunt, plant gardens, harvest chicle, extract lumber, provide firewood, maintain the house and yard, and care for goats and cattle in outlying corrals. Women attend to the house and the children, tend house gardens (cf., Forrest, 1991), and raise and sell domestic animals (e.g., pigs, turkeys, and chickens) in order to supplement family income. Some of these products are sold locally, while others are sold in the town market at Felipe Carrillo Puerto. Maya children at X-Hazil Sur often assist their parents with household chores. A young boy may accompany his father to the garden, but will do little manual work (e.g., planting, weeding, or harvesting) until he attains 10-15 years of age. Young boys, however, often are sent on minor errands, for example, to find firewood or to scare animals from the gardens. Boys usually don't begin hunting until about 15 years of age. Young Maya girls, on the other hand, experience a vastly different childhood than boys do. From the time she is about 5 years old, a young Maya girl assists her mother by helping in the kitchen, caring for younger siblings, and by bringing corn to a local store to be ground into meal. Girls don't hunt or clean game, but are expected to cook the meat and prepare tortillas after the men return. By about the age of 15 years, most Maya girls have learned their adult roles and are able to manage a house and care for a family, while most boys are just beginning to learn their corresponding roles. Social su-ucture in Maya communities is Yucatan has been a difficult subject among anUiropologists. While clear economic differences exist among individuals and families, these have not resulted in structured hierarchies of social classes. Redfield (1960) notes that the ideology of Yucatec Maya villages promotes egalitarian relationships. At the same time, the history of the 19th century Caste War in Uie area (Bums, 1977) suggests that social stratification along patrilineal family lines was an important feature of Maya political organization (Jones, 1977). One thing many authors have 40 noticed, however, is the pervasive importance of factions in village life (Re Cruz, 1992; Redfield, 1960). In short, village social structure in Quintana Roo sometimes appears to anthropologists as a classic case of a closed corporate peasant community (Wolf, 1957) where distinctions of wealth and status are downplayed. At other times, remnants of the Caste War political leadership system are used to structure social relations. Finally, factions and schisms rise and fall in villages, resulting in the appearance of economic stratification that might be just an accident of recent history. The Maya of Quintana Roo are not just Mexican peasants, as Foster (1967) and others describe. But neither are they pristine tribal peoples. They are villagers who have developed systems of social relations that exhibit community differentiation, but are not mirrors of Western social classes. The social hierarchy or stratification at X-Hazil Sur can be seen in men who occupy the major administrative positions and some families that tend to be more prosperous than other families. At X- Hazil Sur and throughout the Yucatan Peninsula, village officials are elected, usually for single terms of office, by ejidatarios. Single adult women at X-Hazil Sur are proscribed by community vote from independent or separate economic activities, such as operating a restaurant or having an individual lot for a house or garden in Ejido X-Hazil y Anexos. Although Maya villagers do not have a social hierarchy or stratification that mirrors western social classes, there are local individuals who perform special functions. For example, shamans cure illnesses, midwives deliver babies, and religious officials oversee spiritual matters. These people have undergone an apprenticeship locally and continue to serve as long as their services are requested by village residents. There are no shamans in the village of X-Hazil Sur, but there are several midwives (parteras), one male herbalist (yerbatero or curandero), and two men and one young woman who pray at religious ceremonies (rezador). These positions sometimes pass from parent to child after the appropriate training. Several other men direct specific activities at the church, and usually are assisted by their wives. The motivation for undertaking these activities vary. Some men indicated that it was their way of giving thanks to the Maya gods for curing an ill child or granting a bountiful harvest from 41 the garden. Other men indicated that it was a way to placate the Maya gods and avoid potential misfortunes. While these actions superficially resemble the practices of cargos (burdens) or promesas (promises) reported for highland peasant communities (Dumond, 1977), the Maya do not accrue prestige by these activities (Bums, pers. comm.). In adjacent villages, herbalists (both male and female) also treat various illnesses. While these people are treated with respect by village residents and receive a small compensation for their services, they are not afforded permanent or formal recognition and do not accrue prestige or political power over time. Primary school attendance is mandatory in Mexico. Students at X-Hazil Sur can attend grades K-8 locally. Secondary school attendance is voluntary. Students willing to pay tuition may attend school in Felipe Carrillo Puerto, about 30 minutes away by bus. In an effort by ejido residents to reduce the costs of attending secondary school, in 1990 a secondary school was opened at X-Hazil Sur where instruction takes place via television (telesecundaria). X-Hazil Sur residents also may attend post-secondary classes in nearby towns. Technical courses, such as accounting and hotel management, frequently are taught in Canciin, Chetumal, and Felipe Carrillo Puerto through various government programs. Increasingly, young adults are completing their education and obtaining jobs outside of X-Hazil Sur. Settlement Patterns Maya Indians in Quintana Roo presently live in a wide variety of villages, towns, and cities throughout the state. Most of the villages, such as X-Hazil Sur, have little commercial activity and are inhabited ahnost solely by Maya people. Since X-Hazil Sur is larger than other nearby villages, it also has a school, medical clinic, and satellite television system. Village stores are linked to those in towns by traveling salesmen who represent specific product lines, such as soda pop, bicycle parts, and groceries. The variety of merchandise, however, is limited. Towns, such as Felipe Carrillo Puerto, are larger than villages and have local government offices, important commercial activity, and are inhabited by both Maya and mestizos. Cities, such as Chetumal and Canciin, are the largest population 42 centers in the state and have Maya, mestizos, and large numbers of national and international tourists. Most of the Maya in Quintana Roo live in villages, however. According to the 1983 census, there were 2,792 population centers in Quintana Roo, with 90% containing fewer than 50 people each (Dachary and Amaiz Bume, 1984). The 11 largest population centers (> 2,500 residents) have a combined population of 214,935 inhabitants (65% of the total population). Maya at X-Hazil Sur live in two basic styles of houses. The traditional house is composed of poles and thatch, measures about 4 by 6 m, and is rounded on each end (cf., Redfield and Villa Rojas [1962], Webber [1980], and Villa Rojas [1987]). About 70-80% of the houses at X-Hazil Sur are of this type. Almost all houses have electricity (usually one outlet and one 60 watt bulb per house) and running water (usually one faucet located outside, near the front door of the house). Traditional houses may have either a dirt or cement floor, while the walls may be of poles or plaster, and the roof may be of tar paper or palm fronds (including Sabal yapa and Thrinax radiata). The modem house is of masonry, has windows, is rectangular in shape, and usually has a cement roof. The masonry house usually is less comfortable than the wooden house because air circulation is limited and the roof and walls tend to radiate heat after sunset. Only a few of the masonry houses have modem conveniences such as indoor plumbing or a kitchen with stove and running water within the house. Modem and traditional houses are laid out in a grid pattem, about 50-100 m on a side, and 4-6 families usually occupy a 1 ha block. Many households consisting of several nuclear or extended families have both types of houses. Subsistence and Economic Activities The Maya at X-Hazil Sur conduct numerous kinds of subsistence and economic activities. The main subsistence activity is to clear forested areas and plant a garden containing com (Zea mays; Ho loch in Maya), beans (Phaseolus spp.; Buul in Maya), squash (Cucurbita spp.; Kuum in Maya), and numerous other crops (Dachary and Amaiz Bume, 1983; Redfield and Villa Rojas, 1962; Villa Rojas, 43 1987; Webber, 1980). An important aspect of this practice is that most trees survive the clearing process and readily resprout (Rewald, 1989). This ability to resprout helps to facilitate the recovery of the forest after the garden is placed into fallow. Gardens typically are about 2 ha in size, located near the village, and are tended by adult men (see Chapter 6 for additional information about gardens). Women rarely assist the men in tending the garden, but young boys frequently begin to help when about 10-15 years old. Gardens usually produce a single harvest of com in December- January, but other crops may continue to produce for 1-2 years (e.g., camote [Ipomoea batatas; Is in Maya] and macal [Dioscorea sp.; Macal in Maya]). Men occasionally work for wages ($4-5/day) on a temporary basis, tending gardens or building houses locally. While tending a garden is not a daily, full-time task at X-Hazil Sur, the work requires careful planning and must be conducted in a sequence that closely conforms to the weather (Noguez-Galvez, 1991). The men clear (la tumba) the garden site during January-March and take advantage of the bright sun during April-May to dry out the bushes, ground litter, and felled trees. Just before the rainy season begins, in May or June, the men set fire to the site in a carefully controlled bum (la quema) that lasts 1-2 h. The men plant (la siembra) in May- July (depending upon when the rainy season begins) and weed (el chapeo) the garden one or two times during August-October. During October-November, the men double or bend over (doblar) the com stalks to facilitate drying and avoid predation by birds and coatis. The com harvest (la cosecha) occurs during December-January. Other crops are planted and harvested throughout the year. Maya women also engage in subsistence activities (Elmendorf, 1976; Redfield and Villa Rojas, 1962; Villa Rojas, 1987; Webber, 1980). Women at X-Hazil Sur generally care for small numbers of pigs, turkeys, or chickens. These animals are butchered by male members of the household. Some of the meat is consumed within the household, while the rest is sold locally. Usually this money belongs to the woman. Pork costs approximately $2.67 per kilogram of meat, turkeys approximately $3.33 per kilogram live weight, and chickens cost approximately $3.33-5.00 each (live). Maya women also tend 44 small gardens near the house. Some of the herbs, fruits, and vegetables are consumed by the household, while the rest is sold locally, often door-to-door by small children. These activities provide only a limited income to women as these products are commonly produced by most area residents. X-Hazil Sur men also undertake several economic activities in order to supplement their subsistence activities and to earn cash and purchase goods. Some of these activities occur on the ejido, while others take place in nearby towns and cities. Within the ejido, the main economic activity for men is to extract latex from sapodilla trees growing wild in the forest (A. Jorgenson, 1992). Chicle latex is used to produce chewing gum, and has been exported to the United States, Japan, and several European countries for more than 100 years (Otanez Toxqui and Equihua Enriquez, 1981; Dachary and Amaiz Bume, 1983). Maya men harvest chicle latex during the rainy season by cutting canals in the bark and collecting the latex in a bag placed at the base of the tree. Men, working independently but as members of a formal cooperative, may process 5-15 trees daily for 3-5 months and earn about $350 per season (Barrera de Jorgenson, 1993). Usually men form teams of 2-4 individuals and work an area of forest for several days before moving on to another site. Men usually leave home each morning and return each afternoon. Rarely, men establish camps in the forest and harvest chicle for 7-14 days before returning to the village. Men usually bring their guns to the tapping site and frequently shoot wildlife between the village and the work site. Another economic activity for Maya men in Quintana Roo and Ejido X-Hazil y Anexos is to harvest the trees for lumber (Edwards, 1986; Murphy, 1990). Mahogany (Swietenia macrophylla) wood is economically valuable and has been harvested primarily for export in the Yucatan Peninsula since the late 1800s (Negreros, 1991; Snook, 1993). Mahogany wood is processed overseas and is used to make items such as furniture, floor tiles, and wall panels. Other species of valuable tropical woods are harvested as contracts are ^proved by the Government of Mexico. An especially valuable species is chechem (Metopium brownei), which is used for railroad ties in Mexico. Lumber harvesting occurs during the dry season, and separate work teams locate, fell, trim, and load the trees onto trucks. 45 Some workers use gasoline-powered chain saws, while others use axes and machetes. Workers often harvest game in the forest while processing trees. For many years, timber harvested on Ejido X-Hazil y Anexos was shipped to private lumber mills near Chetumal for initial processing before shipping the lumber overseas. In 1989 the ejido negotiated the siting of a lumber mill in the village of X-Hazil Sur. The mill owner agreed to train local ejidatarios as mill workers, process ejido timber there, and after 3-4 years turn over the mill operation to the ejido. Presently the mill provides part-time employment to about 75 workers. These workers cut planks and build tables, chairs, and bee hives. While some of these items are purchased locally, most are shipped to Chetumal. The impact of this mill has not been fully felt by ejido residents, but already several men have reduced the time spent hunting and in other subsistence activities. Maya Indians in Quintana Roo have cared for European domestic livestock (i.e., sheep [Ovis aries), horse [Equus caballus), and cow [Bos taurus]) since about 1519 (Hamblin, 1984, 1985; Hamblin and Rea, 1985). Today, about eight teams of men herd cattle on the ejido. These herds are relatively small and presently the main focus of the owners is to increase herd size. About once a month, however, a cow is butchered in the village and the meat sold locally for approximately $3.33 per kilogram. Goats also are raised. One man has a herd of about 15 goats and feeds them fresh leaves from the ramon (Brosimum alicastrum) tree. Eventually these goats will be butchered and the meat sold in Felipe Carrillo Puerto. Another man has about five domestic rabbits and along with his family consumes the meat as circumstances warrant. Men and women at X-Hazil Sur also tend small stores (five stores in the village offer the basic necessities); weave hammocks from thread bought in stores; sew huipiles, white cotton dresses used by women; and tend to special garden plots (zona fruticola; 1-3 ha, irrigated, with the land deeded to specific individuals) where fruits and vegetables are grown to be sold outside the village under a special government program. The importance of these activities to X-Hazil Sur residents is limited in that 46 transportation costs are high, outside buyers usually pay low prices, and other villages in the area also produce the same crops at the same time. Some X-Hazil Sur residents work outside of the ejido for wages. For example, about 5-10 men work in Felipe Carrillo Puerto as masons and construction laborers. They earn about $7-10 per day, but the work is part-time. These men live on the ejido and tend personal gardens, however, in order to maintain their status as ejidatarios. In another example, two brothers have completed their secondary education, learned English, and now work as waiters in an international hotel complex near Tulum. These young men visit X-Hazil Sur frequently, but have lost their ejidatario status. Other men have attempted to locate employment outside of the ejido, but generally have been unsuccessful. However, in view of the limited economic opportunities in the area, young adults— more than ever before— are graduating from high school, attending technical schools, and obtaining jobs in stores, restaurants, and shoe factories in the large towns and cities (Thompson, 1974). There are two important considerations regarding subsistence and economic activities: One, these activities occur throughout the year (Figure 2-5). For example, after the end of the chicle tapping season a man may switch to logging or gardening, only to resume chicle tapping later in the year. At no time are these people inactive for an extended period. Two, X-Hazil Sur residents usually engage in several activities at a time. For example, a man may weed his garden during the morning, survey potential garden sites, and gather firewood on his way home. Subsistence Hunting Hunting is no longer an important activity for the majority of Maya Indians at X-Hazil Sur in light of the alternative ways to subsist. Many men, however, continue to hunt regularly and harvest a substantial amount of game (J. Jorgenson, 1990). The nature and extent of this hunting will be discussed in subsequent chapters. 47 Ethos As with many indigenous people, the worldview of the Maya of X-Hazil Sur has been influenced by many factors, including their history, physical environment economy, and social changes brought about as these people are confronted with the realities of Mexican and international life. To the Maya, the outside world as well as the immediate environment are a source of uncertainty. Many of the Maya at X-Hazil Sur believe that the world is filled with evil spirits that must be propitiated regularly. Appropriate offerings include praying, lighting a candle, sponsoring a mass, or preparing a special meal. Failure to make an appropriate offering may result in any of a series of unfortunate events, including an accident, crop failure, a sick family member, bad luck while hunting, or the loss of a valuable tool. Offerings often are made on a proactive basis in that a person will make an offering ACTIVITY: Chicle tapping Logging Gardening Railroad tie cutting Hunting -Clear- Har Plant — Dbl Corn- Bum Plant Harvest- WEATHER: Temperature Rainfall Max Min Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month Figure 2-5. Monthly occurrence of subsistence and economic activities in relation to temperature and rainfall (Max = month with highest average temperatures; Min = month with lowest average temperatures; Dbl Com = double over com stalks; Har = harvest activity). 48 in order to avoid potential problems. While many X-Hazil Sur residents considered themselves deeply religious, others professed only limited beliefs. Several important saints are recognized by the Maya at Ejido X-Hazil y Anexos. During some of these celebrations, especially prior to planting the garden or during an extended dry period, game is taken on special group hunts (batidas) and consumed as part of the ceremony. Other saints do not require game. San Juan, for example, is celebrated in June. For several weeks prior to the celebration, representatives of the saint visit surrounding villages and gather food that will be cooked and redistributed during the public ceremony. These representatives go from house to house soliciting com, chickens, and other groceries. San Miguel, another important saint, is celebrated in February with an extensive ceremony that lasts 10 days. There is a great deal of peer pressure to contribute valuable commodities such as a pig or a sack of com to San Miguel. Failure to meet these personal promises is viewed by the Maya as an open invitation for revenge by San Miguel. Religion plays an important role in Maya life. Of the three villages, X-Hazil Sur and Uh-May have small churches, while Chancah Veracruz has a larger church that hosts many regional ceremonies. Separate Catholic masses and traditional Maya ceremonies are held at these sites. Few residents of X- Hazil Sur are fervent believers in the Catholic faith, but parents often bring their children to a priest in order to be baptized. This baptism will facilitate obtaining legal identification papers by the child that are necessary in order to attend school. These Maya see no conflict in adhering to both sets of beliefs and often incorporate Catholic prayers (in Spanish) in the traditional Maya ceremonies. At X-Hazil Sur, the Maya see outsiders as potentially hostile and dangerous. An adult Maya man will seek security through economic independence for his immediate family. This independence, he believes, is achieved by hard work and looking out for one's self. Most X-Hazil Sur residents are reluctant to give or seek economic aid, or to lend or borrow items from others, and, except for mandatory village chores (faenas), they rarely engage in cooperative projects. Initially I, too, was considered an outsider and viewed with great suspicion by the residents of X-Hazil Sur. During my early visits to the village, I accompanied biologists from my host agency who 49 already had established friendships with X-Hazil Sur residents over several years of visits. In order to undertake my study at X-Hazil Sur, I had to make a formal request at a village meeting (asemblea) where the issue was discussed and voted upon. Other outsiders also have been allowed to move into X- Hazil Sur on the condition that they would leave if their presence became objectionable to a large number of residents. This actually occurred a few years prior to my study when an evangelical minister from the United States was asked to leave after problems developed. I fortunately was able to complete my study and left the village on good terms with the residents. Personal and official relationships for the Maya of X-Hazil Sur operate on a mixture of fear, respect, and confidence. Maya usually can count on their immediate family for support and assistance. Friendships with neighbors and former schoolmates are not as reliable and arguments frequently occur. Shamans, herbalists, midwives, the elderly, shopkeepers, and people in audiority generally are treated with respect by all. In dealing with teachers, doctors, government officials, and other non-Maya outsiders, the Maya have a sense of fear and inferiority, and they frequently are abused by those in power. Ejido residents live a tightly controlled life, and their interpersonal behavior is guarded. Adult men generally are somber and rarely offer important information to other residents or outsiders. This conduct is similar to peasant behavior in Mesoamerica and central Java (Wolf, 1957). At X-Hazil Sur, various groups of men form nightly at stores and the church to discuss village matters. Adult women primarily remain at home where they visit with friends and family while attending to household chores. Social relationships are especially difficult for young adults to establish. After finishing the eighth grade, young men and women have limited opportunities to date and have a hard time meeting potential spouses due to a lack of time, money, and meeting sites. Children also are guarded in their behavior and most children do not greet or smile at outsiders. These behaviors are changing, however, as children learn about other cultures in school and the adults gain more experience with outsiders through visitors such as myself, traveling salesmen, culturally sensitive tourists, and while working outside of the village for wages. 50 Creativity and artistic expression are not encouraged by the Maya of Ejido X-Hazil y Anexos. During my study, I was aware of only one local painter and a handful of musicians. None of these artists was highly regarded by X-Hazil Sur residents for their talents. In school, children frequently were encouraged to draw, but their parents usually did not appreciate their artwork. A potential outlet for artisans is the DIF store in Fehpe Carrillo Puerto, where handicrafts such as hammocks, dresses, weavings, and carvings, are sold to tourists, but no X-Hazil Sur residents have sold items at this store. The Maya of Ejido X-Hazil y Anexos show a general lack of concern for the future or for self -improvement. Simple matters, such as leaking roofs and losing one's pigs for a few days while they feed in a neighbor's field, are addressed long after they become a problem. Complex problems, such as the declining abundance of game and the determining the equitable use of land by cattle ranchers and horticulturalists, generally are ignored. With respect to self-improvement, only a few families have encouraged their children to seek further education or employment outside of the village, and those individuals who leave subsequently are treated as outsiders. In spite of a general reluctance by ejido residents to ask or seek economic assistance (except from government officials who usually did not insist that the loans be repaid), I was continually asked to give loans against personal items left as security. Occasionally the money was used to purchase necessary items such as food, medicine, or clothing. More often, however, a man would pawn his wife's jewelry in order to buy beer. The local shopkeepers also loaned money or goods against future earnings by village residents. Fortunately, my clients were able to clear all of their debts before I left. While living in the ejido was an interesting experience and my wife and I made many good friends, we also noted several problems that complicated daily life for all. Perhaps the major problem was the immoderate consumption of beer by men and the numerous fights that resulted. This was aggravated by the fact that the men often did not receive any pay until the work was completed even though the job might last several weeks. Men often received 3-6 months earnings in cash at a time. Unfortunately, many of these men would spend much of this money at the local tavern or in Felipe Carrillo Puerto, and within a few days they would be broke again. 51 A second major problem was that the ejido was divided into two main sociopolitical groups. As a result, many problems that needed cooperation by all in the village were not resolved. Initially the division was between two groups of families. One faction was composed of members of a single, extended family that had become somewhat more politically and economically powerful than others in the village. The other faction was composed of families that wanted to obtain that power. Jealousy and animosity often prevented these groups from cooperating in community affairs. The matter was further complicated by competing interests in the villages of Uh-May and Chancah Veracruz. The conflicts between groups and villages has led to difficulties when it was time to elect village officials, decide on timber harvests, and make work assignments. Usually matters were decided by a vote of ejidatarios and little thought was given to compromise or being fair. The minority group always lost and their resentment was apparent as projects were duplicated, poorly completed, or left undone out of spite for those in charge. For example, the village had two bus lines that simultaneously covered the same route. Likewise, X-Hazil Sur had an unfinished village hall, unchlorinated drinking water, and volleyball and basketball courts that lacked lights because residents of one village did not want to pay for recreational activities in another village. In addition, village officials often received complaints from individuals but were unable to organize an effective committee to complain to municipal officials about village problems, for example, incompetent teachers and frequent closings of the local schools. These problems likely will continue until residents learn how to work together in a more cooperative manner. The picture was not completely bleak, however, as many positive activities also occurred. For example, the village had a baseball team that competed with several other villages and towns. The players had uniforms and provided great entertainment on Sunday afternoons. Village residents also managed to host several dances with live bands during the past few years with the profits going for specific village projects. Further, village officials successfully negotiated the placement of a small lumber mill outside the village in order to process locally harvested timber. In this manner, village residents will obtain greater profits from the trees they harvest. Village officials also initiated a 52 program where about 30 residents will receive government loans and subsidies to grow fruits and vegetables on private holdings. These projects have improved village life greatly during the past few years. Closing Comments Recognizing the multidisciplinary nature of this study, I have tried to show the complexity and interrelatedness of the three basic elements of this project; gardens, wildlife, and hunting. No single element can be understood in isolation from the other two. To appreciate these results, it is also imperative to consider the natural, cultural, and socioeconomic environment of the area, how conditions have changed over the past 25,000 years, and what challenges lie in the future. During the next several years, the Maya of Quintana Roc will have to decide what aspects of their culture they wish to maintain and what will be allowed to fade as the area is further developed for logging, tourism, and cattle ranching. If the costs and benefits of the various alternatives are carefully researched by social and biological scientists and clearly explained to the Maya, I would hope that the Maya would choose wisely, but suspect that they might not. CHAPTER 3 GAME HARVEST BY MAYA HUNTERS IN QUINT ANA ROO, MEXICO Introduction Studies on the nature and extent of hunting have been identified as a critical need to develop resource management strategies (Posey et al., 1984). These alternative strategies are necessary because traditional approaches used in Europe and the United States have not been successful in response to the loss of pristine habitats, especially tropical forests, and the unregulated harvests of wildlife, especially for commercial purposes (Shaw, 1991). While people have modified their surrounding habitat and harvested wildlife for millennia, the rates at which these activities presently occur have placed the needs of many people in conflict with those of wildlife (Redford and Robinson, 1985, 1991; Robinson and Redford, 1991c). The challenge for conservationists is to identify instances where the goals of natural resource conservation and the goals of indigenous people are compatible. Previous studies about hunting have treated hunters, wildlife, and the environment in which they occur as separate, independent elements. Traditionally, anthropologists have studied hunting from the point of view of indigenous people and their associated cultural factors. Studies in the Neotropics, for example, have quantified differences between indigenous groups with respect to hunting technique (Hill and Hawkes, 1983; Yost and Kelley, 1983), taboos (McDonald, 1977; Ross, 1978), use of gardens and habitat modification (Linares, 1976; Nations and Nigh, 1980), and hunting regulation (Balee, 1985; Werner, 1983). Biologists, on the other hand, traditionally have studied hunting from the point of view of its effects on the wildlife. Studies in the Neotropics, for example, have quantified differences between wildlife species with respect to changes in behavior (Crawshaw, 1991; Dallmeier, 1991; Mittermeier, 1991), population density (Freese et al., 1982; Peres, 1990), patterns of habitat use (Bodmer, 1989; Bodmer et al., 1988a, 1988b; Fragoso, 1991), and mammal community structure (Glanz, 1991; Janson and Emmons, 1990; Malcolm, 1990). While these studies have provided much 53 54 specific information about the particular species or indigenous group in question, natural resource managers are still attempting to develop viable, long-term, management strategies. Additional studies are necessary that integrate humans, hunting, and conservation. An analysis of subsistence hunting by Maya Indians may be valuable in understanding how an indigenous people apparently can exploit wildlife for subsistence purposes. A biological analysis will help understand if this hunting is being conducted in a sustainable manner. Maya Indians have practiced shifting cultivation and subsistence hunting in southern Mexico and Central America since at least 1,500 B.C. (Adams, 1991). Archaeological evidence from Cerro Brujo, in northern Panama, indicates that the species composition and relative abundance of game taken by indigenous people at that site during 960-985 AD is similar to that taken by Maya subsistence hunters today (Linares, 1976). Assuming that hunting practices today are similar to those conducted several centuries ago in the region, and that hunting practices throughout the Maya realm were similar, this suggests that subsistence hunting by Maya Indians may be sustainable and could be used to develop alternative resource management strategies. The evidence that this hunting could be sustainable was based on an examination of the behavior, species composition, and relative abundance of wildlife at Cerro Brujo compared with their current abundance at other forested Neotropical sites. Linares (1976) noted that certain wildlife species at Cerro Brujo were relatively more abundant than expected, based on an analysis of bones at a village refuse site, when compared with their abundance at other forested Neotropical sites that were not populated by people. According to Linares (1976), these relatively abundant species used the gardens planted by the Cerro Brujo people and benefitted from the interspersion of small gardens in the surrounding forest. Although no data were presented, Linares (1976) postulated that the population density of the species using the gardens was greater because they regularly fed on cultivated crops planted by the hunters. These people subsequently modified their hunting practices and began to specialize on the wildlife species that foraged in the gardens. Linares (1976) called this practice "garden hunting" and defined it as an association between hunters and prey where the prey, due to the 55 garden, benefitted from the additional food resources available and the hunters, due to the garden, benefitted from the additional game available. The current study was designed specifically to test the relationships comprising garden hunting. More generally, it was designed to compile information about a highly acculturated group of indigenous people practicing subsistence hunting. The results of this and subsequent chapters will be useful in evaluating the specific model of garden hunting. These results also will be useful in developing alternative resource management strategies as the Maya contemplate a shift in subsistence activities from hunting and horticulture to logging, cattle ranching and commercial fruit production. There were three general objectives for this chapter and several specific hypotheses: First objective.— Describe the composition and characteristics of the game harvested by Maya hunters at X-Hazil Sur. Several hypotheses were tested: One, the proportion of male game animals taken was equal to the proportion of females taken, per species. Two, the monthly take of mammals and birds was equal among months. Three, the hourly take of mammals and birds was equal among hours. Four, the distfibution of take of mammals and birds was equal, among specific time intervals. Second objective. --Describe the characteristics of the hunters and their weapons. Several hypotheses were tested: One, the age class distribution of hunters was equal to the age class distribution of the general population of males at X-Hazil Sur. Two, the yield of game per outing and hunter-hour was equal between different-sized groups. Three, the mean prey item weight was equal between different types of weapons and by hunter for the seven main hunters, based on the total number of prey taken. Third objective.— Describe the characteristics of the hunt. Several hypotheses were tested: One, the yield of game per outing and per hunter-hour was equal for different types of hunts. Two, the distribution of kill sites in different vegetation types was equal to the proportion of these types in the study area. Three, the mean kill site distance was equal, by game species and by hunter for the seven main hunters. 56 Methods Study Area The study took place at Ejido X-Hazil y Anexos, Quintana Roo, Mexico, during 1989-1990 (total area = 552.95 km^; Figure 2-1). Hunting data were obtained from hunters at the village of X- HazU Sur (19°23'30"N, 88°05'00"W; population = 1,040), the largest of three villages on the ejido (total population = 1,680). The mean annual temperature is about 26°C and the area typically has one dry season (December-May) and one wet season (June-November). Rainfall during 1 January-12 December 1990 was 1,277.3 mm (Figure 2-4). About 88.52% of the ejido was categorized as Late Secondary Forest, 6.07% as Plots & Gardens, 5.18% as Early Secondary forest, and 0.23% as Other (Table 2-3). Since about 1915 the area has been occupied by Maya Indians, whose main subsistence activity has been shifting cultivation, primarily com. Prior to 1915, Maya Indians did not occur in the area. General Considerations The key to the successful completion of this project was obtaining the cooperation of village residents and individual hunters. This cooperation was gained over several months and required careful and continuous attention to many local political and social matters that were indirectly related to the field research. These matters were important, however, because it was difficult to separate personal and professional relationships in the village. My ability to obtain hunting data was enhanced by living full-time in the village and participating fully in local affairs. The results of this study were composed of data about game species taken during outings by hunters from the village of X-Hazil Sur. Reports from hunters in the adjacent villages of Uh-May and Chancah Veracruz were not included due to time and financial constraints, even though they were part of the ejido and hunted on areas used by X-Hazil Sur hunters. The working arrangement widi hunters was that they would report a kill, regardless of the time or the number of animals taken, and allow it to be weighed and measured before they butchered it. Game processing and hunter interviews usually 57 began within 5-10 min of the notification. In order to avoid problems among hunters and to guarantee hunter anonymity, no details of the hunt were divulged to other village residents. Two unresolvable problems were that not all hunters agreed to participate in the study and not all participating hunters reported all the game they took. These problems will be addressed below. Another matter to be resolved was how to distinguish between game and nongame species. Hunters at X-Hazil Sur took a wide range of animal taxa for subsistence purposes. However, hunters also captured some species to keep as pets. Other species were killed because they were dangerous (e.g., venomous snakes), killed domestic animals (e.g., Didelphis spp.), possessed valuable skins (e.g., Felidae), or were crop predators (e.g., birds of the Psittacidae, Columbidae, and Icteridae families). Although young boys with slingshots frequendy killed small animals for target practice, adult hunters did not kill animals for this purpose. In order to focus on species taken for subsistence purposes, local hunters were interviewed to develop a list of primary game species. Game species were defined as taxa that were taken regularly by a large number of hunters and whose meat would be socially acceptable to most X-Hazil Sur residents. Four taxa were excluded from consideration although Uiey were taken for subsistence purposes by some hunters. Doves (Columbidae) and parrots and parakeets (both Psittacidae) were crop predators and frequently were taken by young boys protecting gardens. These species were excluded because they were not widely consumed by X-Hazil Sur residents. Fish were excluded from this study because only a few residents went fishing, fish generally were available only during a short period of time each year (April-June), and most hunters viewed fishing mainly as a reaeational activity where consumption of the meat was of minor importance. The project had several phases and not all types of data were recorded during the entire study. Initially, data were recorded on hunting oufings for game and nongame species. Subsequently, nongame species were excluded from consideration. Effective 31 July 1990, a decision was made to stop recording data for the pocket gopher (Orthogeomys hispidus [Geomyidae]) and plain chachalaca (Ortalis vetula [Cracidae]). This decision was made for three reasons: One, research priorities had 58 shifted from the hunting inventory to completing other aspects of the project. Two, it became apparent that several hunters were taking these species in order to participate in the study. Three, while frequendy taken, these species were of limited value for subsistence purposes, compared with die other game species, due to dieir small body weight. Number of Species and Individuals Animals usually were examined fresh and whole. Animals were identified, weighed, and measured according to standard field procedures. Scales wiUi different capacities and graduations were used to weigh the animals; however, the weights presented here have been rounded off to 0.1 kg. Hunters were queried about Uie details of the hunt during formal interviews (Appendix D). Interviews usually lasted about 15 min and were conducted either at the field station or die home of die hunter. The primary game identification guides were Leopold (1977) and Peterson and Chalif (1973). Sex of Game Species The sex of an animal was determined primarily based on a visual examination of the carcass, but internal organs also were frequently examined, especially for mammals. The sexing of birds in die field was more difficult dian sexing mammals, but each species had certain identifying characteristics. For the great curassow (Crax rubra [Cracidae]), plumage coloration was used as males are black, while females are reddish-brown (Peterson and Chalif, 1973). For die ocellated turkey (Agriocharis ocellata [Meleagridae]), die presence of leg spurs was used to identify males (Leopold, 1977). For the plain chachalaca, die shape and lengdi of the trachea was used. In male chachalacas, die trachea forms a comparatively long loop that extends over the lower abdomen, while in females the loop is short (Delacour and Amadon, 1973). Sexual identification of the diicket unamou (Crypturellus cinnamomeus [Tinamidae]) was based on examination of internal organs by museum specialists at die Universidad Nacional Aut6noma de Mexico as die sexes are similar in size and plumage. 59 Age Class of Game Species Primary game species were assigned to one of three age class categories: adult, subadult, and young. These categories were based on visual examination of the carcass and on information provided by hunters as strict age class definitions for these species have not been established for Mexico. In general, individuals categorized as adults were defined on the basis of having a large body size; worn fur, teeth, nails, and feathers; and having well-developed reproductive organs. Individuals categorized as young generally were no more than a few months old and were defined on the basis of having a small body size; unworn fur, teeth, nails, and feathers; and having poorly-developed reproductive organs. Individuals categorized as subadults were defined on the basis of having intermediate characteristics. The information provided by experienced hunters was especially helpful in assigning the prey to an age class. Reproductive condition The reproductive condition of female birds and mammals was noted to identify when they were gravid or caring for young. Gravid females were identified by a visual inspection of the reproductive tract to locate eggs, embryos, and fetuses. Lactating mammals were determined by squeezing the teats. Brooding birds were determined by having a brood patch. Reproductive tracts also were collected and preserved in 15% formalin for microscopic studies, but they have yet to be analyzed. Distribution of Take bv Month and Time of Day Data were summarized by month and time of day in order to compare the take of game by hunters. The monthly take of mammals was compared for 17 months, while the take of birds was compared for only 7 months because of apparent reporting problems. At first, most hunters did not want to report taking birds, and then it appeared that some hunters were shooting birds specifically to participate in the study. The hourly take of game animals was compared for 24 h over these time periods. Temporal differences in take between mammals and birds were compared for seven, two-hour 60 intervals: 0600-0759 h, 0800-0959 h, 1000-1159 h, 1200-1359 h, 1400-1559 h, 1600-1759 h, and 1800-0559 h. To ensure adequate sample sizes for statistical comparisons, data were combined in order to have a minimum sample size of five or more individuals per interval. Age of Hunters Information about the ages of hunters was obtained from local medical doctors who had conducted annual censuses of village residents. In most cases, ages were obtained from birth certificates, but in some cases they were based on the memory of the respondent. Identification of Hunters. Hunting Group Members, and Types of Weapons In order to obtain information from the hunters, it was agreed not to divulge dieir names or the details of their kills. Hunters were reluctant to cooperate, at fu-st, for various reasons. One, hunting in the study area was a private matter in that game was not considered to be community property and hunters were not expected to share information about game kills or sightings with other residents (cf., Hames and Vickers, 1983). Two, much of the hunting was illegal according to national legislation because some game was taken out of season or in excess of bag limits (cf., SEDUE, in litt.). In order to protect the identity of hunters, specific individuals are referred to by number. In most cases, cooperating hunters readily acknowledged killing an animal. They also openly provided information about dieir hunting partners and the weapons used. Noncooperating hunters who were reported to me by third parties often either denied making a kill or only provided limited information about the outing. Every attempt was made to encourage hunters to participate in the study, and anonymity was assured, but the wishes of those who chose not to collaborate were respected. Seven hunters obtained substantially more game than did the other 77 hunters. The individual results for the main hunters were described in the text and in Appendix E to contrast hunters who harvested a large amount of game versus those who harvested a small amount of game. In all cases. 61 each game kill was attributed to a single hunter. The arithmetic mean prey weight per hunter was defined as the total weight of the prey divided by the number of individual prey items, for each hunter. The entire list of specimens obtained during the study is listed in Appendix F. This list includes game and nongame species regardless of how obtained. Tvpes of Hunts Hunters at X-Hazil Sur used various mediods to obtain game. Based on information provided by hunters during die interviews, hunts were assigned to one of four categories: tracking/stalking, stand/platform, trapping, and other. According to the tracking/stalking method, hunters search for tracks, feathers, or feeding sign as they quietly and slowly walk through the forest or garden. Periodically the hunter sits on a log or rock that provides a clear view of the area and listens for animal sounds. Hunters use this technique during the day and night. During night hunts, hunters use headlamps. According to Uie second method, hunters construct a stand or platform in an elevated location and begin using the site a few weeks later, after die animals become accustomed to the structure. Stands and platforms usually are 3-5 m above ground level and include natural forks in trees, free- standing tripods made from tree trunks, or branches fastened to convenient trees. Stands and platforms are located in both the forest and gardens. Structures located in forests usually are adjacent to game trails or positioned near water holes or fruiting trees (e.g., sapodilla trees [Manilkara zapota]). Structures located in gardens usually are positioned to take advantage of game trails, garden edges and topography, and die distribution of CTops in die garden. Stands and platforms almost always are used during the night. Two basic types of traps were used to capture mammals: wire snares and deadfall traps. Pocket gophers were trapped using a wire snare placed in the tunnel. Upon activating the snare, the pocket gopher is pulled up against the tunnel ceiling by the wire around its chest and eventually suffocates. The tunnel ceiling is reinforced widi small branches to widistand die pressure exerted by 62 the stick used as a spring. Pacas (Agouti paca [Agoutidae]) and agoutis (Dasyprocta punctata [Dasyproctidae]) were captured using a deadfall trap constructed of rocks and boards at the entrance of their burrows or caves where they sought refuge. Caves and burrows frequently were discovered by dogs that accompanied hunters. Both types of traps often were set near work sites in the forest or garden and along trails between the site and the village. Traps have the advantage of being able to c^ture an animal in the absence of the hunter. Disadvantages with using traps are many and include a tendency for the trap to be activated prematurely and the potential for ants, other hunters, and wild and domestic animals to steal or damage the prey. Hunting Outings For purposes of this study, a hunting outing was defined as any activity that resulted in the taking of a game animal for subsistence purposes. Unsuccessful outings were not recorded. Some outings were exclusively for hunting, while others also included nonhunting activities such as logging, chicle tapping, tending the garden, and gathering firewood. This definition of a hunting outing, based on the taking of a game animal, was necessary for several reasons: One, hunters did not report unsuccessful hunts. Two, hunters often engaged in other activities before or after a hunt. Three, hunters and nonhunters often carried firearms to the forest or garden, even though their intent was not to hunt. Four, hunters did not necessarily have to depend on firearms to take game as they could use dogs, traps, rocks, machetes, and heavy branches. Five, hunters were open to changing their plans to work or hunt in response to changing events. For example, strong winds or a rain storm could delay or terminate a hunt, while the accidental discovery of game animals in the garden could convince hunters to initiate a hunt. For these reasons data were taken only on successful hunts. The duration of a hunting outing was defined as the difference between the departure and arrival times of the hunter to his home. Many hunters had watches and could accurately provide this information. Other hunters would estimate their times of departure, kill, and return, or relate them to 63 local events such as moonrise, sunrise, arrival of the village bus, or a television program that was on when they departed or returned. Outing duration included travel time to and from the kill site, and any additional time spent in activities such as resting, eating, or preparing a stand or platform while away from home, but only during that specific outing. This definition may ^pear excessively inclusive, but is justified for two reasons: One, due to the interspersion of roads, trails, gardens, and forest, game animals potentially could be taken almost anywhere in the ejido. Two, game frequently was encountered while the hunter was engaged in other activities. In those few instances when the hunter indicated that he also engaged in nonhunting activities during the outing and definitely was not able to hunt or pursue game, the duration time was adjusted accordingly. In those few cases when a hunter wounded an animal, returned home, and resumed the search with additional help, the combined hunt, travel, and search times were considered as a single outing. The determination of hunting outing duration was facilitated by the hunter immediately reporting a game kill. Vegetation Tvpes and Land Uses According to aerial photograph interpretation and a ground survey conducted by foresters of the Plan Piloto Forestal (now named Plan Estatal Forestal), there are six vegetation types and three land uses on the ejido (Ing. Marcelo Carre6n Mundo, in litt.). In order to characterize the ejido and kill sites for this study, these categories were condensed into four land use/vegetation types (Table 2-3; Figure 2-5): One, the category identified as "Plots & Gardens" was composed of active gardens (milpas), home gardens (huertos and hortalizas), and fruit and vegetable plots in the special corridor along the highway (parcelas). These areas generally were located near the village and along roads and trails in the forest. Gardens generally were ^ 2 ha in size, used 1-2 y before being abandoned, and were separated from one another (see Chapter 6 for additional information about gardens). Plots were irrigated, 0.5- 64 1.0 ha in size, included a house site, and are used to grow fruits and vegetables for commercial purposes on a full-time basis. All of the plots were adjacent to each other in the corridor along Route 307, between the village of Uh-May and the turn-off to X- Hazil Sur. Two, the category identified as "Early Secondary Forest" (huamiles and acahuales) was composed of regrowth areas on sites that had repeatedly been used as gardens during the past 25-50 y. This category also included cattle corrals and pastures as these areas had undergone the same clearing/burning process used to make gardens. In Early Secondary Forest, the stem density of trees and herbaceous plants was greater than in Late Secondary Forest (see below), while the basal area was less than in Late Secondary Forest (Barrera de Jorgenson, 1993). Early Secondary Forest areas generally were located near roads and trails, but tended to be at greater distances from the village than were the plots & gardens. Left undisturbed. Early Secondary Forest eventually becomes Late Secondary Forest through the process of succession. Three, the category identified as "Late Secondary Forest" was composed primarily of low- and medium-height semi-evergreen forest (jnonte alto and monte bajo), but also included sinkholes (cenotes), ponds (bajos and lagunas), and seasonal wetlands (sabanas). The forest in this category had not been used as gardens for at least 50-75 y (if ever at all), according to the oldest residents of X-Hazil Sur. In Late Secondary Forest, the stem density of trees and herbaceous plants was less than in Early Secondary Forest, while the basal area was greater than in Early Secondary Forest (Barrera de Jorgenson, 1993). Ponds, sinkholes, and seasonal wetlands were included in this category because they were quite small, not used for gardens, and widely 65 interspersed in the forest. Areas of Late Secondary Forest typically were located along the ejido boundary and furthest from the village. The category identified as "Other" included the three village sites and forest clearings (rodales) used during logging to sort, trim, and load trees onto trucks. Sites in this category generally would not be good game habitat due to repeated use by people and the substantial nature and extent of the habitat alteration. Kill Site Locations and Distance from X-Hazil Sur The location of each kill site was determined after evaluating the information provided by hunters, including the route taken, habitat type of the kill site, distances along each road and trail, and any landmarks (e.g., trails, gardens, sinkholes, and road markers). This information was compared with aerial photographs (Instituto Nacional de Estadistica Geografica e Informatica, February 1985, 1:37,000 scale), topographic maps (Instituto Nacional de Estadistica Geogr^ica e Informatica, 1987, 1:50,000 scale), and personal knowledge of the area. Kill sites were plotted on base maps (scale = 1:50,000). The minimum distance between the village and the kill site was measured as a straight line from the water tank, about 500 m west of the village plaza. The water tank was used as a landmark because it was permanent, highly visible, and indicated on topographic maps of the area. Care must be exercised when interpreting the information about kill site locations because many of the sites were based on estimated distances or features that could not be clearly located on the maps or aerial photographs. Locations within 2-3 km of X-Hazil Sur, along Route 307 (marked every kilometer), or near well-known landmarks probably were accurate to within 500 m, while other locations probably were accurate to within 1-2 km. The distribution of kill sites in Late Secondary Forest was compared with the combined total of kills in areas categorized as Plots & Gardens, Early Secondary Forest, and Other. This was necessary in order to have an adequate sample size of kills. Given the small sm-face area categorized 66 as Other and the similarity between Early Secondary Forest and Plots & Gardens, it was decided to combine these categories. For the species analyses, the expected number of kills per vegetation type was determined by multiplying the total number of kills by the proportion of each vegetation type. Minimum Harvest and Catchment Areas The minimum harvest area for each taxa was defined as the region encompassing the kill site locations for that taxa. The minimum catchment area for each hunter was defined as the region encompassing the kill site locations of all game taken by that hunter. The minimum catchment area was determined only for the seven main hunters. Since hunters did not always report the entire area searched, the minimum catchment area reported here is conservative and includes only the region encompassing the actual kill site locations. The harvest and catchment areas were determined using the minimum convex polygon method (Hayne, 1949; Mohr, 1947) and measured using a compensating polar planimeter. Use of Dogs Dogs frequently accompanied hunters to the forest and gardens, and often located or killed game. Usually this game was collected and brought home to be eaten by the hunters. The presence of dogs on hunting outing, the game taken, and the total number of dogs with the hunting party were recorded. Statistical Methods Parametric and nonparametric tests were used to analyze hunting data. Sex of game taken was tested against an expected 1:1 (male:female) ratio using the x^ test. Homogeneity for the distribution of take by month was tested using the x" test. For mammals, the expected monthly take was equal to 1/17 of the total take, while for birds it was 1/7 of the total take because hunters apparently did not report their bird kills during 1989. The distribution of take by hour was tested using the x^ test and assumed 67 that the hourly take was equal to 1/24 of the total take. The distribution of take of birds versus mammals by time of day was tested using the x^ test. The yield of game per outing and per hunter hour by hunter group size and type of hunt was tested using the Kruskal-Wallis test (x^ ^proximation). The actual hunter group size was used to calculate the yields, but data for groups ^ 5 individuals were summarized in the tables. For type of hunt, only tracking/stalking and stand/platform were compared because for the other two types of hunts yield was essentially independent of hunter group size. The distribution of kill sites in Late Secondary Forest versus Plots & Gardens, Early Secondary Forest, and areas categorized as Other were compared for homogeneity with the x^ test. The mean kill site distance was compared among species and hunters for the top seven hunters using the Kruskal-Wallis test (x^ ^proximation). Mean prey body weight by hunter and by type of weapon was compared using the Kruskal-Wallis test (x^ approximation). The following statistical terms were used: x^ = chi-square (X^ approximation for Kruskal-Wallis test); x = mean, SE = standard error, n = sample size, d.f. = degrees of freedom, and P = probability-value. Specific P values were presented, rather than base a determination of significance on a set alpha-value. All statistical analyses were conducted using SAS (SAS Institute Inc., 1988), and unless stated were two-tailed. Voucher Specimens Voucher specimens were deposited at the Museo de Vertebrados (M. en C. Carmen Pozo, Director), Centro de Investigaciones de Quintana Roo, Chetumal, Quintana Roo, Mexico, and the Museo de Zoologia (M. en C. Adolfo G. Navarro Siguenza, Director), Facultad de Ciencias, Universidad Nacional Aut6noma de Mexico. Mexico, DF, Mexico. Specimen identifications were assisted by J. E. Escobedo Cabrera, E. M. Figueroa, H. Flores, A. G. Navarro Siguenza, and L. L. Paniagua (in litt.). 68 Limitations of the Data Sets There are several problems that must be considered in interpreting the data obtained during this study. Perhaps the most serious problem was the procedure used to obtain game kill reports. This study was based on voluntary reports of game kills provided by hunters. As might be expected, not all hunters provided reports and not all cooperating hunters reported all of their kills. This problem was addressed in many ways in order to improve the likelihood of receiving a report, including: employing local residents as field assistants; living full-time in the village of X-Hazil Sur; maintaining good relations with hunters and their families; and repeatedly conducting informal, follow-up interviews with hunters to confirm or clarify information obtained from others. These practices greatly facilitated obtaining timely and accurate reports of game kills. Based on my calculations, conversations with hunters, and on information provided by village residents, the reported number of game animals taken probably represents about two-thirds of the total game harvest, by number of individuals, at X-Hazil Sur during the course of my data collection. A second problem was with the measurement of time and distance (e.g., duration of the hunting outing or the location of the kill site). While in many cases the hunter knew the exact location and distance to the kill site as well as the time of day when the animal was killed, in other cases these measures were estimated. These estimates, however, were probably quite reliable as the Maya at X- Hazil Sur were very familiar wiUi measuring time and distances while conducting other subsistence activities, such as logging and working in gardens. Distance and time estimates recorded during this study were constantly checked against maps, aerial photographs, knowledgeable local residents, and by making site visits to many of the gardens, kill sites, and work areas in the forest. A third problem was the normality and independence of the data observations. It rapidly became apparent during the study that the nature and extent of hunting depended greatly upon the biases and preferences of the hunters. For example, some hunters went hunting on a relatively frequent basis, while others went infrequently, and some hunters had several years of experience, while others were just learning. In addition, some hunters were highly dependent upon the game they obtained as food. 69 while others were not as dependent as they had domestic animals (e.g., pigs, chickens, turkeys, and cattle) to supplement their diet. As a result of these differences, there likely were several types of hunters and hunting outings at X-Hazil Sur, that when quantified, would not be independent or normally distributed. These differences among individuals, however, were not important here as the research focus was on the ejido as a whole and the patterns exhibited by those hunters. The assumptions of data normality, independence, and equal variance were avoided by using the nonparametric, Kruskal-Wallis test when comparing groups of hunters or game species, instead of the more-rigorous, parametric ANOVA test. Given that it was not necessary to meet these assumptions in order to make the various comparisons, the Kruskal-Wallis procedure could test for differences under the circumstances at X-Hazil Sur. Results Composition and Characteristics of the Harvest Number of species and individuals. A total of 584 game animals were reported taken by hunters at X-Hazil Sur during 17 months (June 1989-October 1990; Table 3-1). Mammals comprised 66% (n = 385 individuals) of the reported total and birds 34% (n = 199). No reptiles, amphibians, or insects were collected by hunters for personal consumption, except that honey was consumed when encountered in the forest. A small number of fish (Cichlasoma urophthalmus [Perciformes, Cichlidae]) were taken by hunters during April-June of each year, but these data were not included here. This total also does not include animals collected by local residents for other reasons, harmful or distasteful animals that were killed and abandoned, or game species found dead on the road (except for three instances of fresh game that was consumed by hunters). Eight mammalian and four avian taxa were taken for subsistence purposes. The coati (Nasua nasua [Procyonidae]; n = 167 individuals) was the most frequently taken mammal, followed by the pocket gopher (n = 53) and the paca (n = 47), while the plain chachalaca (n = 167) was the most frequently taken bird (Table 3-1). 70 The game taken by hunters provided residents of X-Hazil Sur with a substantial amount of meat. The total body weight of the 584 animals was 2,700.1 kg (Table 3-1). Of this total, 95% was from mammals and 5% was from birds. Three mammalian taxa combined, white-tailed deer ( Odocoileus virginianus [Cervidae]; 709.0 kg total weight, J = 29.5 kg), collared peccary (Tayassu tajacu [Tayassuidae]; 618.5 kg total weight, x = 15.5 kg), and coati (504.9 kg total weight, x = 3.0 kg) provided 68% of the total weight. This meat was primarily consumed by the individual hunters and their immediate families (ca. 400 people), but small quantities were also sold locally (see below). Sex of game species. Males and females were not taken in equal proportions. For all mammals combined, significantly more females were taken than males (jc = 5.3150, d.f. = 1, P < 0.025; Table 3-2). However, for birds, more males were taken than females (x" = 11.8128. d.f. = 1, P < 0.005). While mammals and birds were each dominated by a single taxon, the patterns these higher taxa groupings exhibited were generally the same as those exhibited by individual species. The ratio of males to females taken varied between game species. For mammals, more females were taken than males for seven of the eight taxa (Table 3-2). The mean sex ratio for all mammals combined was 1:1.3:0.0 (males: females: unknown). For manunals, the greatest disparity between males and females taken was for the white-tailed deer (1:3.8:0.0; x" = 8.1667, d.f. = 1, P < 0.005). The ratio of males to females taken also was significantly different for the agouti (1: 1.8:0. 1; X^ = 2.9412, d.f. = 1, P < 0.10) and pocket gopher (1:1.7:0.1); x' = 3.7692, d.f. = 1, P < 0.10). For birds, more males were taken than females for three taxa, but the only significant difference was for the plain chachalaca (1:0.6:0.1; x' = 11.7771, d.f. = 1, P < 0.005; Table 3-2). The mean sex ratio for all birds combined was 1:0.6:0.1. This suggested that for the plain chachalaca there was a significant difference in the harvest by sex, while there was no difference for the great curassow. 71 Table 3-1. Reported number of individuals taken, mean weight (kg), and total weight (kg) of game taken by Maya hunters at X-Hazil Sur, Quintana Roo, Mexico, during June 1989-October 1990. Rank Mean Total Rank Total number order weight weight order Game species individuals taken (no.)^ (kg) (kg)" (wt.)' a) Mammals Pocket gopher 53 (51)" 3 0.4 22.3 10 Paca 47 4 5.8 274.8 4 Agouti 35 6 2.8 96.8 6 Coati 167 1 3.0 504.9 3 White-lipped peccary 3 12 31.4 94.3 7 Collared peccary 40 (36) 5 17.2 618.5 2 Brocket deer 16 8 15.6 250.0 5 White-tailed deer 24 (22) 7 32.2 709.0 1 Total mammals 385 (377) 2570.6 % of all game 66 95 b) Birds Thicket tinamou 13 9 0.4 4.9 12 Great curassow 13 9 3.1 40.1 9 Plain chachalaca 167 1 0.4 64.9 8 Ocellated turkey 6 11 3.3 19.7 11 Total birds 199 129.5 % of all game 34 5 Mammals & Birds 584 (576) 4.7 2700.1 Rank order based on total number of individuals taken. Total weight was determined by sununing the weights of the individual prey items. The degree of precision varied between species as different scales with assorted capacities and graduations were used. Rank order was based on the total weight of the individual prey items taken. Values in ( ) indicate number of individuals weighed if weight not available for all individuals in taxa. 72 < 00 s s o 3 i: 3 V5 3 0-1 •o t/i a 1) o £ s .'^ U. X a C/2 ■~A ■u 19 S o o o vq 00 o iri o c-. q d 6 oc t' d cs d od C'i fS 00 r-; r~- f. o q q m rn OS uS r-^ c<^ d u-i o o\ 5 1 (N o \c r- as (N o o m fi d cs d d »-^ CO 1/^ - — - - ■* m u-i ts VC tN c vc ON m en «S cs oo tN % NC SO i ■a § so < •3 O U 1 1 13 03 1 1 S 8 ■9 2 o d o d d o d ■^ fs o u~i d ^ l-H 1-^ f^ ON 3 O S •3 i3 — 'J NO CN ■<3- 00 ON d V On NO o ON 00 IT) o o d V r- O — ' NO vn O o o O 79 u s ■a ■3 W5 ■o '> 79 1 § in V3 3 V c •o ^ "S Srt ^ 3 "o "u u c o ■^ 5 c o o i> § -3 (9 aj c ^ g « w 1) V5 u 3 ^ ^ tf5 3 li X oj nj •T3 X Vi u ^ 1 Im tM 5 ^ u 5*— 79 o. 1-^ 3 *«l ■o o II '> B •^i ^ V3 1) ■di M H 73 Age class of game species. The proportion of adults, subadults, and young taken by hunters varied between species. For all mammals combined, adults were taken more frequently than subadults or young (55% adults, 43% subadults, and 2% young; Table 3-2). The greatest difference by species was with the pocket gopher (87% adult and 13% subadult), but the percentage of adults versus subadults and young also were high for brocket deer (Mazama americana [Cervidae]; 75% adult and 25% subadult) and white-tailed deer (67% adult, 25% subadult, and 8% young). Adult birds also were taken more frequently than subadults or young overall and for each of the four taxa (83% adults, 17% subadults, and 0% young (Table 3-2). Few subadult or young birds were taken. Reproductive condition. Reproductively active females were harvested by hunters. Among mammals, none of the pocket gophers (n = 33 females; see Table 3-2 for numbers of females) or white-tailed deer that were harvested were determined to be gravid. Pregnant white-lipped peccaries (100% with embryos or fetuses), agoutis (27%), pacas (20%), brocket deer (17%), collared peccaries (15%), and coatis (2%) were noted (Figure 3-1). About 32% of the white-tailed deer and 20% of the pacas taken were lactating. Among birds, none of the great curassows or ocellated turkeys that were harvested were determined to be gravid. Thicket tinamous (25%; see Table 3-2 for numbers of females) and plain chachalacas (5%) with eggs were noted (Figure 3-1). None of the birds taken had a brood patch. The reproductive condition of many birds and mammals was impossible to determine due to damage that resulted from stomach shots by hunters. Distribution of take by month. The take of game animals by month varied during the study. For both mammals (x^ = 157.54, d.f. = 16, P < 0.005) and birds (x^ = 28.56, d.f. = 6, P < 0.005), there were significant differences among months. Game species generally were taken throughout the year, but the majority of individuals of a taxon tended to be taken by hunters during shorter periods of three-five months each (Figures 3-2 & 3- 3). It is important to note the length and timing of these periods were different for each species. The 74 100 03 Cl, K 40- Ohi Apa Dpu Nna Tpe Tta Mam Ovi SPECIES Cci Ou Ove Aoc I GRAVE) ^3 UCt/HDOa I I NOT G/I/B ^S UHOOWK Figure 3-1. Reproductive status of female birds and mammals. Animals were categorized as gravid, lactating/brooding, not gravid/lactating/brooding, or unknown (abbreviations correspond with scientific name; Ohi = pocket gopher, Apa = paca, Dpu = agouti, Nna = coati, Tpe = white-lipped peccary, Tta = collared peccary. Mam = brocket deer, Ovi = white-tailed deer, Cci = thicket tinamou, Cru = great curassow, Ove = plain chachalaca, and Aoc = ocellated turkey). coati, for example, was taken primarily during 9/90-10/90 (48% of the take during two months), and to a lesser extent during 11/89-3/90 (37% of the take during five months; Figure 3-2D). The brocket deer and white-tailed deer also followed this pattern in that take was relatively high during a 2-4 months and then low or absent during the remaining months (Figure 3-3C & 3-3D). Birds were taken primarily during 1/90-7/90 (Figure 3-4). For the thicket tinamou, great curassow, and the ocellated turkey take was primarily during one-two month periods that differed between species. Take of the plain chachalaca was primarily during 1/90-3/90 (50% of total), but data collection for this species was discontinued after 7/90. Gravid female mammals were taken throughout the study. Agoutis with embryos or fetuses were noted during 5 months (January, March, April, June, and August 1990), while gravid pacas were 75 I INDIVIDUALS 0"^ — I ' r ' I ' 'i ' ' I ' I I I ' I ' ' I ' I ' I ' 'i ' JJA80NDJFMAMJJA80 MONTH A) POCKET GOPHER (hi -53) 20 18 10 % INDIVIDUALS n n JJA8ON0JFMAMJJA80 MONTH B) PACA (N ■ 47) « INDIVIDUALS K INDIVIDUALS JJA30NDJFMAMJJAS0 MONTH C) AGOUTI (N ■ 35) JJA80NDJFMAMJJASO MONTH D)COATI(N'167) Figure 3-2. Reported monthly harvest levels for a) pocket gopher, b) paca, c) agouti, and d) coati. 76 100 80 60 40 20 « INDIVIDUALS % INDIVIDUALS JJA80NDJFMAMJJA80 MONTH A) WHITE-LIPPED PECCARY (N- 3) JJA80N0JFMAMJJA80 MONTH B) COLLARED PECCARY (N ■ 40) 20 15 10- 5 % INDIVIDUALS % INDIVIDUALS 0 I ' r I ^ ' I I — I— ~i — I I ' I ' 'i ' ' I ' 'r I ' I ' ' I JJA90NDJFMAMJJAS0 MONTH C) BROCKET DEER (N ■ 16) ' r ' I ' ' r ' r 'I ' i JJA30NDJFMAMJJAS0 MONTH D) WHITE-TAILED DEER (N ■ 24) Figure 3-3. Reported monthly harvest levels for a) white-lipped peccary, b) collared peccary, c) brocket deer, and d) white-tailed deer. 77 40 30 20 10 » INDIVIDUALS 0 I I I I I I I ' I " I ' ' r ' I ' 'I JJAS0NDJFMAMJJA30 MONTH A) THICKET TINAMOU (N • 13) 50 40 30 n 10 K INDIVIDUALS Q-'-i — I — I — r \' '\' — I ' r I I ' r I — r— r JJA8ONDJFMAMJJA80 MONTH B) GREAT CURASSOW(N« 13) U 20 16 10 % INDIVIDUALS n T*— I — I— I — I — I — I — ' I " r ' r ' I ' ' r ' I ' '1 JJASONDJFMAMJJASO MONTH C) PLAIN CHACHALACA(N '1671 36 % INDIVIDUALS 30 N 20- 16 10 6 0^" r— ; — I ' r ' r 'I ' I ' I ' I I I — I I ' I JJASONDJFMAMJJASO MONTH D) OCELLATED TURKEY (N • 6) Figure 3-4. Reported monthly harvest levels for a) thicket tinamou, b) great curassow, c) plain chachalaca, and d) ocellated turkey. 78 recorded during 4 months (November 1989, January, April, and September 1990; Figure 3-5). Gravid females for other species of mammals were recorded only during 1-3 months, each. Among birds, plain chachalacas were encountered with eggs during 3 months (January, April, and May 1990), while the thicket tinamou was recorded with eggs during 1 month (May 1990; Figure 3-6). No eggs were encountered in the other species of birds. Distribution of take by time of day. The distribution of take by time of day for game animals varied. For both mammals (x" = 289.12, d.f. = 23, P < 0.005) and birds (x^ = 444.90, d.f. = 23, P < 0.005), there were significant differences among hours. This suggested that there were times of the day when mammals and birds were more likely to be taken than during other times. Mammals were taken during 22 of the 24 hours of the day (Figure 3-7). The greatest number of mammals was taken during 0800-0859 h when 56 (15% of mammals) individuals were taken. Mammals usually were taken during 0600-1759 h when 325 (84%) individuals were taken. This period roughly corresponded to daytime. Birds were taken during 17 of the 24 hours. Take of birds occurred primarily during two periods: 0600-0859 h, when 109 (55% of birds) individuals were taken, and 1400- 1759 h, when 48 (24%) individuals were taken. These two periods corresponded to early morning and late afternoon. For the seven intervals, there was a significant difference in the proportion of take by time interval between mammals and birds (x" = 166.17, d.f. = 6, P < 0.005). This suggested that the times when mammals were taken were not the same times as when birds were taken. The distribution of take by time of day varied by species. The ocellated turkey was taken primarily at dawn (50% of take during 0400-0659 h). Two species, pocket gopher (100% during 0400- 1659 h) and plain chachalaca (71% during 0400-1159 h), were taken primarily at dawn and during the day. The coati, white-lipped peccary, collared peccary, brocket deer, thicket tinamou, and great curassow, were taken primarily during the day (> 50% during 0700-1659 h). One species,agouti, was taken primarily at dusk and during the day (89% during 0700-1959 h). The paca (70% during 1700- 0359 h) and white-tailed deer (63% during 1700-2359 h) were taken primarily at dusk and during die night. 79 o I oo 0\ ■a a o C/3 o O u > •3 u C 3 o "S ■a & '•3 o ^j t^ u SJ 'D .! r3 O s 1) .9 »« S o o li > 3 •o .5 60 ^ o I 3 SO 80 ^-v < o 0\ ON dv s oc o\ ^"^ b ■s M s ^^ Q Z o c/; O V5 O PQ 3 O O o o a M •3 lO O ca a. u O CO O 3 c o •a o 6 so OO u c > o w o ^ k> O 03 T3 C o o > 8 B 3 U o •3 O a \^ 3 60 81 NO. INDS./HOUR INTERVAL O'M'I r'l r'P'P'P'F'piF'F'F'F'F'P'F'F'plMlniMin 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 ?? TIME OF DAY MAMMALS (N - 385) BIRDS (N - 199) Figure 3-7. Number of game birds and mammals taken per hour. Cliaracteristics of the Hunters and their Weapons Age and identification of hunters. The 1989 population of X-Hazil Sur was 950 residents, 479 males and 471 females in about 200 households (Dr. Juan Chi and Dr. Norbierto Ramirez Morales, in litt.). A total of 86 hunters (84 men from about 70 households and 2 women from 2 households) reported taking game. One woman was 16 y old and reported taking only one collared peccary, while the other woman was 36 y old and reported taking only one agouti. The male hunters ranged in age from 10 to 60 y old (J = 29.1 y, SE = 1.2 y, n = 84; median = 28 y) and constituted 18% of the male population. The distribution of hunter's ages differed from that of the male population of X-Hazil Sur as a whole. The age distribution of males at X-Hazil Sur was pyramidal in shape with relatively few men in the older age classes and proportionally more in the younger age classes (Figure 3-8). However, the age distribution of hunters at X-Hazil Sur was different in that 79% of the male hunters were in age 82 classes 15-39 y, versus 40% of the total male population. This suggested that there is an age range when men practice hunting, but outside of that range few continue the activity. Age Class % RESIDENTS % HUNTERS Figure 3-8. Age class (years) distribution of male residents (539 residents based on 1992 census) and male hunters (84 hunters during 1989-1990 study). The number of game kills per hunter varied greatly. The mean number of kills per hunter during the 17 month period was 7.0 (SE = 1.4; n = 86). Hunter number 35 took the greatest number of animals at 85 individuals, while 35 hunters reported only 1 kill each (Figure 3-9). Seven hunters reported 27 or more kills each and accounted for 54% of the total number of individuals taken (Appendix E). There was a significant difference in tlie mean prey weight among the seven main hunters (x" approximation = 96.15, d.f. = 6, P < 0.0001). Hunter # 6 had the greatest mean prey weight (J = 7.8 kg, SE = 1.4 kg, n = 61), followed by hunter # 1 (J = 3.8 kg, SE = 0.7 kg, n = 44) and 83 100 80 60 40 20 NUMBER OF PREY 1 1 1 1 nnnnnilnnnnnnnnnnnnnnnnnnn 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 HUNTERS Figure 3-9. Number of kills per hunter. A total of 47 hunters had only one or two kills each and are not shown here. hunter # 35 (J = 3.2 kg, SE = 0.3 kg, n = 85; Appendix E). This suggested that even among the main hunters there were substantial differences in the size of the prey they harvested. Hunter group size. Hunters pursued game individually or collectively. The mean group size of hunters was 1.7 individuals (SE = 0.0.05, n = 417). Individual hunters conducted 58% of the hunts and took 55% of the game by number, while hunters in groups > 5 people conducted 3% of the hunts and took 3% of the game (Table 3-3). The mean number of prey taken per hunting outing ranged from 1.3 to 1.8, but there was no significant difference among these values with respect to hunter group size (x^ approximation = 5.6909, d.f. = 4, P < 0.2235; Table 3-3). There was, however, a significant difference between mean number of prey taken per hunter hour with respect to hunter group size (x^ approximation = 109.66, d.f. = 4, P < 0.0001). Hunter groups with a single hunter harvested 0.4 prey per hunter hour, while larger groups harvested game at less than half that rate. This suggested that one-two prey 84 are taken per outing and that single hunters take game at a higher rate per hunter hour than hunters in larger groups. Types of weapons. Hunters used various types of weapons to take game. Approximately 39% of the game was taken by 22 caliber rifles. Game also was taken by traps (9%), shotguns (23% by 16 gauge shotguns, 19% by 20 gauge shotguns, < 1% by 12 gauge shotguns), and methods classified as unknown or other (e.g., dogs, rocks, and n\achetes; 10%). Rifles and shotguns have been used by hunters at X-Hazil Sur since the early 1900s, when the area was settled (M. Aban Noh, pers. comm.). With the exception of traps, snares, and machetes, none of the older hunters could recall what kinds of weapons their ancestors used. Rifle hunters now primarily used long-rifle and long-rifle hollow point shells (E. Cab Can, pers. comm.). Shotgun hunters primarily used 3F (9 balls per cartridge) and 4F buckshot (12 balls per cartridge; B. Can Chi and M. Aban Noh, pers. comm.). This size of buckshot was sufficiently powerful to kill a deer at 40 m distance in an open field. Hunters generally did not use a specific type of weapon to take a specific type of game. With the exception of the white-lipped peccary (Tayassu pecan [Tayassuidae]; 100% taken by 20 gauge shotgun) and the pocket gopher (100% taken by traps or other/unknown), game taxa were taken with several types of we^ons (Figure 3-10). For the white-lipped peccary, brocket deer, collared peccary, and paca, shotguns accounted for 62-100% of the kills per taxon. For the coati (58% by individuals) and agouti (54% by individuals), rifles and weapons categorized as other (especially dogs and machetes) accounted for more than half of the kills per taxon, but shotguns also were used. White- tailed deer were taken by shotguns (83%) and weapons classified as other/unknown (17%), but not by rifle. Among birds, the ocellated turkey (100% by individuals) and the great curassow (69% by individuals) were taken primarily by shotguns, while the plain chachalaca (75% by individuals) and thicket tinamou (62% by individuals) were taken primarily by rifle. 85 A) MAMMAL TAXA (BY BODY MASS) Tpe Ovi y///////////////A Mam - Tta Nna OhI ^-y///////A ^ 1^ ^ ^/////////^i^mxx^y^^^^:^^ ^ -y///////////Am$m$i$im 0% 2S% 50% 75% 100% B) BIRD TAXA (BY BODY MASS) Aoc Cru Ove Ccl - 0% 25% S0% 75% % INDIVIDUALS 100% WEAPONS mi 12 GAUGE ^ZZl 16 GAUGE ^ 20 GAUGE ^ 22 CAL RIFLE LZl OTHER/UNK. Figure 3-10. Take of mammals and birds according to type of weapon used. Species arranged by mean body mass (see Table 3-1 for body mass; se Figure 3-1 for key to species abbreviations). 86 There was a significant difference in the mean prey weight for each weapon type (x^ approximation = 167.49, d.f. = 5, P < 0.0001; Table 3-4). The 12 gauge shotgun, though used only twice, had the greatest mean prey weight (J = 21.4 kg, SE = 18.65, n = 2), while the 20 gauge shotgun had the second greatest mean prey weight (J = 10.4 kg, SE = 1.06, n = 110). This suggested that shotguns were used to take the heaviest prey, while 22 caliber rifles and traps were used to take the lightest prey. Characteristics of the Hunt The typical hunt was a short foray that lasted only a few hours, took place within the ejido, and involved only a minimal amount of preparation. Hunters generally used bicycles to move about and carry game. Except for hunters working overnight at chicle camps or those working at gardens or ranches located far from the village, game was brought back to the village. Hunting outings. Hunters reported taking game on 419 hunting outings (Table 3-5). The tracking/stalking method was used by successful hunters about 79% of the time, by outing, and comprised 64% of the time, by duration of hunt. The total duration of successful hunts was 2,858.1 h (J = 7.1 h, SE = 0.33, n = 409). The mean group size was 1.7 hunters and was similar for the three hunting methods: tracking/stalking, stand/platform, and trapping. The mean number of game animals taken was 1.39 individuals per outing and 0.29 individuals per h. About 82% of the game was taken during tracking/stalking hunts. Hunting yield was compared for two methods; tracking/stalking and stand/platform. There was no significant difference between the mean number of prey taken per outing (x" approximation = 0.4632, d.f. = 1, P < 0.4961) and the mean number of prey taken per hunter hour (x^ approximation = 0.6791, d.f. = 1, P < 0.4099). This suggested that each of the two methods, where hunters were specifically involved in hunting and were not distracted by other activities, were equally effective in obtaining game. 87 1 IS I X s 3 60 B ■a 3 o U s ■a a. s o c 9 X 2 a h o ? § " o d d d ^ o — Tt r- 00 o o o ^^ VO ON (N VO m CN IT) VD (N ^ t~ 00 ts TT r- vo o\ oo r- w-i u-i r- ^ IT) s ,.-v 00 00 o r<^ 00 Wl o J3 ■* ^ 00 q onl 3 V5 C 1 d d c oo o\ 1 p- o H o »o vd 00 d vd «3 i"^ ^ 00 c 13 •3 e2 c o 3 00 vq en lo o r-« pi-i -o d ^ to r^ 00 >o t~ VD 00 iri r-- ts ^-H D c 1-^ -o ■a fS VD ^ cs en fS OS 00 1/^ en en 00 in 00 O •3 o H Ik 13 oo s a 3 O 3 13 > so s 3 O o 60 1) 3 o 00 o it 60 60 88 Table 3-4. Mean prey body weight (kg) by type of weapon for game species taken at X-Hazil Sur. Prey Type of weapon weight (kg) SE n 12 gauge shotgun 21.4 18.65 2 16 gauge shotgun 5.9 0.73 130 20 gauge shotgun 10.4 1.06 110 22 caliber rifle 1.8 0.17 226 Traps 0.6 0.14 51 Other/unknown 5/7 1_31 57 Total 2700.1 576_ r 4.7 0.34 The number of game animals taken per outing generally was low. During about 75% of the outings, only a single mammal or bird was taken (Table 3-6). The greatest number of mammals taken in a single outing was seven (all coatis), while the greatest number of birds taken in a single outing was four (all plain chachalacas). Mixed bags, containing both mammals and birds, were reported for only nine outings. Kill site locations and vegetation types. Game birds and mammals were not taken in the four vegetation types in the proportion in which these types occurred (Appendix G). For five of the eight mammalian taxa and all of die avian taxa, take was primarily in a single vegetation type, while the remaining three mammalian taxa frequently were taken in two or three vegetation types. When analyzed descriptively, several generalizations were noted. For mammals, the white- lipped peccary (100.0% of kill sites), collared peccary (58%), and brocket deer (50%) were taken primarily in sites categorized as Late Secondary Forest (Appendix G). Sites categorized as Plots & Gardens were the main kill sites for two taxa, the white tail deer (63%) and agouti (49%), but die coati (47%), pocket gopher (42%), and paca (34%) also were frequently taken in these areas. For birds, the great curassow (92%) and thicket tinamou (77%) were taken primarily in sites categorized as Late Secondary Forest, while the plain chachalaca (87%) was taken primarily in sites categorized as Early 89 flj c/5 O QJ C S S J3 3 ^ c ° so ^ 3 2 h 31 i 2 ^ -I oj a o S 5 ^- SQ iH 2 3 :j g S ■-' o c o •a O •a ^ S3 a 3 H re oj =« H 5 o d ON 00 r- 00 00 o UI 00 00 o o o \D q 00 ON O f1 oo — U-1 c^ ■^ fS 60 s He/: ;/2 CL, o\ (N d a, 3 o o so r<-i 2 c 3 •a 1 ■* O 00 •3 U-i SO c ■a 3 o fS 3 ^' T— < m v: SO 3 1 s O d u o s 3 S 2 S5 r~ .- s in ( ) ind nt category ,.^ ,.^ U 3 oo' r-^ ■^ C- in 00 > o fS outings, each type ^ .s o w ^ Tl- O 60 s-^ 0= c On "O ■S »— t S 3 ■t ling duration was not obtain^ Dwn if not available for all o 1 o5 H IK ' 00 -- — 1 90 Table 3-6. Combinations of game mammals and birds taken per outing by hunters at X-Hazil Sur. Number of birds/outing Total'' 0 1 2 3 4 Total' 0 95 20 12 4 131 1 218 5 3 226 2 43 1 44 Number of mammals/ outing 3 4 5 6 7 10 3 2 2 1 10 3 2 2 1 279 101 23 12 419 ' Number of outings with from 0 to 7 mammal kills per outing. *" Number of outings with from 0 to 4 bird kills per outing. Secondary Forest and the ocellated turkey (50%) was taken primarily in sites categorized as Plots & Gardens (Appendix G). When analyzed statistically, the distribution of take, by species, for seven species of game for Late Secondary Forest versus the other three vegetation categories combined was significantly different from that expected, based on the proportions of those two vegetation types in the ejido (x^ = 23.4- 358.1, P < 0.005; Table 3-7). For the pocket gopher, paca, agouti, coati, collared peccary, brocket deer, and plain chachalaca, a significantly greater proportion of the take took place in areas categorized as Plots & Gardens, Early Secondary Forest, or Other/unknown than would be expected. Among the species with an adequate sample size to permit testing, no species was taken in Late Secondary Forest at the expected frequency. Minimum catchment area and mean kill site distances. The minimum catchment area over which hunting occurred could not be precisely determined because only successful outings were reported and the total extent of outings was not always reported. Recognizing this, perhaps the best 91 estimate for minimum catchment area would be the actual size of the ejido (552.95 km^; O. Guatemala Biempica, in litt.) as 581 of the 584 kills were on the ejido. The problem with using this value is that hunters from X-Hazil Sur generally hunted only in the western and southern two-thirds of the ejido (ca. 370 km^), while hunters from the villages of Uh-May and Chancah Veracruz (not surveyed during this study) usually hunted in the northeastern one-third of the ejido. The minimum harvest area over which game was taken differed among game species. The two largest harvest areas were observed for the coati (251.8 km^) and the collared peccary (211.5 km^), while the smallest harvest area observed was for the ocellated turkey (4.8 km^ Table 3-8). This suggested that species such as the coati and the collared peccary were widely distributed and harvested in the ejido, while others, such as the ocellated turkey, either had a limited distribution or were widely distributed, but harvested only in a limited area. On a more general level, this suggested that game was taken on only about 45-67% of the study area. There was a significant difference between the mean kill site distances by species (x^ approximation = 58.924, d.f. = 11, P < 0.0001). The three species with the greatest mean distances between the village and the kill sites were birds: ocellated turkey (7.6 km), great curassow (7.5 km), and thicket tinamou (6.6 km), while the pocket gopher (3.4 km) had the smallest mean distance (Table 3-8). The ocellated turkey had the greatest minimum kill site distance (5.6 km), followed by the great curassow (2.5 km). This suggested that, on average, game birds were taken at greater distances from X-Hazil Sur than were game mammals. The greatest distance between a kill site and X-Hazil Sur was 17.5 km for a collared peccary, while four species were taken within 1 km of the village (Table 3-8). Twenty-five percent of all kills were within 2.5 km of X-Hazil Sur, 50% within 4.8 km, and 75% within 7.2 km. 92 Table 3-7. Percent frequency of game kill sites in Late Secondary Forest versus all Combined/Early Secondary Forest and x^ test results for the game taken by hunters at X-Hazil Sur. Game species Percentage of kill sites by vegetation type Combined/ Late Early Secondary Secondary Forest Forest (88.5%)" (11.5%)" ^ P n 5.7 94.3 358.1 < 0.005 53 34.0 65.9 137.3 < 0.005 47 31.4 68.6 112.3 < 0.005 35 44.9 55.1 312.5 < 0.005 167 100.0 0.0 c 3 57.5 42.5 37.9 < 0.005 40 50.0 50.0 23.4 < 0.005 16 12.5 83.3 c 26 a) Mammals Pocket gopher Paca Agouti Coati White-lipped peccary Collared peccary Brocket deer White-tailed deer % Manmials combined 36.9 63.2 1010.2 < 0.005 385 b) Birds Thicket tinamou Great curassow Plain chachalaca Ocellated turkey 76.9 23.1 92.3 7.7 6.0 94.0 33.3 66.7 1119.4 < 0.005 13 13 167 6 % Birds combined 17.1 82.9 999.3 < 0.005 199 % All game species combined 30.1 69.7 1958.9 < 0.005 584 ' Values in ( ) indicate the percentage of the study area in each vegetation type. Total area = 552.95 km-. *" The category Combined/Early Secondary Forest was composed of the vegetation types classified as Early Secondary Forest, Plots & Gardens, and Other. ' X^ test not performed because at least one of the two expected cells had < 5 observations. 93 Table 3-8. Minimum harvest area (km-) and mean distance (km) between X-Hazil Sur (water tank) and kill sites of primary game species taken by hunters at X-Hazil Sur. Minimum Kill site distance (km) harvest Game species area (km^) j SE Range n a) Manmials Pocket gopher Paca Agouti Coati White-lipped peccary Collared peccary Brocket deer White-tailed deer b) Birds Thicket tinamou Great curassow Plain chachalaca Ocellated turkey All species combined 5J 0.17 0.3-17.5 583_ ° All three individuals killed at a single site. *" n = 23 instead of 24 due to one unknown kill site. 66.8 3.4 0.41 0.3-16.4 53 131.7 5.6 0.49 1.2-15.8 47 54.7 4.3 0.50 0.3-11.5 35 251.8 5.8 0.26 1.4-16.6 167 a 4.0 0.00 4.0 3' 211.5 6.3 0.62 1.3-17.5 40 89.3 6.5 1.25 1.3-16.4 16 63.0 4.6 0.90 0.7-16.4 23" 73.5 6.6 1.33 1.0-16.8 13 59.0 7.5 1.21 2.5-16.2 13 112.0 5.5 0.40 0.4-16.3 167 4.8 7.6 0.89 5.6-11.3 6 94 Table 3-9. Minimum catchment area (km^) and mean kill site distance (km) from X-Hazil Sur (water tank), by hunter, for 315 game animals taken by the seven main hunters versus all other hunters combined at X-Hazil Sur. Minimum Kill site distance (km) catchment Hunter No.' area (km^) X SE Range n_ 35 53.0 4.9 0.22 1.2-12.0 85 6 28.3 5.0 0.34 0.4-10.0 eo" 1 37.3 5.4 0.35 1.0-9.2 44 18 51.9 2.9 0.29 1.2-9.8 40 9 68.1 4.3 0.84 1.1-16.2 31 62 7.2 4.3 0.64 0.5-11.3 28 98 45.1 13.3 1.04 1.8-16.3 27 All other 5.5 0.17 268 hunters All hunters combined 5.6 0.27 0.3-17.5 583 " Arranged in rank order by number of kills. '' n = 60 instead of 61 due to one unknown kill site. The minimum catchment areas differed among the seven main hunters. The individuals with the largest catchment areas were hunter # 9 (68.1 km^), hunter # 35 (53.0 km^), and hunter # 18 (51.9 km^), while hunter # 62 (7.2 km^) had the smallest minimum catchment area (Table 3-9). There was a significant difference in mean kill site distances by hunter for the seven main hunters (x^ approximation = 83.786, d.f. = 6, P < 0.0001). The individual with die largest mean distance between the village and the kill sites was hunter # 98 (13.3 km), while hunter # 18 (2.9 km) had the smallest mean distance (Table 3-9). Use of dogs. Dogs assisted hunters by locating or killing game. Seventeen hunters reported using a total of 59 named or individually recognizable dogs. Dogs accompanied hunters on 72 outings (17% of outings) and a total of 127 game animals were taken. The mean number of dogs per outing for this subset was 3.9 (SE = 0.19, n = 72). The coati (95 individuals) was the most frequently taken 95 game species when dogs accompanied hunters, followed by the paca (12), agouti (7), collared peccary (7), plain chachalaca (3), brocket deer (1), white-tailed deer (I), and great curassow (1). Five of the seven main hunters reported using dogs (hunters # 6, 18, 35, 63, and 98), while two did not (hunters # 1 and 9). Discussion Composition and Characteristics of the Harvest Maya Indians have practiced subsistence hunting for at least several thousand years. This study demonstrated that although the Maya of X-Hazil Sur, Quintana Roo, Mexico, have become highly acculturated, many still practice subsistence hunting. In addition, this study also demonstrated those species of game that were hunted during 1989-1990 were essentially the same as those hunted by the prehistoric Cerro Brujo people about 1,000 y ago (cf., Linares, 1976). This suggested that subsistence hunting continues to be an important social and economic activity for the Maya. It also implied that the hunting was conducted in such a way that the wildlife resource was not extirpated. Thus, Maya hunters and the techniques they employed may provide insight into how to manage hunting and wildlife populations in areas where the game has been extirpated or seriously depleted. Number of species and individuals. The primary game species for Maya hunters at X-Hazil Sur were mammals and birds. Some fish were harvested, while reptiles, amphibians, and insects were ' not taken for subsistence purposes, except that honey was consumed when encountered. This contrasted with other indigenous groups in Mexico that-in addition to birds and mammals-took substantial quantities of fish, insects, reptiles, and amphibians. For example, Lacandon Maya Indians in the state of Chiapas consumed a wide variety of amphibians and reptiles (Galleti, n.d.; Gongora- Arones, 1987; March M., 1987), while Mixteca Indians in the state of Oaxaca (Parra Lara, 1986) consumed large quantities of iguanas (Ctenosaura pectinata) and crabs (Cardisoma crassum). Mestizos throughout Mexico also consumed a wide variety of fish, reptiles, and crustaceans (Mellink et al., 1986; Reyes Castillo, 1981; Santana et al., 1990). 96 Elsewhere in the Neotropics, indigenous and mestizo groups also consumed prey other than manmials and birds. For example, iguanas (Iguana iguana) and river turtles (Dermatemys spp. , Kinostemon spp., and Pseudemys spp.) were taken by Maya hunters in Belize for meat (Frost, 1974, 1977). In Nicaragua, Miskito Indians consumed large quantities of fish (e.g., Arius melanopus and Bagre marinus), shrimp (Penaeus spp.), and marine (Chelonia mydas and Eretmochelys imbricata) and freshwater (Pseudemys spp.) turtles (Nietschmann, 1972, 1973, 1979). Fish (27 spp.) and turtle and iguana eggs were taken by Creole mestizos in Costa Rica (Frost, 1974). In Guatemala, Maya Indians consume ants (J. E. Jimenez, pers. comm.). Fish, insects, reptiles, and amphibians were also widely taken by mestizos and indigenous peoples in South America (Beckerman, 1983; Chemela, 1985; Dufour, 1987, 1990; Hames, 1979; Hill and Hawkes, 1983; Posey, 1987; Redford and Robinson, 1987; Slocks, 1983; Vickers, 1984). These studies suggested that Maya hunters at X-Hazil Sur used fewer classes of animals as game than were used by other Indian or mestizo groups in the Neotropics. Among the birds and mammals taken by X-Hazil Sur hunters, relatively few species were harvested compared with the number of species taken by other indigenous peoples. Maya hunters In Quintana Roo took four species of birds and eight species of mammals. Miskito Indians in Nicaragua also took relatively few prey species (about 25 species; Nietschmann, 1972), while Siona-Secoya Indians in Ecuador (Vickers, 1991), Ye'kwana and Yanomamo Indians in Venezuela (Hames, 1979; Saffirio and Hames, 1983), Yuqui Indians in Bolivia (Stearman, 1984, 1990), and Ach6 Indians in Paraguay (Hill and Hawkes, 1983) each took about 50-60 species of game. Clearly, Maya hunters at X-Hazil Sur took fewer species of mammals and birds as game compared with other indigenous groups. _ While a subsistence hunter's diet would depend in great measure upon the potential prey species available, the concentration upon a few species birds and mammals by Maya hunters at X-Hazil Sur, although other potential game taxa also were found in the area, but not taken, suggested at least three possibilities; one, Maya hunters were quite selective in the prey they took; two, the other taxa were not available; and three, hunters were able to supplement their diet by purchasing canned meat and caring for domestic animals. 97 The possibility that Maya hunters were selective in the game they harvested cannot be addressed fully as the study was not designed to measure hunter game preferences. The coati and plain chachalaca were the two taxa most frequently taken by hunters at X-Hazil Sur, by number of individuals, but there was no indication that these species were preferred by hunters. Other, less frequently taken species, such as the ocellated turkey, great curassow, brocket deer, white-tailed deer, and white-lipped peccary, however, were highly prized by Maya hunters, but rarely taken due to their scarcity and difficulty with which they were encountered. If given a choice, most hunters at X-Hazil Sur indicated that they would prefer to harvest a great curassow, brocket deer, or white-tailed deer. These choices correspond with Maya game preferences throughout the Yucatan Peninsula (Bums, 1983; Medi'z Bolio, 1987). The reasons why these taxa were preferred by hunters included the large amount of meat obtained from the deer (see Table 3-1) and the skill necessary to obtain a great curassow due to its reclusive habits (see Table 3-8 and Appendix G). The possibility that many game species widely taken elsewhere in the Neotropics, but not by Maya hunters at X-Hazil Sur, suggest cultural differences among indigenous groups and will be discussed more fully in Chapter 4, where data about the number of game species and their densities in the study area will be presented. This will permit a comparison of the availability of game with the harvest of game. However, data for plants (Flores-Villela and Gerez-Femandez, 1989) and reptiles (Lee, 1980) suggested that species diversity and density may be naturally low in the Yucatan Peninsula due to several abiotic factors, including rainfall gradients, recent geologic age, and relatively flat topography. These factors also may have influenced Maya hunters through the number of species of birds and mammals on the peninsula and their population densities. The relatively small number of game species taken by hunters at X-Hazil Sur was unexpected when compared with (1) the archeological record and (2) the other species of birds and mammals found locally, but not taken for subsistence purposes. This evidence suggests a change in Maya hunting patterns over time. According to archeological studies at Cozumel, a large island off the northeastern 98 coast of Quintana Roo, the following taxa were taken by the Maya during 100-1500 A.D., prior to the arrival of the Spanish: about 40 taxa of fishes and aabs, 4 taxa of amphibians (frogs and toads), 10 taxa of reptiles, 38 taxa of birds, and 16 taxa of mammals (Hamblin, 1984, 1985; Hamblin and Rea, 1985). While many of these taxa were found in ceremonial and administrative buildings and may have been taken for these purposes, the fact that others were found in housemounds suggested that these specimens were taken for subsistence purposes. Although Cozumel is located on the Caribbean Sea and X-Hazil Sur is located about 50 km inland, a comparison between sites is appropriate because of their proximity (ca. 175 km), similar climate and vegetation, and settlement patterns. Other archeological studies in the Yucatan also indicated that the diet of the prehistoric Maya was much more diverse than that of present day Maya (Wing, 1974; Wing and Steadman, 1980). The relatively small number of game species taken by hunters at X-Hazil Sur was also unexpected when compared with the other species of birds and mammals found locally, but not taken for subsistence purposes. For example, the following mammals occur in Quintana Roo (Leopold, 1977; Navarro et al., 1990) and are widely taken for subsistence purposes throughout the Neotropics (e.g., Hames, 1979; Stearman, 1990), but were not taken by Maya hunters at X-Hazil Sur; squirrel {Sciurus spp.), armadillo (Dasypus novemcincius), tapir (Tapirus bairdii), howler monkey (Alouatta pigra), spider monkey (Ateles geoffroyi), kinkajou (Potosflavus), tayra (Eira barbara), and opossum (Didelphis virginiana). This pattern of occurring in Quintana Roo, but not being used for subsistence purposes also was the case for many taxa of birds. Hunters at X-Hazil Sur, for example, generally did not take parrots (Psittacidae) or doves (Columbidae) for subsistence purposes, but these taxa occur at X-Hazil Sur and are widely hunted throughout the Neotropics. Parrots and doves at X-Hazil Sur, however, were killed as noxious animals when they fed on crops. That these species were present at X-Hazil Sur, but not harvested, could suggest that they were not acceptable to Maya hunters or that these kills were not reported during the study. More likely, however, the limited number of game species taken was sufficient to meet the subsistence needs of hunters. 99 When asked why certain wildlife species were harvested and others not harvested, hunters at ^ X-Hazil Sur replied with a variety of responses, including: "too big" (e.g., tapir), "too small" (e.g., squirrel), "the meat doesn't taste good" (e.g., tayra and opossum), and "its not right to kill a harmless animal" (e.g., armadillo and monkeys). Taboos did not appear to explain any of these statements, and y^i/ responses varied among hunters. The availability of canned meat and domestic animals also will affect the number and variety of animals taken by hunters in at least two ways. One, hunters can depend on purchased meat to compensate for outings when no game is taken. This probably will reduce the number and variety of game taken as the hunter will not necessarily go hungry that day. Two, hunters can evaluate the cost and desirability of killing a wild animal versus purchasing meat. This also should reduce the number and variety of game taken by giving the hunter the choice on not taking relatively small or poor-tasting species. Given the availability of canned meat and domestic animals, and recognizing the low number of kills per hunter (ca. 7 per hunter during 17 months), subsistence hunting at X-Hazil Sur was quite opportunistic. Sex and age class of game species. Information about the sex and age classes of game is important in evaluating hunter game preferences, the structure of local game populations, and the likelihood that one specific sex or age group might be more vulnerable than another group to hunters. These elements are important in evaluating game use by subsistence hunters. At X-Hazil Sur, the proportions of male and female birds and mammals harvested by hunters differed. For mammals, more females were taken by hunters than males for all species except the brocket deer, while for birds, more males were taken by hunters than females for three of the four species (equal numbers of males and females taken for the ocellated turkey). Hunters at X-Hazil Sur attributed this to additional nutritional demands on female mammals while suckling their young and the consequent greater use of gardens to obtain food, not on any selection for females by hunters. Maya hunters also indicated that when feeding in gardens, females accompanied by their young were less vigilant than normal. This was especially true for female coatis and white-tailed deer, two species 100 frequently taken in gardens when lactating. Male mammals, according to hunters, were less vulnerable than females to hunters for behavioral reasons; adult males tended to occur individually and could quickly and quietly escape, while adult females tended to occur in groups that included young and other females. Large groups with young, according to hunters, could not escape detection easily as the young frequently made loud noises as they foraged or attempted to flee hunters. Male birds, according to hunters, were easier to take than females because males often displayed from tree tops. While displaying, males were highly visible and less wary of hunters than they normally were. This suggested that female mammals and male birds were more frequently taken as a consequence of their behavior, including different habitat use patterns between sexes, and not as a result of active selection by hunters. At X-Hazil Sur, birds and mammals exhibited similar patterns of take by age in that adults composed > 70% of the harvest although young and subadults also were taken. For mammals, substantial numbers of subadults also were taken (ca. 28%), but few subadult birds were taken (ca. 17%). Hunters at X-Hazil Sur offered no explanation about these differences other than to suggest that birds matured faster than mammals. It was clear, however, that game of any age class, regardless of species, was potential prey for hunters. Only a few studies have documented the age or sex of game taken by subsistence hunters. Age and sex are summarized, however, for three studies in lowland Peru. Pacheco (1987) showed that for both mammals and birds, slightly more females than males were taken by forestry technicians conducting a tree survey, but the study was short (13 days) and sample sizes small (14 species, n = 46 game animals). At the second site, Alvard and Kaplan (1991) reported that adults composed 57% of the pacas taken, 83% of the brocket deer, and 74-76% of the peccaries. At the third site, Bodmer (in litt.) reported that males were more frequently taken than females for the collared peccary (1:0.66, males; females, n = 164 individuals), grey brocket deer {Mazama gouazoubira, 1;0.75, n = 28), and paca (1:0.57, n = 174), while females were more frequently taken than males for the white-lipped peccary (1:1.10, n = 166), red brocket deer (1:1.31, n = 60), agouti (1:1.31, n = 97), and acouchi 101 (Myoprocta spp., 1:1.17, n = 13), while take by sex was equal for the capybara (Hydrochaeris hydrochaeris, 1:1.00, n = 10). For Valley Bisa hunters pursuing large mammals in Zambia, Marks (1973) reported that hunters took more males than females (47 vs. 32 individuals) and more adults than subadults and young (79 vs. 11 individuals). As additional studies are completed, they likely will show differences in rate of take by age and sex between species taken by subsistence hunters. Reproductive condition. Information on the reproductive condition of game is important in assessing the impact of any hunter selection for particular age and sex classes of game (Redford and Robinson, 1990). If gravid, lactating, or brooding females are taken by hunters, then an evaluation of the total harvest must also consider the young that will die. At X-Hazil Sur, hunters harvested female mammals regardless of their reproductive condition. While most female mammals were not gravid or lactating, three species had relatively large proportions of females that were pregnant or caring for young: paca (40% gravid or lactating), agouti (32%), and white-tailed deer (32%; Figure 3-1). These species were among the main game species by number of individuals and total body mass (Table 3-1). Due to small sample sizes, it was not possible to assess the reproductive condition of female game birds. The lack of developing eggs or a brood patch suggest that gravid or brooding females rarely were taken by hunters. Little has been written about the reproductive condition of game animals taken by hunters in the Neotropics. In Mexico, Lazcano-Barrero et al. (1988) reported that reptiles, especially turtles and crocodiles, frequently are taken while guarding their nest. Also in Mexico, Santana et al. (1990) reported that subsistence hunters at the Sierra de Manantlan Biosphere Reserve refrained from hunting during specific periods of time in order to avoid harvesting pregnant or nursing birds and mammals. For lowland Peru, Bodmer (1989) reported that reproductively active ungulates frequently were harvested by hunters, as follows; collared peccary (41% females taken were reproductively active), white-lipped peccary (62%), red brocket deer (40%), grey brocket deer (Mazama gouazoubira; 33%), and lowland tapir (Tapirus terrestris; 37%). Most likely subsistence hunters throughout the Neotropics 102 take substantial numbers of gravid, lactating, or brooding females, but until further studies are conducted the impact of this hunting on game populations cannot be assessed. Distribution of take by month. Seasonality data, when combined with information about the age and sex composition of harvested game can be used to determine whether different age or reproductive classes of game are more susceptible to hunting at certain seasons (Redford and Robinson, 1990). Harvest of game animals varied during the year following a regular pattern that was recognized by hunters at X-Hazil Sur. According to hunters, each species had a time during the year when it was more likely to be taken. The factors that influenced the timing of this period were different for each species, but according to hunters included: season (dry vs. wet), garden cycle, mating period, fruit and seed availabihty in the forest, and activities by people (e.g., chicle tapping, selective logging, and clearing new garden sites). For mammals, the harvest of brocket and white-tailed deer occurred primarily in areas categorized as Plots & Gardens and took place during the early wet season (June-July), just after crop seeds planted in the garden began to sprout. Deer also were taken during September-October. According to hunters, this was the rutting season and deer were less vigilant than normal. Coatis were harvested during the late wet season and early dry season. Beginning in September, during the late wet season, coatis were taken by hunters as they entered gardens to feed on com, and beginning in January, during the early dry season, they were harvested as they entered areas being cleared as new gardens. The harvest of pacas and agoutis also took place during the dry season and was associated with the harvest of com and clearing new gardens. Hunters often took coatis, agoutis, and pacas while logging or tapping sapodilla trees. Pocket gopher harvest took place during both seasons as hunters worked in their gardens. The harvest of collared peccaries was atypical of game mammals in that take occurred throughout the year. Gravid female mammals were taken throughout the year. The monthly distribution of pregnancies suggested that game mammals had two reproductive patterns (Figure 3-5). In the first group, species reproduced throughout the year. This group was composed of the agouti, paca, and 103 collared peccary. In the second group, species reproduced during discrete periods. For all mammals combined, the highest frequency of gravid females occurred during January-April 1990, during the dry season (see Figure 2-4). For birds, the most obvious seasonality of take was for the great curassow and the ocellated turkey. These two species primarily were taken during the early dry season. According to hunters, great curassows and ocellated turkeys seemed to be more active at this time as it was their mating period. Due to small sample sizes and the limited period over which bird data were recorded, it was difficult to assess monthly differences in the reproductive condition of birds. Eggs were reported only for the thicket tinamou (May 1990) and the plain chachalaca (January, April, and May 1990; Figure 3- 6). This period corresponded with the dry season. Seasonal differences in the composition of harvested game harvest have been reported for several long-term studies. For Bari Indians in Colombia, for example, Beckerman (1980) noted differences between months in the composition of mammalian prey taken. Each of the nine families of mammals taken during that study dominated hunter inventories for 1-3 months. During some months, no mammalian prey were taken, while in others, prey from two or three families were taken. Smith (1976) and Yost and Kelley (1983) noted a similar seasonal variation in number of animals killed for Brazil and Ecuador, respectively. These differences were attributed to several factors, including fruit availability to game animals and the need by hunters to plant or harvest a garden. These were some of the same factors mentioned by Maya hunters at X-Hazil Sur to explain monthly differences in the composition of the game harvest. Distribution of take bv time of day. Behavioral studies suggest that species are primarily either diurnal, nocturnal, or crepuscular. This pattern of discrete activity periods for game was not observed in Ejido X-Hazil y Anexos as indicated by the broad range of time over which individuals of each species were taken. Eleven of the 12 game species were taken during the day as well as during 104 the night or at dawn/dusk. The white-lipped peccary, however, was taken only during the day, but this likely was due to a small sample size (three individuals killed during a single outing). There are at least two possible explanations for this broad temporal pattern of take. One, game species in the study area may have expanded their activity periods since hunters were active throughout the day and night. Many game species in other areas are known to change their behavior during the hunting season. For example, ducks, geese, and white- tailed deer hunted on refuges and wildlife management areas in the United States are known to limit their activity until after the daily hunting period has ended. Two, the pattern of take may not reflect the species' activity period if the animal was shot while inactive or if disturbed while inactive and killed while attempting to escape. Hunters in Ejido X-Hazil y Anexos may have disturbed resting animals while enroute to logging or chicle tapping sites or when they used headlamps and dogs to locate roosting birds and or game hidden in caves or dens. These explanations suggested that the pattern of take did not reflect die activity patterns of the species, but instead reflected the activity period of hunters. Characteristics of the Hunters and their Weapons Age and identification of hunters. Among indigenous peoples in the Neotropics, subsistence hunting generally is practiced by all of the adult males of the community (cf., Stearman, 1989, 1990; Vickers, 1983, 1984, 1988). With limited exceptions (e.g., Ach6 in Paraguay [Hurtado et al., 1985]; Matses in Peru [Romanoff, 1984]), women do not participate in hunting (i.e., kill game or carry weapons). While 84 of 86 hunters at X-Hazil Sur were males, only about a third of the adult males reported taking game. Conversations with village residents suggested that most of the remaining two- thirds of the male villagers did not hunt, but this was not confirmed. Among the reasons offered by nonhunters for not hunting, especially by young men, were that they didn't have a gun; didn't know how to track or identify game sign; or didn't like the insects, rain, or late hours associated with hunting. 105 The age class distribution of known hunters at X-Hazil Sur (79% were 15-39 y old) compared with the age class distribution of all men (40% were 15-39 y old) suggested that the population of hunters was aging and that older hunters were not being replaced by younger hunters. This may seem counter-intuitive in view of the relatively young ages of the seven main hunters (see Appendix E), but conversations with both older and younger hunters indicated that few young men were interested in becoming serious hunters. The skills associated with hunting are acquired by young men over several years when they learn how to use a gun, interpret game sign, and develop the self confidence to pursue game while alone in distant parts of the ejido. Many young Maya men no longer are interested in developing these skills. This follows a recent trend among young Maya in the area to attend outside technical and secondary schools, and to seek employment in nearby towns, instead of remaining in the village and learning traditional subsistence activities. / The number of kills per hunter and the amount of game harvested per hunter indicated diat hunters at X-Hazil Sur did not take similar numbers of game. At X-Hazil Sur, the seven main hunters (8% of hunters) took 54% of the game by number and 40% by weight. This disproportionate harvest by a few hunters was similar to that reported for other indigenous groups in the Neotropics. For example, Hames (1979) reported that among the Ye'kwana in southern Venezuela, 36% of the hunters took 75% of the game by weight, while among the Yanomamo, 50% of the hunters took 85% of the game. For the Yuqui in Bolivia, Stearman (1990) reported that 25% of the hunters took 61% of the game by weight in 1983 and 32% of the hunters took 75% of the game by weight in 1988. Maya hunters were different from other subsistence hunters, however, in at least one regard. At X-Hazil Sur, 47 of 86 hunters took only one or two game animals during the study, while subsistence hunters described in other studies regularly undertook outings and obtained game. The reason for this difference probably is the number of domestic animals and opportunity to engage in wage labor available to Maya hunters. Given the availability of canned meat and domestic animals, and recognizing the low number of kills per hunter (ca. 7 per hunter during 17 months), subsistence hunting at X-Hazil Sur was quite opportunistic. 106 Hunter group size. The number of hunters participating in a hunting outing varies somewhat according to the type of hunt and the cultural background of the people, but hunter group size can be generalized across the Neotropics. At X-Hazil Sur, small-sized groups of hunters were common, while large groups were infrequent. About 85% of the 419 outings were conducted by groups containing one or two hunters, who divided the game among themselves. This followed the general pattern at X-Hazil Sur of adult men working alone or with a single partner and sharing part of the harvest or earnings. Large hunting groups at X-Hazil Sur usually were organized at the community level to obtain game for a specific religious occasion. On these outings, participants offered information about recent game sightings or tracks to others in the group, and the hunter making the kill generally donated the game to the feast organizers. These results indicated that hunter group size at X-Hazil Sur closely followed the pattern of few large groups and numerous small groups exhibited by other subsistence hunters in the Neotropics. The average number of game animals harvested per outing, at X-Hazil Sur, regardless of hunter group size, was relatively low (1.00 - 1.81 animals/outing). This indicated that the harvest of one or two game animals per outing was acceptable to hunters. Among the highly traditional Yekuana Indians in Venezuela (Sponsel, 1986), most hunting outings were conducted by one or two hunters. Group hunts were organized only when tracks of large- sized prey (e.g., deer and tapir) were located. Among the traditional Waorani Indians in Ecuador (Yost and Kelley, 1983), 63% of the outings were conducted by individuals, while groups of three or more hunters conducted only 10% of the outings. In northern Brazil (Saffirio and Scaglion, 1982), hunting group size varied between Yanomami Indian groups. For traditional Yanomami hunters, individual hunters conducted 78% of the outings, groups of two conducted 21% of the outings, and three or more hunters conducted only 1% of the outings. For acculturated Yanomami hunters, however, individual hunters conducted 9% of the outings and groups of two conducted 73% of the outings, while groups of three or more conducted 18% of the outings. Saffirio and Scaglion (1982) attributed these differences among the Yanomamo in part to the scarcity of game along the highway in 107 the area used by the acculturated hunters. In eastern Brazil (Smith, 1976), individual mestizo hunters conducted 76% of the outings, while two or more hunters conducted 24% of the outings among three conununities of settlers along the Transamazon Highway. These studies suggested that among settlers as well as traditional and acculturated indigenous peoples in the Neotropics, the vast majority of hunting outings were conducted by individual hunters or pairs of hunters. Types of weapons. The type of weapons used by hunters will affect the size, amount, and type of game harvested. Hunters at X-Hazil Sur used traps, rifles, and shotguns to harvest game. Traps were used for 9% by individuals of the game taken (mean prey weight = 0.6 kg). Rifles were used for 39% of the game (mean prey weight = 1.8 kg). Shotguns were used for 42% of the game (mean prey weight = 5.9-21.4 kg). In general, traps and rifles were used for smaller prey and shotguns were used for larger prey. However, 22 caliber rifles were used to kill prey up to the size of a brocket deer (mean weight = 15.6 kg), while 12 gauge shotguns were used to kill prey as small as the great curassow (mean weight = 3.1 kg). This indicated that Maya hunters did not follow a strict rule about using more powerful weapons for larger game and less powerful weapons for smaller game. When questioned about how they selected the weapon to use for a particular prey item, Maya hunters usually laughed and responded that they only had a single, very old gun from which to choose. Further, hunters often could not know what size prey they would encounter during a hunting outing and thus could not select one weapon over another. Thus, for most hunters at X-Hazil Sur, the only question was whether their weapon could kill the prey item at hand. The best explanation about how hunters select their weapons may deal less with preferences and more with what types of weapons are readily available. Hunters at X-Hazil Sur reported that due to Mexican legislation the only nearby location to purchase weapons or ammunition was at M^rida, about 250 km away. At Merida, State of Yucatan, licensed dealers offered a limited selection of items to buyers with the necessary permits. Hunters at X-Hazil Sur indicated that 22 caliber rifles and 16 and 20 gauge shotguns were best for hunting and most easily obtained, but it is also possible that sales of 12 gauge shotguns and larger caliber rifles were restricted by the Government of Mexico. Whatever 108 the reason, 22 caliber rifles and 16 and 20 gauge shotguns and the corresponding ammunition were more easily obtained than 12 gauge shotguns and larger caliber rifles. Among indigenous people in the Neotropics, shotguns, blowguns, and spears are widely used to harvest game (Hames, 1979; Hill and Hawkes, 1983; Yost and Kelley, 1983). While contemporary Maya Indians in the Mexican state of Chiapas mainly use rifles and shotguns (March M., 1987), archeological studies and descriptions of Maya hunters at the time of the conquest suggested the that pre-historic Maya used a broad array of weapons including dogs, blowguns, spears, atlatls, bows and arrows, snares, traps, torches, whistles, nets, and slingshots to attract and kill game (Flores, 1984; Hambhn, 1984, 1985; March M., 1987; Pohl, 1976). The type of weapon used will have a direct impact on the type of game taken by hunters. At X-Hazil Sur, whereas hunters generally did not use a specific type of we^on to take a specific type of game, shotguns generally were used to take the heaviest prey (e.g., deer and peccaries) and rifles and traps were used to take the hghtest prey (e.g., pocket gophers and thicket tinamous; Figure 3-10). These results differ somewhat from those presented in two studies of tribal peoples in South America. Hames (1979) compared the efficiencies of the shotgun and the bow in Neotropical forest hunting by the Ye'kwana and Yanomamo Indians along the upper Orinoco River in Venezuela. Hames (1979) noted several differences between the two types of weapons (for example, mean prey weight and number of individuals taken), but the most significant difference between the two types of weapons, according to Hames, is the larger number of arboreal and volant animals killed by the Ye'kwana (shotgun) hunters compared with the Yanomamo (bow) hunters. The arboreal and volant animals taken by the Ye'kwana included birds (especially taxa in the Cracidae Family), monkeys (especially taxa in the Cebidae Family), sloths (Bradypus tridactylus), and collared anteaters (Tamandua tetradactyla). In the second study, Yost and Kelley (1983) compared the efficiency of shotguns, blowguns, and spears in forest hunting by the Waorani Indians in eastern Ecuador. Yost and Kelley (1983) determined that 36% by weight of the prey was taken with blowguns, 51% with shotguns, and 13% with spears. Blowguns were used almost exclusively for arboreal animals under 10 kg, the spear was used for large 109 mammals, and the shotgun was used for both classes of animals. These two studies show that tribal hunters generally consider the type of prey at hand when selecting the type of weapon to be used. Maya hunters, however, appear to be less concerned about prey type than are tribal hunters when > selecting the type of weapon to be used. Characteristics of the Hunt Several techniques have been described for bumian hunters in the Neotropics. For example, the Ach^ Indians conducted day hunts through the forest, walking about 15 km on an average outing (Hill and Hawkes, 1983). Matses Indians in the Peruvian Amazon conducted day hunts, but also undertook extended trips lasting several days and conducted at great distances from their village (Romanoff, 1984). In Colombia, the Maraca Indians hunted at night; undertook day-long trips as well as extended hunting expeditions lasting several weeks; and constructed and used a number of special hunting aids, including blinds, stands, and deadfalls (Ruddle, 1970). Settlers and colonists in eastern Brazil used dogs and flashlights, pursued game during the day, waited in stands for game during the evening, and patiently waited along garden edges or near fruiting palms (Orbygnia martiana; Smith, 1976). The variety of these techniques indicated that game could be successfully taken by various methods. Hunting outings. Maya hunters at X-Hazil Sur used several techniques to obtain game. The most frequently used technique was to track/stalk game (79% of 419 outings). About 82% of the game, by individual, was taken by this method. Game also was taken by traps (10% of outings) and while hunters were positioned in stands/platforms (8% of outings). The average hunter group size for these three techniques was 1.6-1.8 hunters/outing, while the average quantity of game harvested was 1.2-1.4 individuals/outing. Hunters reported that they used the different techniques under different circumstances. Tracking/stalking was used primarily during the day and occurred in both gardens and forested areas. Usually an outing of this type was undertaken only after a hunter, based on the type of tracks or 110 feeding sign made by the animal, had ascertained over several days the species and general behavior of the prey. Game was tracked throughout the year, but was less frequently so during the height of the dry season, according to hunters, when noise from dry leaves and ground litter could alert the game. Stands and platforms were used primarily as sites from which to kill animals at night while they were feeding in gardens or on fallen fruit in the forest. Deer (during April-July) and pacas (during December-March) were especially vulnerable to this technique. Usually stands and platforms were constructed on an elevated location in Uie garden or at a location with a clear view of the game trail and the fallen fruit. Stands and platforms also were constructed at waterholes and at the entrances of caves used by pacas and agoutis. Traps were used primarily to capture pocket gophers, but pacas and agoutis also were taken. Normally, traps were set in gardens where the hunter was working or along trails enroute to the work site. Traps usually were checked twice a day while the hunter was in the area. According to hunters, traps were inefficient as they often misfired, the prey often escaped, other hunters frequently stole the game, or the hunter forgot the location of the trap. When compared with other subsistence hunters, the Maya should be considered as indigenous peasants and not as indigenous hunters. Maya hunters at X-Hazil Sur were quite similar to Brazilian settlers and colonists with respect to the techniques used in that both groups used dogs, tracked/stalked game, hunted by day and by night, and constructed stands/platforms. The Maya at X-Hazil Sur were unlike most of the more traditional indigenous people in the Neotropics, however, in that the Maya usually did not hunt in village groups, undertake extended expeditions, or even use blowguns or bows and arrows as we^ons. Kill site locations and vegetation types. The vegetation types of game kill sites can be invaluable in interpreting human hunting patterns as well as patterns of habitat use by game (Redford and Robinson, 1990). Hunters do not just randomly undertake a hunting. Rather, the timing and location of hunting outings takes into account game behavior and feeding patterns on fruiting trees or crops in gardens. Ill At X-Hazil Sur, the distribution of kill sites suggested that game species generally were taken in all major vegetation types. Rather than using a single vegetation type, the harvested game used highly-disturbed and early-successional vegetation types (e.g., Plots & Gardens and Early Secondary Forest) as well as little-disturbed and late-successional vegetation types (i.e., Late Secondary Forest; Table 3-7; Appendix G). Nine of the 12 game species used three of the vegetation types; Plots & Gardens, Early Secondary Forest, and Late Secondary Forest. Except for the white-lipped peccary (n = 3; 100.0% of locations in Late Secondary Forest), no species were taken exclusively in a single vegetation type. According to hunters, the reason for the wide variety of vegetation types in which game species were taken was due to the fact that game species used different vegetation types for different activities, for example, to rest, mate, forage, or seek drinking water. The significant relationship between game kill sites and areas categorized as Plots & Gardens, Early Secondary Forest, and Other was supported by results from two other studies (Table 3-10). At X-Hazil Sur, 63% of mammals and 70% of birds, by number of individuals, were taken in these three vegetation types although these types composed only 11.5% of the study area (see Table 2-1). In Brazil, Ka'apor Indian hunters at one site harvested 14.7% by number of the total game in areas categorized as gardens (0.3% of the total catchment area), while hunters at another site harvested 36.6% by number of the total game in gardens (1.7% of the total catchment area; Balee, 1985). Balee suggested that game used gardens as refuge and foraging areas. In Ecuador, Runa Indian hunters harvested about 45% by number of the pacas and agoutis taken in areas categorized as fallows and gardens (Irvine, 1987). Fallows and gardens also were important game kill sites for the collared peccary (14.3% by number), brocket deer (35.0%), and acouchi (Myoprocta pratti; 28.9%) taken by the Runa. Irvine suggested that gardens and fallows were important to pacas and agoutis due to the crops, while fruiting trees attracted game in the fallows. Minimum catchment area and mean kill site distances. An understanding of the catchment area over which hunters range is important in evaluating harvest yields and the status of game 112 populations in the area (Redford and Robinson, 1990). On a broader scale, this information can be used to analyze human territoriality and resource use at a community level (Vickers, 1983). The minimum catchment area for hunters at X-Hazil Sur was about 370 km", but hunters from this village as well as from Uh-May and Chancah Veracruz (not studied) ranged throughout the ejido (552.95 km-). This catchment area of 370 km^ may be somewhat artificial for two reasons: One, the eastern boundary of the ejido roughly extends to the savanna on Sian Ka'an Biosphere Reserve. The savanna is flooded much of the year, due to rainfall, and hunting is difficult (see Figure 2-4). Two, the ejido boundary is clearly marked. Ejido residents avoid legal problems by not entering the adjacent ejido to hunt, plant gardens, or harvest timber. Catchment area values reported for mestizos and indigenous groups indicated that while some communities did kill some game over relatively large areas during the time of the studies, most use much smaller areas (Table 3-11). For example, among the largest catchment areas reported for an indigenous group were a region of about 1810 km^ used by a community of 500 Mats^s Indians in western Brazil (Romanoff, 1976) and a zone of about 600 km^ used by a group of 155 Achd Indians in eastern Paraguay (Hill and Hawkes, 1983). The reported catchment areas for eight other indigenous groups ranged from 79 to ca. 400 kml Among mestizos, the reported catchment area was 100-500 km^ per community. While these areas and hunter groups differed, for example, with respect to vegetation type, use of weapons, wildlife populations, degree of acculturation, human population density, and importance of fishing and shifting cultivation, the catchment areas-excluding the extremely large and small areas-were remarkably similar in size at 200-600 km^. The relationship between catchment area and specific harvest areas around communities of indigenous and mestizo subsistence hunters are poorly known. At X-Hazil Sur, harvest areas in which individual species were taken ranged from 4.8 km^ to 251.8 km^ and were substantially less than the total area available on the ejido (Table 3-8). Only two species were taken over relatively large areas; coati (251.8 km") and collared peccary (211.5 km^). Both coatis and collared peccaries are known to be highly mobile and occur in groups, but their home range sizes in Mesoamerica are poorly known. 113 Mean kill site distances varied between species and ranged from 3.4 km (pocket gopher) to 7.6 km (ocellated turkey; Table 3-8). According to hunters at X-Hazil Sur, the ocellated turkey and great curassow (7.5 km) were quite wary and easily disturbed by people. These observations agreed with those from other sites where researchers reported that the ocellated turkey and great curassow were highly sensitive to human disturbances. Two possible explanations may account for the limited harvest areas in which individual game species were taken: One, individual species did not occur except in those areas where they were harvested. Two, individual game species occurred beyond the catchment area, but were not harvested by hunters in those locations. Comments offered by hunters and wildlife census data presented in Chapter 4 supported the second possible explanation. According to several hunters, it was easier to locate and hunt game nearby, in the vicinity of a garden or at a work site, than it was to track or stalk game in distant areas that were poorly known by the hunter. A consequence of this belief was that hunters may have harvested less game. Kill site distances also suggested that many species were highly tolerant of human disturbances. Of the 11 species for which a range could be calculated, nine game species had a minimum kill site distance < 1.4 km (Table 3-8). At this distance from the village, children frequently played, human voices and barking dogs often could be heard, street lights were easily seen at night, and dogs and men frequently passed enroute to the gardens (see Chapter 6). Apparendy these kinds of disturbances were insufficient to cause most game species to avoid the area around the village. The minimum catchment area and mean kill site distances for individual subsistence hunters in the Nootropics, as compared with areas used by the village as a whole, are poorly known. For Maya hunters at X-Hazil Sur, the minimum catchment areas for the seven main hunters ranged from 7.2 km^ (# 62) to 68.1 km^ (# 9; Table 3-9; Appendix E). Hunter # 62 (23 y old) had the smallest catchment area and reported that he only hunted in the vicinity of his garden, which was near X-Hazil Sur, and at nearby chicle tapping sites. Hunter # 9 (21 y old) had the largest catchment area. Although this 114 B 3 in 00 n CO s o s s -2? o u a a 1 1> >N u c 1) 3 O" OJ tM -v ^ r~ C3 00 8 ON Wn -J a. 2i s d 1 cA vT 1» i 1 1 H ^ 00 ^ -1^ 2 S ^ 5 te c/: £ o. u < fN , 30 >< SV ^ '^ 3 >^ a Sg 3 u fN £ i) ^^ o. CI so ng o CI CI 00 C i ON 1 ■S S cs s 'J 5J u J= ^ t^ > s > 3, z W5 W5 TJ OJ ^ B a 0.05) were detected for sighting distances between forest successional stages by species (four species tested), taxonomic group (five groups tested), or game versus nongame status (four game status combinations tested; Appendix J). Number of Taxa and Sightings Sightings. A total of 23 taxa were sighted (16 mammals and 7 birds; Appendix J; Figure 4-2) during the 121 censuses (total distance censused = 240.460 km; total time censused = 244.27 h; Appendix H). Transects were censused five times each during the sunrise period and five or six times each during the sunset period. Censuses on Transect # 1 were discontinued after six censuses due to problems with a local resident. More than 50% of the taxa had been sighted by census number 22, while no new taxa were sighted after census number 75. A total of 240 sightings were recorded (150 mammals and 90 birds; Appendix J). Among the mammals, the most frequently sighted taxa were squirrels (47 sightings; Sciurus deppei and S. 133 2S 20 15 lO NUMBBR TAXA SIGHTED 20 40 60 80 lOO CENSUS NUMBER TAXA -H— MAMMALS O BIRDS — ^ 120 140 — ^— TOTAL Figure 4-2. Cumulative number of avian (7) and mammalian (16) taxa sighted during 121 censuses in 1990. yucatanensis, combined; see Appendix L for scientific and common names), kinkajous (31 sightings), and coatis (17 sightings). The plain chachalaca was the most frequently sighted bird (64 sightings). Other birds rarely were sighted. Animals not sighted or not present. Several taxa of mammals and birds, that potentially could have been taken for subsistence purposes, were not observed during the censuses. Some of these taxa were observed at other times during the study (see Chapter 3 for an inventory of game species) or their presence was reported by local residents, while other taxa were expected to occur in the area, based on 134 published range m^s (e.g., Leopold [1977] and Peterson and Chalif [1973]), but were not observed or reported. Among mammals, several taxa were not observed during censuses. Two primates were reported by hunters, but not observed during censuses: howler monkey (Alouatta pigrd) and spider monkey {Ateles geoffroyi). The rabbit iSylvilagus floridanus) and the ring-tailed cat (Bassariscus sumichrasti) were not reported. The porcupine (Coendu mexicanus), short-tailed weasel (Mustela frenata) and Baird's tapir (Tapirus bairdii) were reported (pers. obs.). Among birds, three taxa were expected, but not observed during censuses; the great tinamou (Tinamus major) was not reported by local residents, while the crested guan (Penelope purpurascens) and the black-throated (Yucatan) bobwhite (Colinus nigrogularis) were reported, but not observed. Relative Abundance Since the number of individuals sighted during the censuses was relatively low, three levels of analyses were conducted: species (four species compared), taxonomic group (five groups compared), and game versus nongame birds and mammals (four categories compared). These comparisons allowed conclusions based on the available data. Species. Significant differences were detected for plain chachalaca sighting frequencies between Late Secondary Forest without Gardens, Late Secondary Forest with Gardens, and Combined/Early Secondary Forest (mean sighting frequencies were 1.9 individuals/ 10,000 km, 0.6/10,000 km, and 5.2/10,000 km, respectively; x^ approximauon = 8.4900, d.f. = 2, P = 0.0143; Table 4-1). No significant differences in sighting frequencies were detected between forest successional stages for squirrels (P = 0.1596), coatis (P = 0.0877), or kinkajous (P = 0.0665). Groups. Significant differences were detected for bird sighting frequencies between Late Secondary Forest without Gardens, Late Secondary Forest with Gardens, and Combined/Early Secondary Forest (mean sighting frequencies were 5.2 individuals/ 10,000 km, 1.3/10,000 km, and 6.8/10,000 km, respectively; x' approximation = 7.2692, d.f. = 2, P = 0.0264; Table 4-1). 135 Significant differences also were detected for carnivore sighting frequencies between Late Secondary Forest without Gardens, Late Secondary Forest with Gardens, and Combined/Early Secondary Forest (mean sighting frequencies were 10.2 individuals/10,000 km, 3.4/10,000 km, and 0.9/10,000 km, respectively; x^ approximation = 6.2482, d.f. = 2, P = 0.0440; Table 4-1). No significant differences in sighting frequencies were detected between forest successional stages for marsupials and edentates (P = 0.9100), rodents (P = 0.0954), or artiodactyls (P = 0.4337). Game versus nongame taxa. Significant differences were detected for game bird sighting frequencies between Late Secondary Forest without Gardens, Late Secondary Forest with Gardens, and Combined/Early Secondary Forest (mean sighting frequencies were 4.8 individuals/ 10,000 km, 1.2/10,000 km, and 5.8/10,000 km, respectively; x^ approximation = 7.2947, d.f. = 2, P = 0.0261; Table 4-1). No significant differences in sighting frequencies were detected between forest successional stages for game mammals (P = 0.1972), nongame mammals (P = 0.1752), or nongame birds (P = 0.1994). Population Density Estimates The number of wildlife sightings were sufficient to calculate several population density estimates, but not sufficient to justify a statistical analysis of population density estimates among the three successional stages of forest. In the following section, as in the previous section, the data for population density estimates were summarized by species, taxonomic group, and game and nongame birds and mammals. Species. Among mammals, the squirrel was the most abundant species and had a higher population density in Late Secondary Forest without Gardens (10.3 individuals/km^) than in Late Secondary Forest with Gardens (7.2 individuals/km^) or Combined/Early Secondary Forest (4.5 individuals/km-; Table 4-2). The plain chachalaca was the most abundant bird and had a higher population density in Combined/Early Secondary Forest (20.3 individuals/km^) than in the other successional stages. 136 Table 4-1. Average sighting frequency (mean number of sightings per 10,000 km) of wildlife, summarized by species, groups, and game and nongame birds and mammals censused along 12 transects in three successional stages of forest at Ejido X-Hazil y Anexos, Quintana Roo, Mexico, during 1990 (d.f. = 2 throughout). Successional stages x' approx. Late Secondary Forest without Gardens Late Secondary Forest with Gardens Combined/ Early Secondary Forest Taxa' — b X SE X SE X SE P a) Species Squirrels (47) 2.7 (0.4) 1.8 (0.3) 1.4 (0.4) 3.6699 0.1596 Coati (17)*' 0.3 (0.1) 1.9 (1.4) 0.0 (0.0) 4.8673 0.0877 Kinkajou (31) 2.0 (0.5) 1.4 (0.5) 0.4 (0.2) 5.4220 0.0665 Plain chachalaca (64)* 1.9 (0.7) 0.6 (0.6) 5.2 (0.2) 8.4900 0.0143 b) Groups Marsupials & Edentates (18) 1.6 (1.1) 0.9 (0.4) 0.7 (0.3) 0.1886 0.9100 Rodents (68) 10.3 (6.6) 3.0 (0.6) 1.8 (0.5) 4.6991 0.0954 Carnivores (55) 10.2 (8.3) 3.4 (1.7) 0.9 (0.2) 6.2482 0.0440 Artiodactyls (9) 0.3 (0.3) 0.3 (0.1) 0.6 (0.2) 1.6545 0.4337 Birds (90) 5.2 (1.7) 1.3 (0.8) 6.8 (0.5) 7.2692 0.0264 c) Game versus nongame species' Game mammals (47)* 1.4 (0.8) 3.3 (1.8) 1.0 (0.3) 3.2473 0.1972 Nongame mammals (103) 21.0 (16.3) 4.3 (0.9) 3.0 (0.4) 3.4833 0.1752 Game birds (78)* 4.8 (1.8) 1.2 (0.7) 5.8 (0.2) 7.2947 0.0261 Nongame birds (12) 0.4 (0.2) 0.1 (0.1) 1.0 (0.4) 3.2249 0.1994 See Appendix L for common and scientific names, (n) = number of sightings. X = (10,000) * (average of four replicates of sightings/km). * = Game species. See Chapter 3 for additional information about game species. 137 Groups. Birds were more abundant in Combined/Early Secondary Forest than in Late Secondary Forest without Gardens or Late Secondary Forest with Gardens. Among the mammalian groups, rodents were most abundant and had a higher density in Late Secondary Forest without Gardens (13.5 individuals/km^) than in Late Secondary Forest with Gardens or Combined/Early Secondary Forest. Birds had a higher population density in Combined/Early Secondary Forest (26.3 individuals/km^) than in Late Secondary Forest without Gardens or Late Secondary Forest with Gardens. Game versus nongame taxa. Game mammals and birds were different from each other in that game mammals were most abundant in Late Secondary Forest with Gardens (8.4 individuals/km^), while game birds were most abundant in Combined/Early Secondary Forest (21.8 individuals/km-; Table 4-2). Nongame birds and mammals were similar in that their respective population densities were comparable in Late Secondary Forest without Gardens and Late Secondary Forest with Gardens. Discussion Number of Taxa and Sightings A total of 16 taxa and 150 individuals of mammals and 7 taxa and 90 individuals of birds were sighted during 121 censuses at Ejido X-Hazil y Anexos (total distance censuses = 240.460 km; total time censused = 244.27 h; Appendices H and J). These sightings did not include all of the taxa in the study area, however. About 25 taxa of non volant manmials, including game and nongame species, were killed by hunters (see Appendix F) or sighted during censuses, while another 4 taxa were sighted incidental to other activities during the study. Thus, excluding small rodents, there were about 29 taxa of nonvolant mammals at the study area. Compared with the number of nonvolant mammals sighted at other Neotropical sites, the number of taxa sighted at Ejido X-Hazil y Anexos was low. For example, Glanz (1990) reported 39 species for Barro Colorado Island, Panama. In Costa Rica, Wilson (1983) reported 43 species for La 138 Selva and 59 species for Osa Peninsula. In Peru, Patton et al. (1982) and Terborgh et al. (1984) reported 59 species for Rio Cenepa and 68 species for Cocha Cashu. While these sites differ in many Table 4-2. Population density estimates (individuals/km^ total number of sightings [n]) determined by King's Method and sununarized by species, taxonomic groups, and game and nongame birds and mammals. Successional Stages Taxa' Late Secondary Forest without Gardens Late Secondary Forest with Gardens Combined/Early Secondary Forest a) Species Squirrels'" 10.3 (22)' 7.2(13) 4.5 (12) Coati*" 1.0 (2) 3.9 (15) - (0) Kinkajou 5.6(16) 5.5 (12) 1.6 (3) Plain chachalaca* 8.7 (16) 2.4 (5) 20.3 (43) b) Groups Marsupials & Edentates 2.8 (4) 4.9 (8) 3.4 (6) Rodents 13.5 (29) 10.8 (24) 5.7(15) Carnivores 6.8 (19) 9.5 (29) 2.6 (7) Artiodactyls 0.4 (2) 0.8 (2) 1.7 (5) Birds 10.4 (24) 5.0(10) 26.3 (56) c) Game versus nongame species'* Game manmials* 4.2(11) 8.4 (28) 3.0 (8) Nongame mammals 18.2 (43) 18.4 (35) 10.1 (25) Game birds* 11.7(21) 4.7 (9) 21.8(48) Nongame birds 0.5 (3) 0.4 (1) 4.6 (8) See Appendix L for common and scientific names. ^ Composed of Sciurus deppei, S. yucatanensis, and an unidentified squirrel (not a new taxon). ' (n) = Number of sightings. '' * = Game species. See Chapter 3 for additional information about game species. 139 aspects, they are similar in being forested Neotropical areas that share comparable faunas. The most frequently sighted taxa at Ejido X-Hazil y Anexos were the plain chachalaca (n = 64 sightings), squirrels (n = 47), kinkajou (n = 31), and coati (n = 17). Of these three, however, only the coati and plain chachalaca were considered as game species by Maya hunters (see Chapter 3 for a description of game species). At least three environmental variables may be responsible for the abundance of wildlife at Neotropical forest sites (Emmons, 1984): soil type, undergrowth density, and rainfall and seasonality. Neither undergrowth density nor rainfall and seasonality were analyzed with respect to wildlife densities at Ejido X-Hazil y Anexos. Soil properties, however, were studied by Noguez-Galvez (1991) and can help explain the low number of wildlife taxa in the study area as soils help determine the type of vegetation that an area potentially can support. The soils at Ejido X-Hazil y Anexos, according to the FAO classification system, correspond to the Order Rendzinas, Suborder Calcomorphics, while according to the USDA classification system, these soils correspond to the Order Mollisols, Suborder RendoUs (United States Department of Agriculture, 1960). At Ejido X-Hazil y Anexos, these soils have good structure; drainage; and a high content of organic matter, calcium, carbonates, and bicarbonates (Noguez-Galvez, 1991). As a result of a shortened fallow period, from 20 y previously to 10 y at present, however, soils at X-Hazil now are nutrient-poor. Noguez-Galvez (1991) predicted that serious degradation of the soils and cover vegetation will occur within a short period of time if tbis frequency of forest clearing continues. While the length and severity of this nutrient deficiency problem are unknown, it is quite possible that the soils and cover vegetation at the study area consequently are unable to support wildlife population densities found elsewhere in the Neotropics. Soil fertility has been used to explain differences in animal biomass levels in Neotropical forests in South America. Janzen (1974) proposed that animal biomass should be less on nutrient-poor soils that produce black- water rivers in tropical forests than on soils that produce white- water rivers. The mechanism predicted for this difference was the high level of toxic secondary chemicals in plant 140 tissues. Eisenberg (1979) and Emmons (1984), based on an analysis of variations in mammal densities among various Neotropical sites, supported this hypothesis that areas with poor soils have fewer individuals than areas with rich soils, but did not necessarily agree with the mechanism proposed by Janzen (1974). Historical factors can also affect animal species abundance and diversity. Due to its location on the land corridor that joins North and South America, the flora and fauna of the Yucatan Peninsula strongly reflect the mingling of species that resulted after the closing of the Panamanian land bridge about three million years ago ((jraham, 1993; Haffer, 1974; Stelhi and Webb, 1985). The impact of the land bridge was especially notable on terrestrial mammals as nearly half of the families and genera that now occur in South America belong to groups that emigrated from North America during the last three million years (Marshall, 1988). A substantial number of South American mammalian taxa likewise migrated to North America, but were less successful in establishing themselves than their North American counterparts (Webb and Marshall, 1982). These movements continue today as the distribution of the North American coyote (Canis latrans) approaches Colombia and the distribution of the South American armadillo (Dasypus novemcinctus) moves northward into southern portions of the United States (Hall, 1981). A second historical factor affecting the distribution and abundance of plants and animals is the arrival of people to the New World. Crossing the Bering land bridge from Asia to North America about 11,000-12,000 y ago, humans occupied most or all of the New World within possibly 1,000 y (Martin, 1973). The human impact on the fauna was immediate and widespread as most of the large- bodied mammals in North and South America went extinct (Marshall et al., 1984). Although Markgraf (1985) proposed that climate change was responsible for these extinctions, Marshall (1988) attributes the loss of these species to human hunters. The third historical factor important here is the development of the Maya civilization. While there is some debate about when people first arrived in Mesoamerica (cf., Adams, 1991; Martin, 1973; Nesbitt, 1980; Pipemo et al., 1990), a tangible Maya civilization dates from about 1,500 B.C. 141 (Andrews, 1960, 1965; Deevey et al., 1979). At the height of the Maya civilization, roads, villages, large ceremonial centers, and extensive agricultural fields supplied with water by canals occurred throughout the Yucatan Peninsula to Guatemala and Honduras (Hammond, 1982a, 1982b, 1986; Turner, 1974, 1990; Turner and Harrison, 1983). The impact of these activities on plants and animals in Mesoamerica can only be surmised. Likewise, the impact of the fall of the Maya civilization about 900 A.D. on plants and animals and the resulting decrease in human activities has yet to be determined. Change persists, however, as humans now exploit lumber, chicle, and other nontimber forest products in Quintana Roo and the southern portion of the peninsula (Edwards, 1986; Konrad, 1988). Two additional factors affecting animal populations in other areas may also be affecting animal populations at Ejido X-Hazil y Anexos; human hunters and large predators. In a study of tapir (Japirus bairdii) abundance in Belize, Fragoso (1991, 1992) determined that among such factors as logging, disease, habitat destruction, and shifting cultivation, the most parsimonious explanation for the low number of tapirs was human hunting. Similarly, Glanz (1991) compared mammal abundances at protected and unprotected areas in central Panama and compared them with Barro Colorado Island, which is a protected area, but also lacks large predators such as jaguars (Panthera onca) and pumas {Felis concolor). Glanz (1991) noted dramatic differences in mammal densities among sites and attributed these differences to the impacts of hunting by humans rather than due to the absence of large predators, as had been noted by Emmons (1987) at other Neotropical forested sites. Forest Successional Stages and Relative Abundance of Birds and Mammals The study area had three forest successional stages that differed with respect to degree of human disturbance, the abundance of gardens, and the number and kinds of crops planted in the gardens. Birds and mammals were expected to select for or against the different stages according to their individual requirements. Few clear patterns, however, were noted. At the species level, a significant difference in sighting frequency between forest successional stages was observed only for the plain chachalaca (P = 0.0143; Table 4-1). No significant differences 142 were observed for the other three species that were compared. The plain chachalaca, a game species, was most frequently sighted in Combined/Early Secondary Forest (J = 5.2 sightings/ 10,000 km) and least frequently sighted in Late Secondary Forest with Gardens (J = 0.6 sightings/10,000 km; Table 4- 1). This agreed with hunter reports that the plain chachalaca foraged, displayed, and nested in the vicinity of gardens and roads, but rarely ventured into Late Secondary Forest. When game and nongame birds and mammals were compared, few clear patterns likewise were noted. Among the four animal categories, significant differences in sighting frequencies between successional stages were observed only for game birds (P = 0.0261; Table 4-1). The category game birds, however, was composed primarily of plain chachalaca sightings (64 of 78 sightings). These findings suggested that at Ejido X-Hazil y Anexos the frequency with which birds and mammals were sighted did not vary greatly between the three successional stages. This was unexpected given the distribution of game kill sites in the four main vegetation types (see Appendix G), especially in the vegetation type categorized as Plots & Gardens, and the comments of hunters that animals were more noticeably abundant in some areas and less abundant in others. One possible explanation for an apparent lack of clear patterns of habitat use by wildlife of different forest successional stages was the small sample size of sightings. Only 240 animals were sighted; about 1 individual/km or 1 individual/h. A larger sample size likely would identify significant differences in use of different successional forest stages by wildlife. Another possible explanation for an apparent lack of clear habitat use patterns by wildlife was the mobility of the animals, location of the 12 census transects (see Figure 4-1), and high degree of interspersion of the various vegetation types (see Figure 2-4). Although four transects were located in Late Secondary Forest without Gardens, they may have been located too close to other vegetation types to differentiate clearly any habitat use differences by wildlife. While it may have been desirable to have located the Late Secondary Forest without Gardens transects further to the east of Route 307, the distribution of roads and trails in the forest and the transition from medium- to low-height forest and savannah may have introduced other complications. 143 Rank Order of Take Versus Sighting Frequency A comparison of game harvest at Ejido X-Hazil y Anexos with the frequency at which game animals were sighted during censuses suggests that the more-frequently-taken game animals also were the more-frequently-sighted game animals. The coati and plain chachalaca were the most frequently taken game species, at 167 kills each (Table 4-3). The plain chachalaca was the most frequently sighted game species (64 sightings), while the coati ranked second (17 sightings; Table 4-3). Conversely, the white-lipped peccary, great curassow, and ocellated turkey were rarely taken or sighted. While not analyzed statistically, these comparisons indicate that Maya hunters at Ejido X- Hazil y Anexos either concentrated their efforts more on the common game species and less on the rare species or, perhaps more simply, harvested the species that were more frequently encountered. Comparative Population Density Estimates Manunal population densities have been calculated for several sites in Neotropical forests and permit a comparison with some taxa that occur at Ejido X-Hazil y Anexos. At Ejido X-Hazil y Anexos, when both game and nongame mammals are considered, the squirrel (two taxa combined) was the most abundant taxon at 4.5-10.3 inds./km^, followed by the kinkajou (1.6-5.6 inds./km*) and the coati (1.0-3.9 inds./km^; Table 4-4). Squirrels also had the greatest population densities at two other sites: Panama (180 inds./km^ Glanz [1982]) and Sierra Chame, Guatemala (100 inds./km^; Hendrichs [1977]), while they were relatively uncommon at Guatopo, Venezuela (Eisenberg et al., 1979), Cocha Cashu, Peru (Emmons, 1987; Terborgh, 1983), Cabassou, French Guiana (Charles-Dominique et al., 1981), and Tikal, Guatemala (Cant, 1977). On the other hand, excluding bats and small rodents, opossums were among the most abundant mammals at three of the six Neotropical sites (25-65 inds./km-; Venezuela, Peru, and French Guiana), while they were relatively uncommon at Ejido X- Hazil y Anexos. When the population density estimates are compared, it is clear that mammal populations at Ejido X-Hazil y Anexos were much less dense than those at the other Neotropical sites mentioned above. For example, squirrel densities at Panama (Glanz, 1982) were about 17 times greater than 144 Table 4-3. Reported number of individuals taken by Maya hunters at X-Hazil Sur versus number of sightings during animal censuses (game species only). Total number of Rank order Total number of Rank order Taxa individuals taken (take) sightings (sightings) a) Mammals Pocket gopher 53 3 1 10 Paca 47 4 8 5 Agouti 35 6 12 3 Coati 167 1 17 2 White-lipped 3 12 0 11 peccary Collared peccary 40 5 2 8 Brocket deer 16 8 4 6 White-tailed deer 24 7 3 7 Game mammals 385 47 b) Birds Thicket tinamou 13 9 12 3 Great curassow 13 9 0 11 Plain chachalaca 167 1 64 1 Ocellated turkey 6 11 2 8 Game birds 199 78 Total (mammals 584 125 -1- birds) those at the study area (Table 4-4). For the coati and kinkajou, population densities at the other sites were about five times as great at the other sites as at Ejido X-Hazil y Anexos, while it is not possible to generalize numerically about the population densities of the remaining taxa of mammals, it is clear that the densities at Ejido X-Hazil y Anexos were much lower. 145 Table 4-4. Estimated population densities of mammals at Ejido X-Hazii y Anexos compared with those at six other study areas in Neotropical forests (summarized from Glanz [1990]). Study area . (no. inds./km^) Sierra Cocha Cabassou, Chame, Tikal, This BCl, Guatopo, Cashu, French Guate- Guate- Taxa^ study' Panama' Venezuela'^ Peru' Guiana^ mala' mala" Didelphis Present 47 65 55 25-50 Tamaiidua Present 5 6 f 7 Dasypus Present 53 4 7 7 Sciurus 4.5-10.3 180 25 25 Common 100 > 9 Orthogeomys* Present Agouti* Present 40 18 3.5 Common 30 8 Dasyprocta* Present 100 63 5.2 Common 30 8 Urocyon Present Nasua* 1.0-3.9 24 i < 1 Present 15 20 Potos 1.6-5.6 20 — 20 20-30 20 74 Eira Present 1.6 2 ? Present 1 2 Conepatus Present Tayassu* Present 9.3 0.19 5.6 10 Mazama* Present 2 5.3 2.6 20^ Odocoileus* Present 0.7 — — 20^ Includes only those mammals observed at Ejido X-Hazil y Anexos. * = game species at Ejido X-Hazil y Anexos. Range of values presented for three vegetation types. Barro Colorado Island; Glanz (1982). Eisenberg et al. (1979). Emmons (1987) and Terborgh (1983). Charles-Dominique et al. (1981). Hendrichs (1977). Cant (1977). ? = Unknown. — = Does not occur. Mazama and Odocoileus combined. A final consideration involves the practice of keeping domestic animals. Maya Indians in the Yucatan Peninsula have had access to pigs, chickens, horses, cattle, and other kinds of livestock since 146 the early 1500s, following the Spanish conquest (Redfield and Villa Rojas, 1962; Villa Rojas, 1987). In order to protect these animals, the Maya likely killed the large predators, venomous snakes, and other noxious animals in the vicinity of their villages. The accumulated effects of these attempts to eradicate certain kinds of dangerous wild animals in the area over 450 y are unknown, but likely have affected the population densities of many game birds and mammals. In conclusion, in this assessment of wildlife populations at Ejido X-Hazil y Anexos, I have shown that densities are very low, compared with other Neotropical forested sites. This was true for birds as well as mammals, especially game species. While the reasons for these low densities are unknown, it is possible that low soil nutrient levels due to increasingly shorter fallow periods have negatively affected the vegetation of the area. It is also quite possible that the occurrence of jaguars and pumas in the area (Navarro L. et al., 1990) and the occurrence of Maya hunters during 2,000- 3,000 y are also responsible in part for low animal abundance levels. Most likely low animal densities are a result of the combination of many of these factors. With respect to garden hunting, the results obtained during this study generally did not detect differences in population densities of game species between areas with and without gardens. This could mean either that differences did not occur or that differences did not exist. Local hunters reported that wildlife populations were denser in areas with gardens than in areas without gardens, but could not quantify the magnitude of these differences. Density differences were indirectly confirmed by the relatively large proportion of game taken in areas categorized as Plots & Gardens (see Chapter 3). However, given the small size of the study area, close proximity of census transects, and large home ranges of many game species, it may not have been possible to detect differences in population densities of game species. CHAPTER 5 CONSUMPTION OF GARDEN CROPS BY GAME ANIMALS IN QUINT ANA ROO, MEXICO Introduction As a game-procurement system, garden hunting has been described as more similar to harvesting vegetable products and marine resources than to hunting in tropical forests (Linares, 1976). Basing her conclusions on an archeological study of a group of indigenous people in northwestern Panama who practiced shifting cultivation, Linares (1976) described the garden hunting system and determined that it focused on specific taxa of terrestrial mammals. One element of this system, it is hypothesized, is that many of the mammals taken are dependent upon crops found in gardens. A second element of this hunting system, it is hypothesized, is that those mammals, as a consequence of consuming crops, are found at greater population densities in forested areas with gardens than in forested areas without gardens. In response to the higher densities of these mammals around gardens, hunters have shifted their methods for pursuing wild animals from tropical forest hunting, where a wide variety of arboreal and terrestrial birds and manmials are harvested, to garden hunting, where a narrow range of terrestrial mammals are taken. Although Linares (1976) did not present corroboration for either hypothesis, the model of garden hunting has been applied to game harvest practices by hunters in other areas. It has been suggested, for example, that Maya hunters in Mexico practice garden hunting (Greenberg, 1992; Nations and Nigh, 1980). Given that many of the species taken by the Maya also were reported for Panama (see Chapter 3), it may be possible to use the example of subsistence hunting by Maya Indians in Mexico to support the model of garden hunting. The data about food habits of game animals must be considered within the context of the overall study. In Chapter 3, data were presented about the distribution of game kills in the various 147 148 vegetation types that occur on the study area. Given that many animals were taken while they were feeding, their diet and the vegetation type of the kill site are related. In Chapter 4, data were presented about the differences in wildlife densities in three forest successional stages: Late Secondary Forest without Gardens (areas with little habitat or human disturbance), Late Secondary Forest with Gardens (moderately disturbed areas), and Combined/Early Secondary Forest (highly disturbed areas). In Ch^ter 6, data will be presented about the characteristics of Maya gardens and their use by game. In this chapter, data about the nature, extent, and seasonality of consumption of garden crops by wild animals at Ejido X-Hazil y Anexos will be presented. There were four general objectives for this study and several hypotheses: First objective. --Quantify and characterize food items consumed by game species, identifying the most frequently taken food items, plant parts consumed, and proportion of plant and animal matter. Second objective. -Compare the percent occurrence of major crop species in the diets of game species. The hypothesis was that there were no differences in diets, by game species, with respect to either the number of crop taxa consumed or the percent occurrence of crops in stomach samples. Third objective. -Compare seasonal variation in consumption patterns of major CTop species by selected game species. Two hypotheses were examined: One, for each major crop species, there was no difference, by garden season for all game species combined, in the percent occurrence or the percent volume by plant part of the individual major crop species in the diet. Two, for each selected game species, there was no difference, by garden season for each individual crop species, in the percent volume or percent occurrence of the individual major crop species in the diet. 149 Methods Study Area The study took place at Ejido X-Hazil y Anexos, Quintana Roo, Mexico, during 1989-1990 (total area = 552.95 km^; Figure 2-1). Crop consumption data were obtained from game taken by residents at the village of X-Hazil Sur (19°23'30"N, 88°05'00"W; population = 1,040), the largest of three villages on the ejido (total population = 1,680). The mean annual temperature is about 26°C and the area typically has one dry season (December-May) and one wet season (June-November). Rainfall during 1 January-12 December 1990 was 1,277.3 mm (Figure 2-4). About 88.52% of the ejido was categorized as Late Secondary Forest, 6.07% as Plots & Gardens, 5.18% as Early Secondary Forest, and 0.23% as Other (Table 2-3). Since about 1915 the area has been occupied by Maya Indians, whose main subsistence activity has been shifting cultivation, primarily com. Prior to 1915, Maya Indians did not occur in the area. Collection and Measurement of Stomach Contents The results of this chapter were based on stomach contents obtained from freshly killed game taken by hunters for subsistence purposes at X-Hazil Sur during June 1989-October 1990. Each reported game kill was measured and weighed, stomach contents were collected, and the hunter was interviewed regarding the particulars of the hunt (see Chapter 3). For mammals, the term stomach contents is taken to mean the contents of the stomach proper, while for birds this term includes the contents of the esophagus, crop, proventriculus, and gizzard. For each stomach sample, including parasites and digestive juices, the total wet volume of the material was measured using a beaker, small syringe, graduated cylinder, or calibrated plastic bucket. For small quantities, ^ about 100 ml, the volume was measured to within 1 ml. For large quantities, ^ about 500 ml, the volume was measured to within 50-100 ml, depending upon the total volume of the specimen. After the total volume of the stomach sample was determined, the contents were washed with water in a food strainer (mesh size about 1 mm) to remove the digestive juices and fine food 150 particles. For small volumes, the entire contents (minus juices and fine particles) were preserved, while for large volumes, about 25-50% of the contents (up to about 500 ml, minus juices and fine particles) was preserved. Stomach contents were preserved in 15% formalin solution. Identification of Stomach Contents Stomach contents were identified on-site at X-Hazil Sur. These identifications were based on a visual examination of the washed stomach contents in which the food items were separated by size, shape, color, and plant or animal part. Food items were categorized as fruits or seeds (abbreviated as PS in tables and figures), leaves (L), animal material (AM; e.g., hair, snail shells, or insect antennae), and other (O; e.g., tubers, roots, stems, plastic, rocks, or soil; Appendix M). Two local Maya hunters, each about 45 y old and with extensive hunting, logging, and gardening experience in the area, identified the stomach contents. These men were provided with the local common name of the game animal, the vegetation type of the kill site, and the time of year when the animal was killed. Supported by this information, these individuals studied the sample, identified the food items, and indicated the local common name in Maya for the material. The men were not pressured to make determinations when they were not sure of the identity of the food item. In some cases, the men consulted with other village residents to identify an unknown food item. By using the same two individuals to identify all of the stomach content samples and by allowing these men to consult with others, the consistency of the identifications was enhanced. The identification of stomach contents and subsequent measurement by species and food part were based on a complete examination of some specimens and on a partial examination of other specimens. For stomach contents of relatively small volume, the entire contents were examined, while for large volumes, a random subsample of about 50-100 ml was examined. After each food item had been identified, its volimie was measured to within 0.5 ml, and the specific plant or animal part that had been consumed was recorded. Food items that measured < 0.5 ml (e.g., body parts of insects or 151 snails) were categorized as "trace" and considered to have a volume of 0.0001 ml for analytical purposes. Trace items were reported in the tables and appendices as " < 0.0%." The scientific name of each food item was determined from a reference collection of local plants and animals. Most of these specimens were initially identified by hunters with the local common name in Maya. About 265 specimens of animals and plants (most with flowers or fruits) were collected and identified. Reference plant specimens were identified by J. Chavelas Polito, Director, Centro Experimental Forestal "San Felipe Bacalar," and E. Cabrera Cano, Botanist, Centre de Investigaciones de Quintana Roo (CIQRO). Reference animal specimens were identified by various specialists from CIQRO and Universidad Nacional Aut6noma de Mexico (see Acknowledgments). Plants usually were identified to the species level. Insects were identified to the level of order, while other animals usually were identified only to the level of class due to the damage they had incurred while being eaten. The reference collection and the stomach contents, including additional specimens not identified during this study due to time constraints, were deposited in the scientific collections at CIQRO, Chetumal, Quintana Roo, Mexico. The data are presented by food item as percent occurrence and percent volume of the material that was analyzed (entire stomachs or subsamples). Empty stomachs (volume = 0.0 ml) were excluded from these analyses. Percent occurrence was defined as the number of stomachs in which that food item occurred. Percent volume was defined as the ratio between the total volume of a separate food taxon and the total volume of all food taxa combined. All of the volumes were based on preserved stomach contents. Game and Crop Species A total of 12 game species and 13 main crop species were considered in this study (Table 5-1). Game species were the taxa taken locally for subsistence purposes (see Chapter 3). The main crop species (11 species and 2 combined categories [i.e.. Other and Fruits]) were the taxa grown locally for 152 Table 5-1. Game species and main crop species at Ejido X-Hazil y Anexos, Quintana Roo, Mexico, during 1989-1990. Scientific name Class or Order/Family' Common name a) Game species'" Orthogeomys hispiduf Agouti paca* Dasyprocta punctata* Nasua nasua* Tayassu pecari Tayassu tajacu* Mazama americana Odocoileus virginianus Crypturellus cintiamomeus Crax rubra Ortalis vetula Agriocharis ocellata Mammalia/Geomyidae Mammalia/ Agoutidae Mammalia/Dasyproctidae Mammalia/Procyonidae Mammalia/Tayassuidae Mammalia/Tayassuidae Mammalia/Cervidae Mammalia/Cervidae Aves/Tinamidae Aves/Cracidae Aves/Cracidae Aves/Meleagrididae Pocket gopher Paca Agouti Coati White-lipped peccary Collared peccary Brocket deer White-tailed deer Thicket tinamou Great curassow Plain chachalaca Ocellated turkey b) Main crop species'^ Ipomoea batatas* Tubiflorales/ Convolvulaceae Sweet potato Cucurbita moschata* Campanulales/ Cucurbitaceae Squash Dioscorea alata' Liliales/Dioscoreaceae Yam Cnidoscolus chayamansa Geraniales/ Euphorbiaceae Chaya Manihot esculenta* Geraniales/ Euphorbiaceae Yuca Cajanus cajan Resales/ Fabaceae (Leguminosae) Lentil Zea mays* Graminales/Gramineae Com Pachyrrhizus erosus var. palmatilobus Rosales/Leguminosae Jicama Phaseolus lunatus Rosales/Leguminosae lb bean Phaseolus vulgaris* Rosales/Leguminosae Kidney bean 153 Scientific name Class or Order/Family" Common name Capsicum annum Tubiflorales/ Chili Solanaceae * Other/Zacate^ Fruits* ' Class and Family for game species (sorted taxonomically). Order and Family for main crop species (sorted alphabetically by Family). *■ Game species based on hunter survey during 1989-1990 (see Chapter 3). * = Selected game or main crop species. ' Pocket gophers were excluded from food item analyses because stomach contents were not readily available or identifiable. '' Main crop species based on gardener survey during 1989-1990 (see Chapter 6). ' The term "yam" refers primarily to Dioscorea alata, but X-Hazil Sur residents also cultivated another macal that was not identified, but may be either Colocasia esculenta or Xanthosoma yucatanense (Arales/Araceae). ' The term "other" refers primarily to zacate {Scleria lithosperma [Graminales/Cyperaceae]), but also includes an unidentified bean (Phaseolus sp.), an unidentified squash (Cucurbita sp.), and an unidentified cucumber (Cucumis sp.). * The term "fruits" was used by X-Hazil Sur residents to refer to several species of plants. For analytical purposes, these species were combined but included the following; Anacardiaceae, mango (Mangifera indica); Bromeliaceae, pineapple {Ananas sativa); Cariacaceae, papaya (Carica papaya); Cucurbitaceae, watermelon (Citrullus vulgaris); Gramineae, sugarcane (Saccharum officinarum); Lauraceae, avocado {Persea americana); Leguminosae, peanut (Arachis hypodaea); Musaceae, banana (Musa sp.); Rutaceae, lime (Citrus aurantifolia) and orange (C. sinensis); and Sapotaceae (Pouteria nuimmosa). subsistence purposes (see Chapter 6). Crops grown in house gardens for subsistence purposes or commercially in the truck gardens along Mexican Route 307 (see Chapter 2) were not included. In order to focus on the best available data during the analyses, several game species were excluded from further consideration eidier because they ate few, if any, crops or their consumption of crops was only for a limited period of time during the year. The following game species frequently were taken in gardens by hunters or frequently consumed crop species and were designated as selected game species: paca, agouti, coati, and collared peccary. The following crop species were consumed by game species and were designated as major crop species: sweet potato, squash, yuca, com, kidney bean, and other/zacate. 154 Garden Season Categories The consumption of major crop species by selected game species was analyzed with respect to time periods that reflected seasonal variations in weather as well as human activities. Three garden season categories were designated, based on weather, garden activities, timber harvest, and chicle tapping (cf., Barrera de Jorgenson, 1993; see Ch^ter 2 for additional information): One, "Clear and bum" was the period from January to April. During this period, the weather was dry and getting wanner, gardeners cleared and burned new garden sites, old gardens and any leftover crops were left unattended, timber was harvested, and by about the end of February the chicle tapping season was over. Two, "Plant and weed" was the period from May to August. During this period, the weather was rainy and hot, gardeners planted and weeded new gardens, little timber was harvested, and by late August the chicle tapping season was just beginning. Three, "Weed and harvest" was the period from September to December. During this period, the weather was rainy and getting cooler, gardeners weeded and harvested their gardens, new garden sites were selected, no timber was harvested, and the chicle tapping season was underway. Statistical Analyses The analyses in this chapter were primarily descriptive. This was due to the fact that sample sizes were too small for comparisons. The following terms were used: J = mean, SE = standard error, and n = sample size. 155 Limitations of the Data Sets Several factors must be considered when reviewing these results. One, these data were based on reports provided by cooperating hunters (see Chapter 3 for additional information). Not all game kills were reported to me. Based on my calculations, conversations with hunters, and information provided by village residents, the reported number of game animals taken probably represents about two-thirds of the total game harvest at X-Hazil Sur during the course of my data collection. Two, the hunters selected when to hunt, what to shoot, and where to go. The reported kills, thus, do not represent a random sample of the game available or the foods they would consume. Three, these data were based on animals that were killed by hunters and do not reflect animals that escaped being killed. For most game species, there were seasonal differences in the age, sex, and number of individuals taken. These results, then, could be biased if individual preferences for or against certain crop species placed a wild animal at a different risk level than other individuals in the population. As a consequence, some of the results may be biased due to the actions of hunters or the behavior of individual game animals. Another factor that must be considered while reviewing these data is that many animals, especially those taken in gardens, were killed while they had empty stomachs. Hunters reported that many of the animals taken in gardens obviously were looking for food and probably would have eaten garden crops. As a consequence, these results of crop consumption should be taken as minimum values for the importance of crops to game species. This problem will be discussed below. Results Food Items Consumed by Game Species Food items were identified for 11 of the 12 game species taken by Maya hunters at X-Hazil Sur. For two reasons, food items were not identified for the pocket gopher: One, local residents always cooked and consumed the entire pocket gopher; thus, stomach contents normally were not available for collection. Two, pocket gophers normally ground up their food into extremely fine 156 particles that could not be readily identified through visual, macroscopic examination of the material (personal observation). For each of the remaining 11 game species, the following variables were simmiarized: the number of stomachs analyzed, the volimie of stomach contents, proportion of sample analyzed (for large samples), the most frequent food items by percent occurrence and percent volume, the percent volume of plant and animal material, and the percent volume of crop species (see Chapter 6 for a description of crop species). Game Species Accounts A total of 28 food items were recorded for the paca (n = 36 stomachs; total volume analyzed = 1886.5 ml; Appendix N). By percent occurrence, the most frequent food item was fruits/seeds from Manilkara zapota (S^xjtaceae) at 47.2%. By percent volume, fruits/seeds from Manilkara zapota also was the most commonly taken food item at 26.6%. Plant material composed 99.8% (23 taxa), by percent volume, of paca stomach contents, while animal material composed 0.2% (five taxa). About 94%, by percent volume, of the paca diet consisted of fruits and seeds (Figure 5-1). By percent volume, crop species composed a total of 35.9% (four taxa: com, squash, yuca, and sweet potato) of the stomach contents. For the agouti, a total of 21 food items were recorded (n = 26 stomachs; total volume analyzed = 1167.5 ml; Appendix O). The most frequent food item was fruits/seeds from Manilkara zapota at 38.5%, by percent occurrence. Com at 25.5% and sweet potato at 20.2%, both by percent volume, were the two most commonly taken food items. By percent volume, plant material composed 99.9% (19 taxa) of agouti stomach contents, while animal material composed 0.1% (two taxa). About 98%, by percent volume, of the agouti diet consisted of fruits and seeds (Figure 5-1). Crop species (four taxa: com, sweet potato, squash, and zacate) composed a total of 59.3%, by percent volume, of the stomach contents. A total of 49 food items were recorded for the coati (n = 129 stomachs; total volume analyzed = 4678.5 ml; Appendix P). By percent occurrence, the most frequent food items were an unidentified 157 100 BO- 60 fe? 40 H 20- ^ ^ iss ^iaf¥^ Apa Dpu Nna Tpe Tta Mam Ovi GAME SPECIES r V V T- Ccl Cm Ove Aoc OTHER ANIMAL MAT LEAVES FRUITS/SEEDS Figure 5-1. Proportion (percent volume) of food items identified from stomach contents of game species taken by Maya hunters at X-Hazil Sur, Quintana Roo, Mexico, during June 1989-October 1990. (Food items: FRUITS/SEEDS = fruits or seeds, LEAVES = leaves, ANIMAL MAT = animal material, and OTHER = other [see Mediods for further information]. See Table 5-2 for key to species abbreviations.) slug (Class Gastropoda) at 50.4%, an unidentified insect (Order Orthoptera) at 44.2%, and com at 40.3%. By percent volume, com at 53.5% and fruits/seeds from Manilkara zapota at 19.0% were the two most commonly taken food items. Plant material composed 91.6% (29 taxa), by percent volume, of the stomach contents, while animal material composed 8.4% (19 taxa). About 92%, by percent volume, of the coati diet consisted of fruits and seeds, while 8% consisted of animal matter (Figure 5- 1). By percent volume, the single crop species com composed 53.5% of the stomach contents. For the while-lipped peccary, a total of 17 food items were recorded (n = 3 stomachs; total volume analyzed = 208.0 ml; Appendix Q). The most frequent food items were fruits/seeds from Brosimum alicastrum (Moraceae) at 100.0%, stems/roots from an unidentified plant called Can pe tzii at 100.0%, and leaves from Psychotria sp. (Rubiaceae) at 100.0%, by percent occurrence. Brosimum 158 alicastrum at 87.7%, by percent volume, was the most common taken food item. By percent volume, plant material composed 100.0% (14 taxa) of white-lipped peccary stomach contents, while animal material composed < 0.0% (trace amounts of three taxa). About 93%, by percent volume, of the white-lipped peccary diet consisted of fruits and seeds (Figure 5-1). No crop species were recorded for the three white-lipped peccaries. A total of 38 food items were recorded for the collared peccary (n = 29 stomachs; total volume analyzed = 1829.0 ml; Appendix R). By percent occurrence, the most frequent food items were fruits/seeds, leaves, and stems/roots from Psychotria sp. at 79.3% and fruits/seeds from Brosimum alicastrum at 75.9%. By percent volume, Psychotria sp. at 23.8% was the most commonly taken food item. Plant material composed 100.0% (32 taxa), by percent volume, of collared peccary stomach contents, while animal material composed < 0.0% (trace amounts of five taxa). About 58%, by percent volume, of the collared peccary diet consisted of fruits and seeds, while 21% consisted of items categorized as other, and 20% consisted of leaves (Figure 5-1). By percent volume, crop species (four taxa: squash, com, sweet potato, and zacate) composed a total of 25.5% of the stomach contents. For the brocket deer, a total of 32 food items were recorded (n = 8 stomachs; total volume analyzed = 446.0 ml; Appendix S). The most frequent food items were fruits/seeds, leaves, and stems/roots from Psychotria sp. at 87.5%, fruits/seeds and leaves from Brosimum alicastrum at 62.5%, leaves from Eugenia sp. (Myrtaceae) at 50.0%, and fruits/seeds from an unidentified plant at 50.0%, by percent occurrence. Psychotria sp. at 58.3%, by percent volume, was the most commonly taken food item. By percent volume, plant material composed 100.0% (30 taxa) of brocket deer stomach contents, while animal material composed < 0.0% (trace amounts of one taxon). About 42%, by percent volume, of the brocket deer diet consisted of leaves, while 33% consisted of fruits and seeds, and 25% consisted of items categorized as other (Figure 5-1). The single crop species zacate composed a total of 0.2%, by percent volume, of the stomach contents. A total of 43 food items were recorded for the white-tailed deer (n = 1 1 stomachs; total volume analyzed = 601.0 ml; Appendix T). By percent occurrence, the most frequent food item was 159 leaves and stems/roots from Psychotria sp. at 72.7%. By percent volume, Psychotria sp. at 42.7% was the most commonly taken food item. Plant material composed 100.0% (40 taxa), by percent volume, of white-tailed deer stomach contents, while animal material composed < 0.0% (trace amounts of two taxa). About 57%, by percent volume, of the white-tailed deer diet consisted of leaves, while 25% consisted of items categorized as other, and 18% consisted of fruits and seeds (Figure 5-1). By percent volume, the single crop species kidney bean composed a total of 9.0% of the stomach contents. For the thicket tinamou, a total of seven food items were recorded (n = 3 stomachs; total volume analyzed = 13.0 ml; Appendix U). The most frequent food items were fruits/seeds from an unidentified plant at 100.0%, Chrysophyllum mexicanum (Sapotaceae) at 66.7%, and fruits/seeds from Galactia striata (Leguminosae) at 66.7%, by percent occurrence. The unidentified plant at 69.2%, by percent volume, was the most commonly taken food item. By percent volume, plant material composed 96.2% (six taxa) of thicket tinamou stomach contents, while animal material composed 3.8% (one taxon). About 96%, by percent volume, of the thicket tinamou diet consisted of fruits and seeds (Figure 5-1). The single crop species zacate composed a total of < 0.0% (trace amounts), by percent volume, of the stomach contents. A total of 26 food items were recorded for the great curassow (n = 13 stomachs; total volume analyzed = 694.0 ml; Appendix V). By percent occurrence, the most frequent food items were plastic, rocks, and soil at 61.5%; fruits/seeds from Diospyros sp. (Ebenaceae) at 53.8%; and fruits/seeds from Manilkara zapota at 53.8%. By percent volume, Brosimum alicastrum at 34.5% was the most commonly taken food item. Plant material composed 98.5% (23 taxa), by percent volume, of great curassow stomach contents, while animal material composed < 0.0% (trace amounts of two taxa). About 95%, by percent volume, of the great curassow diet consisted of fruits and seeds (Figure 5-1). By percent volume, crop species (three taxa: kidney bean, squash, and zacate) composed a total of 4.2% of the stomach contents. For the plain chachalaca, a total of 20 food items were recorded (n = 21 stomachs; total volume analyzed = 329.5 ml; Appendix W). The most frequent food items were fruits/seeds and 160 leaves from Chrysophyllum mexicanum at 23.8% and fruits/seeds from Metopium brownei (Anacardiaceae) at 23.8%, by percent occurrence. Metopium brownei at 69.2%, by percent volume, and Sabal yapa (Palmae) at 17.9% were the two most commonly taken food items. By percent volume, plant material composed 100.0% (20 taxa) of plain chachalaca stomach contents, while animal material was not recorded. About 86%, by percent volume, of the plain chachalaca diet consisted of fruits and seeds (Figure 5-1). The single crop species zacate composed a total of 6.7%, by percent volume, of the stomach contents. For the ocellated turkey, a total of 17 food items were recorded (n = 5 stomachs; total volume analyzed = 211.5 ml; Appendix X). The most frequent food items at 60.0% each by percent occurrence were fruits/seeds and leaves from zacate; leaves from Psychotria sp.; plastic, rocks, and soil; and fruits/seeds from an unidentified plant. Com at 24.6%, kidney bean at 18.9%, and Manilkara zapota at 12.5%, by percent volume, were the three most commonly taken food items. By percent volume, plant material composed 92.9% (15 taxa) of ocellated turkey stomach contents, while animal material composed 3.8% (one taxon). About 68%, by percent volume, of the ocellated turkey diet consisted of fruits and seeds, while 25% consisted of leaves (Figure 5-1). Crop species (three taxa: zacate, com, and kidney bean) composed a total of 51.1%, by percent volume, of the stomach contents. In summary, the 11 game species consumed collectively a total of 86 plant taxa (80 noncrop species and 6 crop species [sweet potato, squash, yuca, com, kidney bean, and zacate]), 25 animal taxa, and one "taxon" that consisted of plastic, rocks, and soil. The average number of food items per game species was 27.1 taxa (SE = 3.8, n = 11 game species, range = 7-49 food items per game species). By percent volume, plant matter (primarily fruits and seeds) composed 92-100% each of the diet for each species. The coati consumed the greatest proportion of animal matter at 8%, by percent volume. The average number of crop taxa per game species was 2.1 taxa (SE = 0.5, n = 11 game species, range = 0-4 crops per game species). 161 Crop Species Consumed by Game Species Empty stomachs. Not all of the stomachs obtained from harvested wild animals contained food. A total of 22 individuals (ca. 4% of 584 game animals taken) had empty stomachs at the time they were taken (Table 5-2). Coatis had the greatest number of empty stomachs at 14 (ca. 8% of 167 individuals), while no empty stomachs were reported for seven game taxa. Major crop species. Many of the main crop species were not consumed by game species. Of the 13 main crop species planted by gardeners (Table 5-1), seven taxa were not recorded in the stomach contents of game species, and will not be considered here further: yam, chaya, lentil, jicama, ib bean, chili, and fruits. For purposes of this ch^ter, the remaining six crop species will be renamed here as major crop species and will form the basis for the following discussion: sweet potato, squash, yuca, com, kidney bean, and other/zacate. Among the major crop species, com and squash were the two most frequent taxa in the stomachs of game species with food, by percent occurrence, at 25% and 6%, respectively (n = 284 stomachs analyzed; Table 5-2). Zacate (taken by 7 of 11 game species) and com (taken by 5 of 11 game species) were the major crop species most widely taken by game species. The number of crop taxa in stomach contents of game species varied by game taxon (Table 5- 2). The paca, agouti, and collared peccary consumed the greatest number of crop taxa, at four each, whereas the white-lipped peccary consumed no crop taxon and the coati, brocket deer, white-tailed deer, thicket tinamou, and plain chachalaca consumed a single crop taxon each. The percent occurrence of major crop species in stomach contents of game species varied by taxon (Table 5-2). The ocellated turkey had the highest percent occurrence of major crop species at 80% (n = 5 stomachs), followed by the agouti (percent occurrence = 50%, n = 26 stomachs analyzed), paca (percent occurrence = 42%, n = 36 stomachs analyzed), and coati (percent occurrence = 40%, n = 129 stomachs analyzed). Overall, 35% of the stomachs that were not empty contained at least one major crop species. 162 o > 3 •s N B C o •s 00 3 t u u ^ s Q. c3 O ©X) 60 w^ «^ o 5J V5 O .* 1 c- B O 1> •3 u so B 3 '■5 3 "u c/3 X (U U 3 o* OV B 2 C/5 y s c X: H. " O o a. o o o\ o 00 ON b .o_ u 2 B Q 3 (M •— » o SO 8 B B 3 S T3 3 5" O u a o -^ i 2 ri Cu •-H II r4 lA "S N _5J >> 3 19 e2 i o o < > O 3 u u o H 3 Q '* o fS 00 ^ m vC fO «N '" d o o 00 m m o d so in p4 o. o S ^ c« W u s g- — ■x ^ 1 C/5 H 3 00 •3 O 13 2 163 Seasonal Variation in Plant Parts Consumed by Game Species Among the major CTop species, the primary plant parts consumed were fruits and seeds (Table 5-3). However, game species also consumed tubers (e.g., sweet potato), roots (e.g., yuca), and leaves (e.g., kidney bean and zacate). The data suggested that there were seasonal differences in the frequency at which the different plant parts were consumed. Fruits and seeds primarily were consumed during two seasons: (1) plant and weed and (2) weed and harvest. Leaves and plant parts categorized as other, however, primarily were consumed during the season called clear and bum. Due to small sample sizes, it was not possible to test for seasonal differences in consumption by plant part. Selected Game Species Not all game species at X-Hazil Sur consumed substantial amounts of crop species. This suggests that many game species obtained at least some of their food from the forest and were not dependent upon garden crops for food. For example, no crop species were recorded for stomach contents of the white-lipped peccary (n = 3 stomachs; Table 5-2). Further, of the 11 game species (excluding pocket gophers; Table 5-1), four taxa (coati, brocket deer, white-tailed deer, thicket tinamou, and plain chachalaca) consumed only a single crop species each. Except for the coati, which consumed a substantial amount of com, the contribution of crops to the diets of the remaining three game species was relatively minor. For this reason, these game species were not included in the remaining analyses. In addition, the recorded consumption of crop species by the great curassow (n = 13 crop/gizzard samples) and the ocellated turkey (n = 5 crop/gizzard samples) were based on small sample sizes. In order to focus primarily on those game species that frequently ate crops, except for the coati, which consumed substantial amounts of com, these seven species will not be considered here further. For purposes of this chapter, the remaining four game species (paca, agouti, coati, and collared peccary) will be renamed here as selected game species and will form the basis for the following discussion. 164 & > •o a N 13 c o oc 3 u "o so O u o 2 I s ■« O 3 f o — OS .a u sr 2 E B o ^ so c s , (U s fc -o i^ OJ ^ i "^ ■^ "3 Q > o. o C/2 ^— ' «^ 6 s Z 2 > !/i i3 J3 ■o i 1^3 a. u ^ lume (ml) Plan Q so 3 > < >< CS 2 Z 2 i •g aS ^ U B » 'J rs J9 % '>; O ^ 1 Q. Cu olume (ml)' Plan _^_^ > ka D. < r- H . t) 1 No torn. X> lys T3 i w CO 2 u 1 1 P3 i 3 o (N C/0 ^^ II c<-, T5 vA s JJ ^ •3 ^ g ■51- \C m [^ m ^•^ ^^ 0 J U- Si [i, ^ u. ^ ^ ^ ^ s q o 2 d o d ON in ^ Q r- tN vO <^' -" V ^ 3 3" C/5 O B o -§ i2 m oc o fn in O en o q en - oo o o 8 73 i c/ t c 3 3 Total volume (ml) 1 £ s 1 f2 >> 3 i B O B 13 B O > .o X u •3 o X 60 c 3 -Si •5 ■5 > ^ -3 ■r: -g u D. re o B d sog 1) ^ C/5 B IE S o u S a so s B ^ O •O 1) 3 -o ^ re o so B O U >s s o 73 -"SI! i§ S ^ 33 1/1 O >> D. q/5 fe so y s> •a "^ ^ 3 " r w« 3 U u o _ IS -^ '/! ™ _^ b u c o •« o 're' >> '5 a -^2 rs . S £ 3 a va •O = .S .^ g in O B u B K O V2 ?73 w a B o « 8 :2 u -a Cu ^ § o « Z H jg 165 Seasonal Variation in Consumption of Major Crop Species by Selected Game Species The proportion of major crop species in the stomach contents of selected game species varied by season. The paca consimied four major crop species and exhibited seasonal differences in the proportion of these taxa in the diet, by percent volume (Table 5-4). For example, during the weed and harvest season, 64% of the stomach contents were major crop species (three taxa: primarily com [33% by percent volume], squash [25% by percent volume], and sweet potato [7% by percent volume]). During the other two seasons, major crop species represented about 18-20% of the stomach contents (one-two taxa), by percent volume. The agouti consumed four major crop species per season and exhibited seasonal differences in the proportion of these taxa in the diet, by percent volume (Table 5-4). For example, during the weed and harvest season, 68% of the stomach contents were major crop species (three taxa: primarily com [52% by percent volume], squash [16% by percent volume], and other/zacate). During the clear and bum season, agoutis primarily consumed sweet potatoes (27% by percent volume) and com (15% by percent volume), while during the plant and weed season, agoutis primarily consumed sweet potatoes (33% by percent volume) and squash (22% by percent volume). The coati consumed a single major crop species, com, during two of the three seasons and exhibited seasonal differences in the proportion of this taxon in Uie diet, by percent volume (Table 5-5). During the weed and harvest season, 66% by percent volume of the stomach contents were com, while during the plant and weed season com represented 16% by percent volume of the stomach contents. No major crop species were consumed by the coati during the clear and bum season. The collared peccary consumed four taxa of major crop species and exhibited seasonal differences in the proportion of these taxa in the diet, by percent volume (Table 5-5). During the weed and harvest season, 41% by percent volume of the stomach contents were major crop species, primarily com (41% by percent volume). During the plant and weed season, 23% by percent volume of the stomach contents were major crop species, primarily sweet potatoes (12% by percent volume) and 166 Table 5-4. Seasonal variation of major crop species in (a) paca stomach contents (n = 36 stomach samples, total volume analyzed = 1886.5 ml) and (b) agouti stomach contents (n = 26 stomach samples, total volume analyzed = 1167.5 ml), expressed as percent occurrence (% Occ.) and volume (% Vol.). Major crop species'" Garden season" Clear and bum (Jan-Apr) Plant and weed (May-Aug) % Occ. (% Vol.)'' % Occ. (% Vol.) Weed and harvest (Sep-Dec) % Occ. (% Vol.) a) Paca Com Squash Sweet potato Yuca 15.0 15.0 0.0 5.0 (5.2) 0.0 (0.0) (7.5) 100.0 (20.2) (0.0) 0.0 (0.0) (5.2) 0.0 (0.0) 42.9 (32.5) 21.4 (24.5) 7.1 (6.9) 0.0 (0.0) Volume (ml) (1034.1) (124.0) (728.0) No. stomachs 20 14 b) Agouti Com 22.2 (14.5) 11.1 (1.5) 37.5 (51.5) Squash 0.0 (0.0) 22.2 (20.5) 25.0 (16.1) Other/ 0.0 (0.0) 0.0 (0.0) 12.5 « 0.0) Zacate Sweet potato 11.1 (27.3) 33.3 (39.8) 0.0 (0.0) Volume (ml) (289.0) (394.5) (484.0) No. stomachs 9 9 8 " See text for additional information about the three garden season categories. '' See text and Table 5-2 for additional information about how the "major crop species" were determined. For the paca, kidney bean and other/zacate did not occur in stomach contents. For the agouti, yuca and kidney bean did not occur in stomach contents. ' Percent occurrence in stomach contents, by garden season. '' Percent volume in stomach contents, by garden season. 167 Table 5-5. Seasonal variation of major crop species in (a) coati stomach contents (n = 129 stomach samples, total volume analyzed = 4678.5 ml) and (b) collared peccary stomach contents (n = 29 stomach samples, total volume analyzed = 1829.0 ml), expressed as percent occurrence (% Occ.) and volume (% Vol.). Major crop species'" Clear and bum (Jan-Apr) Garden season" Plant and weed (May-Aug) Weed and harvest (Sep-Dec) % Occ' (% Vol.)'' % Occ. (% Vol.) % Occ. (% Vol.) a) Coati Com 0.0 (0.0) 7.7 (16.2) 54.3 (66.0) Volume (ml) (652.0) (309.0) (3717.5) No. stomachs 22 13 94 b) Collared peccary Sweet potato 0.0 (0.0) 11.1 (11.9) 12.5 (< 0.0) Squash 0.0 (0.0) 16.7 (10.7) 12.5 (< 0.0) Cora 0.0 (0.0) 5.6 (< 0.0) 25.0 (40.9) Other/ 0.0 (0.0) 5.6 (0.3) 0.0 (0.0) Zacate Volume (ml) (241.0) (1013.0) (575.0) No. stomachs 18 8 See text for additional information about the three garden season categories. See text and Table 5-2 for additional information about how the "major crop species" were determined. For the coati, sweet potato, squash, yuca, kidney bean, and other/zacate did not occur in stomach contents. For the collared peccary, yuca and kidney bean did not occur in stomach contents. Percent occurrence in stomach contents, by garden season. Percent volume in stomach contents, by garden season. 168 squash (11% by percent volume). No major crop species were consumed by the collared peccary during the clear and bum season. In summary, only a subset of the main crops were consumed in substantial quantities by game species. These were the sweet potato, squash, yuca, com, kidney bean, and other/zacate. In addition, only a subset of the game species consumed substantial quantities of crop species. These were the paca, agouti, coati, and collared peccary. Discussion According to Linares (1976), the practice of garden hunting is based on the fact that certain species of terrestrial mammals are tolerant of human disturbances and feed regularly on cultivated crops. Since these species congregate in areas with cultivated aops, the biomass of these game animals is greater in the vicinity of the gardens than in the adjacent forest. This increased biomass of wild animals, according to Linares (1976), supports a game-procurement system by human hunters that exploits the diet and behavioral characteristics of these mammals. The results obtained during this study confirmed that many game species consumed crops, as well as other plant and animal matter. In the first part of this discussion, I will generally desaibe game food habits at Ejido X-Hazil y Anexos, compare them with food habits reported for other areas, and summarize briefly the behavior and ecology of game species as they relate to food habits. In the second part, I will describe the percent frequency and occurrence of crop species in game diets. In the third part, I will discuss seasonal variation in crop consumption by game species. In conclusion, I will suggest that crops and gardens are important food sources for game species at Ejido X-Hazil y Anexos but that noncrop species also are important to game animals. Food Items Consumed by Game Species Pacas are noctumal frugivore-granivores (Eisenberg, 1981). They are not especially social animals and usually live alone (Collett, 1981; Leopold, 1977). Pacas are found in a variety of habitat 169 types, but usually seek forested areas near water (Grimwood, 1969; Matamoros H., 1982; Nowak and Paradise, 1983). On the Yucatan Peninsula, pacas reportedly do not dig burrows (Hall and Dalquest, 1963; Leopold, 1977), but burrow digging has been noted in other areas (Baker, 1974). At Ejido X-Hazil y Anexos, plant material (primarily fruits and seeds) composed 99.8% (23 taxa reported), by percent volume, of paca stomach contents, whereas animal material composed 0.2% (five taxa reported). A total of 28 taxa of food items were reported (Figure 5-1; Appendix N). Among mammals at Ejido X-Hazil y Anexos, pacas ranked second among seven taxa (excluding pocket gophers) in percent volume of fruits and seeds in the diet and ranked fifth in the total number of taxa of food items consumed. Elsewhere, the diet consists of leaves, stems, roots, seeds, and fallen fruits (especially avocados and mangos) (Chavez Le6n, 1983; Nowak and Paradiso, 1983; Smythe et al., 1982). Agoutis are diurnal frugivore-granivores (Eisenberg, 1981; Smythe, 1983). They are social animals and usually occur as a mated pair (Smythe, 1978). Agoutis live closely associated with water and are found in forests, thick brush, savannahs, and cultivated areas (Grimwood, 1969; Nowak and Paradiso, 1983). Agoutis, unlike pacas, do not dig burrows. Agoutis, however, cache fruits and seeds during periods of food abundance and retrieve those items when food is scarce (Murie, 1977). At Ejido X-Hazil y Anexos, plant material (primarily fruits and seeds) composed 99.9% (19 taxa reported), by percent volume, of agouti stomach contents, whereas animal material composed 0.1% (two taxa reported). A total of 21 taxa of food items were reported (Figure 5-1; Appendix O). Among mammals at Ejido X-Hazil y Anexos, agoutis ranked first among seven taxa in percent volume of fruits and seeds in the diet and ranked sixth in the total number of taxa of food items consumed. Elsewhere, the diet consists of fruits, vegetables, and succulent plants (Chavez Le6n, 1983; Hallwachs, 1986; Nowak and Paradiso, 1983; Smythe, 1983; Smythe et al., 1982). Coatis are diurnal frugivore-omnivores (Eisenberg, 1981), but adult males may forage at night. Primarily frugivores when fruits are available, coatis instead consume rodents and invertebrates during times when fruits are scarce (Smythe, 1970). While adult male coatis usually are solitary, adult 170 females and their young are highly social and occur in large groups that on Barro Colorado Island had 4-20 individuals (Kaufmann, 1962; Smythe, 1970; Russell, 1981). Coatis are found in wooded areas and forage in trees, as well as on the ground (Kaufmann, 1983; Kaufmann et al., 1976; Nowak and Paradiso, 1983). At Ejido X-Hazil y Anexos, plant material (primarily fruits and seeds) composed 91.6% (29 taxa reported), by percent volume, of coati stomach contents, whereas animal material composed 8.4% (19 taxa reported). A total of 49 taxa of food items were reported (Figure 5-1; Appendix P). Among mammals at Ejido X-Hazil y Anexos, coatis ranked fourth among seven taxa in percent volume of fruits and seeds in the diet and ranked first in the total number of taxa of food items consumed. At 8.4% by percent volume, coatis ranked first among seven taxa in the proportion of animal material in the diet. Elsewhere, the diet consists of both plant and animal matter (Bisbal E., 1986; Chavez Le6n, 1983; Nowak and Paradiso, 1983). White-lipped peccaries are nocturnal frugivore-herbivores (Eisenberg, 1981). They are highly social and can occur in large groups of several hundred individuals (Kiltie and Terborgh, 1983; Leopold, 1977; Sowls, 1983, 1984). White-lipped peccaries occupy large home ranges and are found in several habitat types, including desert scrub, arid woodlands, and rain forest (Donkin, 1985; Grimwood, 1969; Nowak and Paradiso, 1983). These peccaries frequent waterholes. At Ejido X-Hazil y Anexos, plant material (primarily fruits and seeds) composed 100.0% (14 taxa reported), by percent volume, of white-lipped peccary stomach contents, whereas animal material composed < 0.0% (trace amounts of three taxa reported). A total of 17 taxa of food items were reported (Figure 5-1; Appendix Q). Among mammals at Ejido X-Hazil y Anexos, white-lipped peccaries ranked third among seven taxa in percent volume of fruits and seeds in the diet and ranked seventh in the total number of taxa of food items consumed. Elsewhere, the diet consists of cactus fruit, berries, tubers, bulbs, rhizomes, grubs, snakes, and small vertebrates (Donkin, 1985; Kiltie, 1981; Nowak and Paradiso, 1983; Sowls, 1984). 171 Collared peccaries, like white-lipped peccaries, are nocturnal frugivore-herbivores (Eisenberg, 1981). These peccaries are highly social but occur in smaller groups than do white-lipped peccaries. Group size ranges from 2-50, but usually is 5-15 individuals (Castellanos, 1983; Donkin, 1985; Nowak and Paradiso, 1983; Robinson and Eisenberg, 1985). Collared peccaries occupy large home ranges and, like white-lipped peccaries, are found in scrub, woodlands, and forest (Bigler, 1974; McCoy et al., 1990; Schweinsburg, 1971; Sowls, 1974, 1978, 1983, 1984). At Ejido X-Hazil y Anexos, plant material (primarily fruits and seeds) composed 100.0% (32 taxa reported), by percent volume, of collared peccary stomach contents, whereas animal material composed < 0.0% (trace amounts of five taxa reported). A total of 38 taxa of food items were reported (Figure 5-1; Appendix R). Among mammals at Ejido X-Hazil y Anexos, collared peccaries ranked second among seven taxa in percent volume of fruits and seeds in the diet and ranked third in the total number of taxa of food items consumed. Elsewhere, the diet consists of fruits, berries, tubers, bulbs, rhizomes, grubs, snakes, vegetables, and small vertebrates (Chavez Le6n, 1983; Day, 1986; Donkin, 1985; KilUe, 1981; McCoy et al., 1983; Nowak and Paradiso, 1983; Sowls, 1984). Brocket deer are frugivore-herbivores (Eisenberg, 1981). Diurnal, nocturnal, and crepuscular activity has been reported (Nowak and Paradiso, 1983). These deer are solitary, except during the mating season. Brocket deer are shy, sedentary, and are found in woodlands and forests (Chavez Le6n, 1983). At Ejido X-Hazil y Anexos, plant material (primarily leaves) composed 100.0% (30 taxa reported), by percent volume, of brocket deer stomach contents, whereas animal material composed < 0.0% (trace amounts of one taxon reported). A total of 32 taxa of food items were reported (Figure 5- 1; Appendix S). Among mammals at Ejido X-Hazil y Anexos, brocket deer ranked sixth among seven taxa in percent volume of fruits and seeds in the diet and ranked fourth in the total number of taxa of food items consumed. At ca. 42% by percent volume, brocket deer ranked second in the proportion of leaves in the diet. Elsewhere, the diet consists of grasses, vines, and tender green shoots (Nowak and Paradiso, 1983). In Suriname (Branan et al., 1985), red brocket deer rumens contained food material 172 from at least 57 plant species (n = 75 rumens collected during 8 months), whereas for white-tailed deer, only 14 plant species were identified (n = 13 rumens). White-tailed deer are crepuscular herbivore-browsers (Eisenberg, 1981). These deer are social, but two different types of groups have been reported. Adult males are solitary or occur in small groups with other males, whereas adult females occur in groups with their yearling daughter and fawns of the year (Marchington and Hirth, 1984). White-tailed deer are found in a great variety of habitat types, but they avoid dense forests (Nowak and Paradise, 1983). At Ejido X-Hazil y Anexos, plant material (primarily leaves) composed 100.0% (40 taxa reported), by percent volume, of white-tailed deer stomach contents, whereas animal material composed < 0.0% (trace amounts of two taxa reported). A total of 43 taxa of food items were reported (Figure 5-1; Appendix T). Among mammals at Ejido X-Hazil y Anexos, white-tailed deer ranked seventh among seven taxa in percent volume of fruits and seeds in the diet and ranked second in the total number of taxa of food items consumed. At ca. 57% by percent volume, white-tailed deer ranked first in the proportion of leaves in the diet. Elsewhere, the diet consists of grasses, weeds, shrubs, twigs, mushrooms, nuts, lichens, vegetables, and vegetable leaves (Chavez Le6n, 1983; Mandujano and Rico- (jray, 1991; Nowak and Paradiso, 1983; Vaughan and Rodriguez, 1991). At Ejido X-Hazil y Anexos, unlike in Suriname, brocket deer (32 taxa reported) and white-tailed deer (43 taxa reported) consumed about the same number of plant taxa. Thicket tinamous are diurnal frugivore-granivores that forage on the ground (Paynter, 1955a, 1955b; Slud, 1964). They are not especially social and usually occur as solitary individuals (Delacour and Amadon, 1973). Thicket tinamous are found in dry woodlands, especially along forest edges, but seldom enter open areas (Leopold, 1977). Tinamous do not scratch for food with their feet (Lancaster, 1983). At Ejido X-Hazil y Anexos, plant material (primarily fruits and seeds) composed 96.2% (six taxa reported), by percent volume, of thicket tinamou stomach contents, whereas animal material composed 3.8% (one taxon reported). A total of 32 taxa of food items were reported (Figure 5-1; 173 Appendix U). Among birds at Ejido X-Hazil y Anexos, thicket tinamous ranked first among four taxa in percent volume of fruits and seeds in the diet and ranked fourth in the total number of taxa of food items consumed. Elsewhere, the diet consists of seeds, fruits, and insects (Chavez Le6n, 1983; Lancaster, 1983). Great curassows are diurnal frugivore-herbivores that forage on the ground (Karr et al., 1990; Paynter, 1955a, 1955b; Slud, 1964). They occur individually or in small groups (Delacour and Amadon, 1973). Primarily terrestrial, great curassows are found in ravines, shrubby areas, and humid woodlands (Amadon, 1983; MacKinnon Vda. de Monies, 1989). At Ejido X-Hazil y Anexos, plant material (primarily fruits and seeds) composed 98.5% (23 taxa reported), by percent volume, of great curassow stomach contents, whereas animal material composed < 0.0% (trace amounts of two taxa reported). A total of 26 taxa of food items were reported (Figure 5-1; Appendix V). Among birds at Ejido X-Hazil y Anexos, great curassows ranked second among four taxa in percent volume of fruits and seeds in the diet and ranked first in the total number of taxa of food items consumed. Elsewhere, the diet consists of fruits, seeds, succulent leaves, and insects (Amadon, 1983; Chavez Le6n, 1983; Mendez, 1979). Plain chachalacas are diurnal frugivore-herbivores that forage on then foliage in the canopy (Paynter, 1955a, 1955b; Slud, 1964). They are social and usually occur in small groups (Delacour and Amadon, 1973). Primarily arboreal, plain chachalacas are found in a variety of dry habitat types, including fallow areas, agricultural fields, and forest edges (Leopold, 1977; MacKinnon Vda. de Montes, 1989; Slud, 1964). At Ejido X-Hazil y Anexos, plant material (primarily fruits and seeds) composed 100.0% (20 taxa reported), by percent volume, of plain chachalaca stomach contents, whereas animal material was not reported. A total of 20 taxa of food items were reported (Figure 5-1; Appendix W). Among birds at Ejido X-Hazil y Anexos, plain chachalacas ranked third among four taxa in percent volume of fruits and seeds in the diet and ranked second in the total number of taxa of food items consumed. Elsewhere, the diet consists of fruits and seeds (Leopold, 1977). 174 Ocellated turkeys, endemic to Mexico, are diurnal frugivore-omnivores that forage on the ground (Paynter, 1955a, 1955b). They occur as solitary individuals or mated pairs. Ocellated turkeys are found in forested areas adjacent to fallow areas or agricultural fields (Leopold, 1977; MacKinnon Vda. de Monies, 1989). At Ejido X-Hazil y Anexos, plant material (primarily fruits and seeds) composed 92.9% (15 taxa reported), by percent volume, of ocellated turkey stomach contents, whereas animal material composed 3.8% (one taxon reported). A total of 17 taxa of food items were reported (Figure 5-1; Appendix X). Among birds at Ejido X-Hazil y Anexos, ocellated turkeys ranked fourth among four taxa in percent volume of fruits and seeds in the diet and ranked third in the total number of taxa of food items consumed. Elsewhere, the diet consists of fruits, seeds, and insects (Leopold, 1977). In simunary, fruits and seeds were the primary food items at Ejido X-Hazil y Anexos for both birds and mammals, by percent occurrence and volume, for 9 of the 11 taxa analyzed (Figure 5-1). Among mammals, the paca, agouti, coati, and white-lipped peccary consumed relatively large proportions of fruits and seeds (^ 92% by volume), while the collared peccary consumed a moderate proportion (58% by volume), and the brocket deer and white-tailed deer consumed relatively low proportions of fruits and seeds (33% and 18% by volume, respectively; Figure 5-1). The brocket deer and white-tailed deer, on the other hand, primarily consumed leaves (42% and 57% by volume, respectively). Among birds, all four taxa primarily consumed fruits and seeds (68-96% by volume). Animal material, primarily invertebrates, was important in the diet of three taxa; coati, thicket tinamou, and ocellated turkey. Food habits for game birds and mammals at Ejido X-Hazil y Anexos generally agree with those reported elsewhere. Crop Species Consumed bv Game Species The consumption of crops is critical to the theory of garden hunting (Linares, 1976). Whereas Linares (1976) described garden hunting only with respect to terrestrial mammals, at Ejido X-Hazil y Anexos, both game birds and mammals consumed crops. 175 Six crops of the 13 planted by Maya gardeners at Ejido X-Hazil y Anexos were consumed by game species: com, squash, other/zacate, sweet potato, kidney bean, and yuca (see Chapter 6). These species are among the most abundant crops by percent occurrence in gardens and number of stalks/m^. An analysis of crop consumption patterns by game species indicates that there are differences among game species. By percent volume of crops in game stomachs, die agouti (59.3%), coati (53.5%), and ocellated turkey (51.1%) ranked highest among game species. By percent occurrence of stomachs with ^ one crop species, Uie ocellated turkey (80.0%), agouti (50.0%), paca (41.7%), and coati (40.3%), ranked highest among game species (Table 5-2). By number of crop taxa consumed, the paca (4 crop taxa), agouti (4), and collared peccary (4), ranked highest among game species. This shows that crops are relatively important to five taxa (paca, agouti, coati, collared peccary, and ocellated turkey), whereas crops are relatively unimportant to six taxa (white-lipped peccary, brocket deer, white-tailed deer, thicket tinamou, great curassow, and plain chachalaca). A comparison of percent volume of crops in game stomachs and percent occurrence of kill sites in areas categorized as Combined/Early Secondary Forest (primarily gardens and fallow areas), shows that gardens may be important to some game species for reasons other than the availability of crops. Four taxa had relatively low consumption levels of crops and relatively high proportions of kill sites in areas categorized as Combined/Early Secondary Forest; brocket deer, white-tailed deer, thicket tinamou, and plain chachalaca; Figure 5-2). One possible reason for this is that these species could be ignoring crops, but feeding on plant or animal material that occurs in gardens and fallows. Another possible reason is that gardens and fallows may be important as nesting sites or for social interactions. Seasonal Variation in Consumption of Major Crop Species by Selected Game Species Crops are not equally available in gardens throughout the year. Depending upon when the seasonal rains begins, crops are planted during May-July (Figure 6-1). Crops are harvested during November- January, but this also depends upon the weather and the specific type of crop (see Chapter 6). Com, for example, is harvested at this time, but squash and sweet potatoes may be left in the 176 TAXA Apa Dpu Nna Tpe Tta Mam Oyl Cci Cru Ove Aoc 100 go 60 40 20 % CROPS "1 1 1 1 1 — 0 20 40 60 80 % KILL SITES 100 Figure 5-2. Percent volume of CTOps in game stomachs (n = 284 stomachs analyzed) and percent occurrence of kill sites in areas categorized as Combined/Early Secondary Forest (n = 584 game kills). gardens for several additional months as they do not spoil quickly. Zacate, on the other hand, grows throughout the year. Among major crop species, there were seasonal differences in the percent occurrence and volume of CTOps in stomach contents for all game species combined (Table 5-3). Three crop taxa were consumed only during a single garden season. The sweet potato, kidney bean, and yuca were 177 consumed only during the clear and bum season, when mature plants, fruits, and seeds from the previous growing season were still available in the garden. Three crop taxa were consumed throughout the year; com, squash, and other/zacate. The plant parts consumed of these three taxa, however, varied by season. Fmits and seeds of com and squash were consumed throughout the year. Com was available throughout the year for several reasons: One, some com plants never were harvested due to garden failures. Two, some cobs were stored in huts near gardens, but crop predators raided these sites. Three, some com germinated in fallow areas. Zacate leaves, however, were consumed during the clear and bum season while zacate fruits and seeds were consumed during the rest of the year. Among selected game species, there were seasonal differences in the percent occurrence and volume of crops in stomach contents for the major crop species. For pacas, the major period of crop consumption was the weed and harvest season when com (42.9% by occurrence and 32.5% by volume) and squash (21.4% by occurrence and 24.5% by volume; Table 5-4a) were the main crops consumed. For agoutis, the major period of crop consumption also was the weed and harvest season when com (37.5% by occurrence and 51.5% by volume; Table 5-4b) was the main crop consumed. For coatis, the major period of crop consumption also was the weed and harvest season when com (54.3% by occurrence and 66.0% by volume; Table 5-5a) was the only crop consumed. For collared peccaries, the major period of crop consumption also was the weed and harvest season when com (25.0% by occurrence and 40.9% by volume; Table 5-5b) was the main crop consumed. While the consumption of other/zacate, sweet potatoes, kidney beans, and yuca by game species cannot be ignored, these data suggest that the consumption of crops by game species is primarily that of com and squash during the weed and harvest season. In conclusion, these data generally support the garden hunting hypothesis (Linares, 1976) by confirming that several species of game birds and mammals consume garden crops. These data indicate, however, that game species do not depend exclusively upon crops. For the 1 1 game taxa considered, on average at least 30% of the stomach contents were not crop species. In addition, crops largely were unavailable during four to eight months of the year. Given the seasonal nature of game 178 kills, it was not always possible to determine game diets throughout the entire year. These data, however, show that crops are important to some game species at least part of the year. CHAPTER 6 CHARACTERISTICS OF MAYA GARDENS AND THEIR USE BY WILDLIFE IN QUINT ANA ROO, MEXICO Introduction For the Maya today, as for the past 3,000-4,000 years, daily life is centered around their gardens and the production of com, the principal CTop (Redfield and Villa Rojas, 1962; Steggerda, 1941). A successful garden provides esteem to the gardener, as well as abundant food resources for the family and its domestic animals (e.g., cattle, pigs, chickens, or turkeys). In addition, a successful crop harvest is evidence to the gardener, as well as to other villagers, that he and his family have fulfilled their spiritual obligations to the numerous Mayan saints and spirits. Natural events such as crop failures or low crop yields, according to the Maya, have spiritual causes and require spiritual solutions (cf., Bums, 1983:202-225). In this sense, the garden has both a practical and a religious context (Villa Rojas, 1987). Recognizing the importance of gardens to the Maya and other subsistence farmers in Latin America, much has been written about gardens and their characteristics (cf. Conklin, 1961; Ewell and Merrill-Sands, 1987; Watters, 1971). These studies suggest that the practice of gardening has changed little in the Yucatan Peninsula over the past 500 years (Hammond, 1982a, 1982b; Landa, 1978; Morley, 1956; Roys, 1972; Turner, 1974, 1990). These studies also give the impression that gardens are important only in terms of providing food to the Maya. Rewald (1989) and Greenberg (1992), however, suggest that Maya gardens and shifting cultivation might be part of a larger ecological process that is dynamic-rather than static-and affects both native plants and wildlife. One important ecological process that has been affected by gardens and shifting cultivation is the distribution and abundance of many important plant species in the Yucatan Peninsula. For example, fruits such as huaya (Jalisia olivaeformis; see comment in the Methods section about Maya and Spanish 179 180 terms), sakpaj (Brysonima bucidaefolid), and mamey (Calocarpus mammosum) are edible native plants that thrive in the early successional stages of the forests that regrow in fallowed garden sites (Edwards, 1986; Murphy, 1990; Puleston, 1982). In addition, sapodilla (Manilkara zapota), mahogany (Swietenia macrophylla), and cedar (Cedrela odorata) are valuable timber species that do not regrow quickly and are consequentiy spared from felling by the Maya when gardens are cleared (Rewald, 1989), whereas copal (Protium copal) and ramon (Brosimum alicastrum) may actually have been grown in plantations by the ancient Maya (Roys, 1972). These examples suggest tiiat the ancient Maya had an extensive knowledge of useful plants in the forest and were able to manage these species during several hundred years. By managing these useful plant species in coordination with their gardening practices, Maya gardeners created a habitat mosaic. This mosaic included gardens, fallow gardens, and managed forests of various sizes and many age classes in close proximity to one another. Given the length of Maya occupation of the Yucatan Peninsula (3,000-4,000 y) and the wide area over which shifting cultivation has been practiced by these people (Mexico, Belize, Guatemala, and Honduras), the present forests of the Yucatan Peninsula are considered by many scientists to be highly anthropogenic (Barrera etal., 1977; Edwards, 1986; G6mez-Pompa et al., 1987; Lundell, 1933). In marked contrast to numerous studies on the impact of shifting cultivation on soils and plants (e.g., Noguez-Galvez, 1991; Nye and Greenland, 1960), Uie impact of shifting cultivation on Uie distribution and abundance of wildlife species in the Yucatan Peninsula, while discussed by some researchers, has not been thoroughly investigated. This deficiency is especially glaring given the importance, wide extent, and long history of subsistence hunting by the Maya (cf. Hamblin, 1984, 1985; Hamblin and Rea, 1985; Pohl, 1976, 1985; Reina, 1967; Wing and Steadman, 1980) and evidence that this hunting may be dependent upon shifting cultivation. From a slightiy different perspective, Uiis deficiency in garden studies is also serious because hunting yields by other indigenous people practicing shifting cultivation have been shown to be a function of settiement age (Vickers, 1980, 1988, 1991). 181 At least three studies have examined hunting with respect to gardens and shifting cultivation in Mesoamerica. Based on an archeological analysis of animal bones found in middens at Cerro Brujo, a forested area in northern Panama, Linares (1976) concluded that subsistence hunting by Cerro Brujo hunters was sustainable and conducted in accordance with ecological relationships that were exploited by these people. According to Linares (1976), the interspersion of small gardens in the forest around the settlement enhanced the populations of certain species of wildlife. The local wildlife populations were enhanced because they regularly fed on the crops planted in the gardens. Cerro Brujo hunters in turn exploited the increased wildlife populations. This relationship, called "garden hunting" by Linares (1976), described an association between hunters and prey that depends upon gardens. In garden hunting, the prey animal species, due to the garden, benefit from the additional food resources available, and the Cerro Brujo hunters, also because of the garden, benefit from the additional game available. The beneficial relationship between wildlife and shifting cultivation also has been documented in contemporary studies beyond Panama. For example, the Lacandon Maya in southern Mexico practice garden hunting. Based upon an analysis of plant and wildlife use of gardens, fallowed gardens, and the surrounding forest. Nations and Nigh (1980:17) made two observations relevant to this ch^ter: "the acahual [fallowed garden] attracts many animals of interest to humans," and "the majority of the animals attracted to the acahual cannot live from that area alone; they must also complete critical periods of their life cycle in other environments." These observations reinforce Linares' (1976) hypothesis that the interspersion of gardens in forest has enhanced hunting and wildlife densities and has resulted in a dependent relationship between gardens and the populations of many wildlife species. The present study was designed to test some of Linares' (1976) hypotheses by examining garden hunting in a forested area in southeastern Mexico that is populated by Maya Indians who practice subsistence hunting and shifting cultivation. Specifically, this study proposed to characterize Maya gardens and to quantify their use by wild animals. In designing this study about gardens and wildlife, several factors had to be considered: One, not all adult Maya ejido residents planted gardens 182 every year. While all ejido members normally were required to plant a garden as one of several requirements to maintain their residency status, sometimes this regulation was not enforced (D. Ake Ayala, pers. comm.). Two, gardeners were free to select any site within the study area on which to plant their garden. The single exception to this was a 25,000 ha parcel of land declared as a permanent forest reserve by ejido and state forestry officials (M. Carre6n Mundo, pers. comm.). Three, gardeners were free to determine the crops planted, the amount of land cleared, and the times to plant and harvest. Four, the distribution of roads, trails, fertile soil, and level lands appropriate for planting varied greatly on the study area. Five, gardeners often employed other workers or formed teams with friends or family members to complete garden-related tasks. Six, gardeners often undertook other activities (e.g., hunting, chide taping, and gathering firewood) while they were tending their gardens. These factors required that the study be carefully planned and the analyses carefully thought out. There were three general objectives for this study and several specific hypotheses: First objective. "Determine the total number of gardeners and number of gardens planted in 1989 by X-Hazil Sur residents and compare this with a subset of gardeners and gardens planted in 1990. The specific hypothesis tested was that there was no difference between 1989 and 1990 with respect to the number of gardens planted per "garden group" (see definition, below). Second objective. -Describe the annual horticultural cycle, quantify garden characteristics, and characterize crop phenology, density, and mortality. The specific hypotheses tested were that there was no differences between 1989 and 1990 with respect to the following variables: garden size, distance from X-Hazil Sur, number of crops planted, and number of years of continuous use to which the garden had been subjected. Third objective. -Characterize crop predation and wildlife use of gardens and adjacent forest. Three specific hypotheses were tested: One, there was no difference between wildlife use of track slicks in gardens and the adjacent forest in 1990. Two, there was no difference between wildlife use of Late Secondary Forest without Gardens (see definition, below). Late Secondary Forest with Gardens, and Early Secondary Forest in 1990 with respect to number of wildlife taxa reported and number of 183 track sets recorded. Three, there was no difference among months in the occurrence of wildlife track sets in the track slicks. Methods Study Area The study took place at Ejido X-Hazil y Anexos, Quintana Roo, Mexico, during 1989-1990 (total area = 552.95 km^; Figure 2-1). Gardening data were obtained from residents at the village of X-Hazil Sur (19°23'30"N, 88°05'00"W; population = 1,040), the largest of three villages on the ejido (total population = 1,680). The mean annual temperature is about 26°C and the area typically has one dry season (December-May) and one wet season (June- November). Rainfall during 1 January-12 December 1990 was 1,277.3 mm (Figure 2-4). About 88.52% of the ejido was categorized as Late Secondary Forest, 6.07% as Plots & Gardens, 5.18% as Early Secondary Forest, and 0.23% as Other (Table 2-3). Since about 1915 the area has been occupied by Maya Indians, whose main subsistence activity has been shifting cultivation, primarily com. Prior to 1915 the area was not inhabited by Maya Indians. Gardener Interviews During 1989 a complete census of X-Hazil Sur households was conducted in order to determine the number of residents who planted gardens and the characteristics of those gardens. Only limited information about the annual horticultural cycle was obtained at that time for two reasons; one, the villagers did not yet trust me, and two, the planting season already had passed and gardeners did not remember when specific garden activities were conducted. In 1990, a random sample of those who planted gardens in 1989 was selected and interviewed to enable comparisons about gardening between years. The 1990 interviews were more successful and included additional questions about the annual horticultural cycle that were not asked in 1989 (e.g., dates of clearing and burning). Interviews usually were conducted at the home of the gardener (see Appendix Y for a copy of the data form). A local 184 Maya assistant translated or explained the questions, if necessary, during the 15 min interview. Questions were asked in Spanish or Maya. If all gardens were the same in the sense that each household planted a single garden, then it would be expected that the relationship between all gardens and wildlife would be comparable. However, if some households shared in planting gardens or if outside labor was used, then the relationship between gardens and wildlife might be different. Originally it was thought that each adult man would plant a garden by himself for his immediate family. This would result in a certain duration of work and frequency of visits to the garden site. During the gardener interviews, however, it became apparent that this single pattern was not universally applicable at X-Hazil Sur and that there were various arrangements by which gardens were planted and the crops harvested and distributed. These differences potentially could affect wildlife use of gardens by substantially decreasing the rate at which workers visited the garden. Another reason for not simmiarizing garden activities per household was because nonfamily members often cooperated in garden activities. For example, an adult man could individually have a single garden for the use of his immediate family, or he could share the work and the crop obtained with a son, brother, or other relative. In other cases, two men might cooperate in planting one garden, while individually planting other gardens. These arrangements changed during the growing season and between years as gardeners died, became ill, or took temporary jobs outside of the ejido. As a result, there was a great deal of flexibility regarding who worked in the garden and who shared in the harvest. This complex situation suggested that a new term should be used to described the gardening unit. In this chapter, instead of trying to categorize and quantify the different gardening arrangements, the single term garden group will be used to denote all of the people, except temporary contract employees, who contributed to the production or consumption of garden crops from a particular garden. Garden group is different from the term household in that garden group members may not necessarily be related or eat or sleep in the same housing compound. Garden group is similar 185 to household in that members have a close economic and social relationship. Garden group thus recognizes that gardeners may or may not be related and may or may not live together as a family unit. During this study, garden was used for the Spanish term milpa to identify the plot of land away from the house compound where crops such as com were planted. Milpas are different from solares, which are located within the house compound and include a great variety of important annual and perennial plants (Forrest, 1991). Soil Type of Gardens and Transects Soil type is an important factor in determining the potential crop production of a garden. In addition, soil type will determine, in part, the type of forest that regenerates in a garden placed in fallow. Soil type was determined by a local forester for the seven selected gardens (garden numbers 5- 11) and the eleven associated transects (transect numbers 2-12; see Figure 4-1), based on 0.5 kg samples collected from the track slicks (see description, below). Identification, using Maya soil terminology, was based on soil color, texture, and size of grains. Garden Size. Location, and Distance from X-Hazil Sur Garden size is important in determining the amount of crops potentially available to wildlife. Location is important in determining the distribution and abundance of gardens around the village where gardeners live. Distance from X-Hazil Sur is important to gardeners in considering the trade- offs between time and effort. For example, gardeners who prefer to plant near the village have shorter travel times, but may have increased competition with other gardeners for favorable sites. On the other hand, gardeners who prefer to plant at greater distances from X-Hazil Sur have longer travel times, but may have reduced competition for sites. The determination of garden size was based, in part, on information provided by gardeners during the interviews. The basic unit of measurement reported by gardeners was the mecate. The mecate had a dual function in that it described a linear measurement of 20 m, as well as a square area 186 measuring 20 m on a side (total area = 400 m^. Some gardeners precisely determined this area using a tape measure, while others estimated this area, for example, by counting paces. A precise measurement of size was important to gardeners because they often contracted out certain phases of the work and used the area worked, instead of an hourly salary, as the basis for determining the payment. Garden size for the seven selected gardens used to monitor crop phenology (see below) was calculated from measurements of the garden edges and their bearings. Garden edge lengths were measured with a tape measure to within 1 m. Garden edge bearings were measured with a hand-held compass to within 2 degrees. An outline of the garden was made (scale 1; 1000), and the area enclosed was measured with a compensating polar planimeter. Gardeners reported the distance and location of their gardens during the interview by indicating the route taken, making reference to nearby landmarks, and referring to locations of nearby gardens. The accuracy of distance estimates greatly benefitted due to the occurrence of numbered road markers at intervals of 1 km along Mexican Route 307 (see Figure 2-5). Based on this information, aerial photographs, and my familiarity with the area, garden locations were plotted on a base map (scale 1:100,000). Individual garden distances from X-Hazil Sur were calculated to within 0.1 km using the water tank at the western edge of X-Hazil Sur as the reference point. The water tank was used because it is a permanent structure and is indicated on many published maps. Distance measurements for gardens near X-Hazil Sur or those close to roads or other landmarks probably were accurate to within 0.5 km, while distances for the others probably were accurate to within 1 km. Garden Type and Years of Previous Use The number of years during which a garden is used is important in determining the rates at which new areas are converted to gardens and gardens are placed in fallow. This will have a direct impact on wildlife and their use of the area. If gardens are planted but a single year, the rate of conversion is much greater than if gardens are used 3-4 y in a row. If an area has a relatively low population density of gardeners and a relatively low forest conversion rate, this rate will not be of 187 much importance because the amount of land used by gardeners is much less than potentially available. However, if the area has a relatively high population density of gardeners and a relatively high forest conversion rate, this rate will be important because overused lands eventually lose their fertility and their ability to grow either crops or forests (Noguez-Galvez, 1991; Nye and Greenland, 1960; Stromgaard, 1986, 1991). This will affect both native plants and wildlife, as well as gardeners. Crops often were planted and harvested in gardens for two or more years in a row. Based on information provided by the gardener, gardens were aged (number of years of previous use immediately prior to the present growing season) and classified as "new" (not used as a garden the previous year) or "old" (used as a garden the previous year). Gardens categorized as old underwent the entire annual horticultural cycle and were different from managed fallows at other sites (see Alcorn, 1984a, 1984b) where crops continued to be weeded and harvested during subsequent years without additional planting. Crop Phenologv and Mortality Maya gardeners reported that different species of wildlife use crops at different developmental stages. Accordingly, crop phenology was measured during the 1990 growing season in order to relate specific wildlife use to a particular crop developmental stage. Crop phenology was monitored in circular plots that were 2 m in diameter (3.1 m-) at the seven selected gardens. The number of plots per garden was proportional to garden size (about 6-7 plots/ha). Plots were randomly located within gardens and were surveyed about once a month during 7/90-1/91 (month/year), the period of time from planting through harvest. Given that gardeners planted by hand a mixture of four-six seeds per clump in rows at roughly 1-m intervals, a plot size of 3.1 m^ was selected to assure a sample of three-five seed clumps per plot. Crop density was determined by converting these values to stalks/m^. For each plot survey, the following variables were measured for each crop: number of plants (stalks), height of plants (determined by measuring tape to within 5 cm), developmental stage of plants. 188 and evidence of crop predation. The following developmental stages were used (listed in order of maturity): seedling, stalk with flowers, stalk with irmnature fruits, stalk with mature fruits, and harvested. The harvest dates for the seven selected gardens complemented the information about the annual horticultural cycle obtained through interviews in 1990 because it was not possible to re- interview all 31 gardeners. Crop predation (species of crop predator and type of crop damage) was recorded in order to demonstrate specific use of a crop by wildlife. Although phenology data were collected for all crops, only the results for com, squash, and kidney beans are presented here. Several criteria were used in choosing the seven selected gardens: One, the gardener had to agree to the selection. Two, the garden had to be reasonably close to a road or trail in order to facilitate access by project personnel to measure the plots. Three, the vegetation either to the west or to the east of the garden had to be fairly uniform and free of other gardens in order to accommodate a 2000-m-long wildlife census transect (see Chapter 4 for additional information). Crop mortality was determined during 7/90-10/90 (the growing season in 1990) by counting the number of stalks per plot during the several readings and subtracting the last stalk count from the greatest stalk count for each garden. These values were averaged for the seven gardens. This was necessary as the first reading did not necessarily have the greatest number of stalks per plot. Seeds germinating after 10/90 were ignored in calculating crop mortality. Wildlife Use of Gardens Maya gardeners reported that different species of wildlife use gardens at different rates and times of the growing season. Accordingly, wildlife use of gardens was measured during 7/90-1/91 in order to compare seasonal use by month and differences in frequency of garden use among different forest successional stages. Wildlife use of gardens was measured in two ways: (1) tracks and feeding signs observed in the circular plots used to monitor crop phenology (see description, above) and (2) track slicks along gardens or along transects in forests adjacent to gardens. The track slicks were 1 m by 20 m in size 189 and were specially located at sites where wild animals were likely to pass. Track slicks were prepared by spading the soil, removing all rocks and roots, and raking the soil to produce a smooth surface. The slicks were surveyed about once a month during 7/90-1/91, usually on the same date as crop phenology was monitored in the corresponding garden. For each track set observed, species and comments about the soil humidity of the slick as it related to clarity and duration of track sets were recorded. Track slicks were maintained, as necessary, by supplemental raking and weeding, but the area around the track slick was not disturbed. Track slicks were surveyed by two local Maya men who also were experienced hunters. Tracks were identified on the basis of their size and shape, as well as associated hairs, feathers, feces, game trails, or feeding sign. When unknown tracks were encountered, the track readers consulted with other knowledgeable hunters at X-Hazil Sur to determine the species that made the tracks. Track slicks were located along the edges of the seven selected gardens (one track slick per garden) and were used to measure wildlife use of gardens. Seven additional track slicks were located in the adjacent forest along the transects extending from the seven selected gardens (one transect per garden and one track slick per transect). The purpose of these track slicks was to determine if there were any differences between wildlife use of gardens and the adjacent forest. Transects were about 2000 m in length and were used during wildlife censuses to determine wildlife densities (see Chapter 4). A single track slick was located on each transect at about 40-220 m from the edge of the garden for comparative purposes. Track slicks along transects were prepared and read in the same manner as the garden track slicks. Four additional track slicks were located in forests on four transects (transects number 2, 3, 4, and 12) that did not extend from gardens. These track slicks were prepared and read as described above, and were used as a control to compare wildlife use of gardens in Early Secondary Forest versus Late Secondary Forest with Gardens. Wildlife species were identified on the basis of descriptions provided by gardeners, consultation with reference books (Leopold, 1977; Peterson and Chalif, 1973), and by comparison with 190 wildlife shot by hunters (see Chapter 3 for additional information). Voucher specimens were identified by Mexican biologists and deposited at the Museo de Vertebrados, Centre de Investigaciones de Quintana Roo, Chetumal, Quintana Roo, Mexico. Statistical Methods The results presented in this chapter were based on three data sets: One, during 1989 and 1990, gardeners were interviewed to determine number of gardeners, number of gardens, annual horticultural cycle, years of previous use, garden size, garden distance from X-Hazil Sur, number of crops planted, number of crops per garden, and reported crop predation. Two, during 1990, seven gardens were selected for an intensive study of crop phenology, density and mortality of crops, and actual crop predation. Three, during 1990, track slicks at 18 sites were used to compare wildlife use of gardens versus use of adjacent forest, wildlife use of the three vegetation types, and monthly variation in wildlife use of the three vegetation types. Parametric and nonparametric tests were used to analyze garden and wildlife data. For the chi-square comparison (x^) of garden aops between years, three crops (chaya, lentils, and other) were combined to obtain an expected cell frequency S 5. The nonparametric Kruskal-Wallis test (x^ approximation) was used for variables where the observations likely were not independent or normally distributed. For example, several gardeners reported that they always planted their fields near those of other family members. Several gardeners also reported that they always planted a single crop of com in large gardens and mixed crops in smaller gardens. The results are presented as x^ = chi-square; x^ approximation = chi-square approximation; J = mean; SE = standard error; n = sample size; d.f. = degrees of freedom; and P = significance level. All statistical analyses were conducted using SAS (SAS Institute Inc., 1988). 191 Results Gardeners Number of gardeners. Data for this part of the study were obtained during 6/89-1/91 (month/year). In 1989, 180 interviews of widows and adult men were conducted during 6/29-8/22 (month/date) to determine the extent to which X-Hazil Sur residents planted gardens. A total of 123 residents (2 women and 121 men, representing about 60% of the total number of households at X-Hazil Sur) reported having planted a total of 150 gardens in 1989; 14 residents (7% households) reported having cleared a total of 17 gardens, but had not planted them by the date when the interview was conducted; and 43 residents (20% households) reported that they had not cleared or planted a garden that year. In 1990, 31 randomly selected gardeners that had planted in 1989 were interviewed during 11/22-11/24 and reported having planted a total of 40 gardens, while another 5 residents who also planted in 1989 reported that they had not planted in 1990. Thus, about 80-85% of the ejidatarios planted gardens. Number of gardens per garden group. Garden groups frequently planted more than one garden per group. In 1989, each group planted an average of 1.2 gardens (SE = 0.05, n = 123 garden groups). One garden group planted 4 gardens, 3 groups planted 3 gardens each, 18 groups planted 2 gardens each, and 101 groups planted 1 garden each. In 1990, each garden group planted an average of 1.3 gardens (SE = 0.11, n = 31 garden groups). Two groups planted 3 gardens each, 5 groups planted 2 gardens each, and 24 groups planted 1 garden each. The number of gardens planted per garden group was not significantly different between years (x^ approximation = 0.4094, d.f. = 1, P = 0.5223). Gardens Annual horticultural cycle. The annual horticultural cycle consisted of six main periods: initial site clearing (felling trees and brush), burning, planting, weeding, doubling over of com stalks, and 192 harvesting (Figure 6-1). These activities frequently overlapped in time and collectively occurred throughout the year. Initial site clearing occurred primarily during January-April and consisted of using an axe or machete to cut down brush and small trees (Figure 6-1). Large trees usually were not cut down, while medium-sized trees usually were cut at waist height (A. Tuz Novelo, pers. comm.). The brush and small trees were left to dry at the site. Clearing occurred over several weeks. Some gardeners cleared sites as late as April-June. Gardens individually were burned on a single day-usually in an aftemoon-and collectively were burned during the month of April (Figure 6-1). Gardeners carefully timed the blaze to maximize bum completeness and ash production, thus facilitating planting and increasing soil fertility. If the bum took place too early in the season, gardeners indicated, the brush and trees would resprout and the ash would blow away with the wind. If the bum took place after the rains had begun, the blaze likely would be incomplete and little ash would be available to fertilize the garden. Gardeners rarely attempted to rebum areas that did not bum initially (M. Cab Cohuo, pers. comm.). The quality of a bum was extremely important as an incomplete or improperly timed bum would reduce crop yields and make planting more difficult. There were two planting periods: an intense period during May-July and another period during the rest of the year when seeds were sown at irregular intervals (Figure 6-1). During the fu-st period, squash (Cucurbita pepo), sweet potato (Ipomoea batatas), bean (Phaseolus spp.), lentil (Lens esculeiua), yam {Dioscorea sp.), and com {Zea mays) were the main crops planted. During the second period, jlcama (Pachyrrhizus erosus) was the main crop planted, while some macal and ibes (Phaseolus lunatus) seeds also were planted at this time. The main planting usually took about 1-2 weeks per garden and began after heavy rains had thoroughly moistened the soil (F. Balam Can, pers. comm.). Usually a mixture of seeds was sown throughout the garden. Often, however, seeds such asji'cama, lentils, and chili (Capsicum annuum) were specifically planted in a small part of the garden that had better soil and was better tended than the rest of the site. 193 GARDEN ACTIVITY: Clear Bum Plant Weed Double com Harvest WEATHER: Temperature Rainfall Max Min Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 6-1. Annual horticultural cycle at Ejido X-Hazil y Anexos during 1989-1990 (Max = month with highest mean temperature and Min = month with lowest mean temperature). Weeding took place during July-September (Figure 6-1). Most gardeners weeded once or twice, as needed and as time permitted (2-4 weeks per garden per weeding; R. Cab Cohuo, pers. comm.). Gardeners either pulled weeds by hand or cut them with a machete or hooked knife. Pesticides and herbicides generally were not used in gardens but were used on plots in the fruit corridor east of X-Hazil Sur, along Highway 307 (E. Che Canul, pers. comm.). Com plants were doubled over during September-October in order to facilitate drying and to protect the cobs from crop predators such as parrots (Family Psittacidae) and coatis (Nasua nasua; see below; Figure 6-1). Other garden crops were not doubled over. Com plants were harvested during November-January (Figure 6-1). Other crops were harvested as they became ripe or were needed by the gardener. For example, yams, squash, and sweet potatoes did not spoil quickly and could be harvested up to 6-12 months after planting (M. Cab Ake, pers. comm.). The 1989 annual horticultural cycle, according to local residents, was highly unusual in that rains were intermittent, late to begin, and less substantial than normal (see Figure 2-4). As a result. 194 many residents delayed planting, replanted, or abandoned their gardens. Some garden crops dried up, while weeds grew faster than did crops in other gardens. Although harvests were not measured, gardeners reported that only about 10-20% of the gardens produced an average crop in 1989, while the other gardens produced below-average yields. Due to the irregular nature of the rains in 1989, data for the timing of horticultural activities that year were limited to the date of first planting. The median first date of planting was 5/12-5/13 (n = 150 gardens). The earliest reported planting was on 4/1, while the latest first planting was on 8/7. At least seven gardens were completely replanted once and two were completely replanted twice. Other gardens were partially replanted as large numbers of seedlings died due to the irregular and inadequate rains. The 1990 annual horticultural cycle, according to residents, was about normal and most gardeners achieved an average harvest. The median date to begin initial site clearing was 2/15 (range = 9/15/89-6/1/90, n = 40 gardens; month/date/year). The median date to bum the garden site was 74 days later, on 4/30 (range = 3/1-6/30, n = 40 gardens). The median date to plant was 32 days later, on 6/1-6/2 (range = 4/30-7/15, n = 40 gardens). Weeding and doubling took place during 8/90- 10/90. The com harvest began about 10/30 and was complete by about 12/8 (n = 7 gardens). Garden type and years of previous use. Most gardens were planted on sites that had not been used as gardens the previous year. In 1989, the average number of years that a site had been used previously as a garden was 0.1 (SE = 0.05, n = 150 gardens, range = 0-5 y). About 91% of the 150 gardens were on sites categorized as new, while 9% were on sites categorized as old. In 1990. the average number of years that a site had been used previously as a garden was 0.0 (SE = 0.02, n = 40 gardens, range = 0-1 y). About 98% of the 40 gardens were on sites categorized as new, while 2% were on sites categorized as old. The number of years of previous use for gardens was not significantly different between years (x' approximation = 1.7769, d.f. = 1, P = 0.1825). One gardener indicated that he had cleared a field in Late Secondary Forest in 1989. All other gardeners reported that they had cleared fields in Early Secondary Forest. The importance of this is 195 that ejido and state forestry rules forbid clearing gardens in Late Secondary Forest (D. Ake Ayala, pers. comm.). Size. Gardens ranged in size from relatively small (minimum = 0.3 ha) to quite large (maximum = 18.0 ha), but 70% were 1.0-2.9 ha in extent in 1989, while 62% were that size in 1990 (Figure 6-2). In 1989, the average garden size was 2.1 ha (SE = 0.11, n = 150 gardens, range = 0.3-8.0 ha, total area = 310.84 ha). In 1990, the average garden size was 2.7 ha (SE = 0.45, n = 40 gardens, range = 0.5-18.0 ha). Garden size was not significantly different between years (x^ approximation = 1.4455, d.f. = 1, P = 0.2293). Based on the 1989 information for X-Hazil Sur, the total area for the ejido in gardens would be about 545 ha (ca. 1% of the total area available; about 346 households times 0.75 gardens per household times 2.1 ha per garden). 0.0-0.9 1,0-1.9 2.0-2.9 3.0-3.9 GARDEN SEE (HA) 4.0-4.9 =>5.0 1989 ^1990 Figure 6-2. Distribution of gardens by size during 1989 (n = 150 gardens, average size = 2.1 ha) and 1990 (n = 40 gardens, average size = 2.7 ha). 196 Soil type of gardens. Soil type was determined for the seven selected gardens and the 11 associated transects. Two types were determined (Mr. P. Macario Mendoza, in litt.); K'ankab and Yaax'hom. K'ankab soil, the most common, was identified for six of the seven selected gardens (numbers 5, 6, 7, 8, 9, and 11). This soil is red due to the concentration of iron oxides and the leaching out of black organic matter. K'ankab soil is relatively poor in phosphorus and nitrogen and is subject to laterization when bared to the sun. This soil is slightly acidic (pH 6.5-7.5) and has 10-15% organic matter and 26% clay. Yaax'hom soil was identified only for garden number 10 (see Figure 4-1 for location of gardens and transects). This soil is black. Yaax'hom soil also is slightly acidic (pH 6.5-7.5) and has 5-20% organic matter and 62% clay (soil description from Snook, 1993). Distance from X-Hazil Sur. Gardens were located at distances that ranged from relatively close to X-Hazil Sur to quite far away (minimum distance = 0.6 km, maximum distance = 16.6 km), but about 50% were at 1.0-2.9 km (Figure 6-3). In 1989, the average garden distance from X-Hazil Sur was 4.0 km iSE = 0.29, n = 150 gardens, range = 0.7-16.6 km). In 1990, the average garden distance was 3.8 km (SE = 0.50, n = 40 gardens, range = 0.6-15.9 km). Garden distance was not significantly different between years (x^ approximation = 0.1730, d.f. = 1, P = 0.6774). Several gardeners indicated that competition with other gardeners had forced them to sow gardens at alternate sites on the study area as the intended site already had been taken. One common tactic used by gardeners to obtain additional land was to plant a garden adjacent to that of another gardener. Usually the adjacent gardener abandoned the vicinity after the harvest season. Another tactic was to clear a narrow strip of land around a potential garden site, in effect laying claim to the area, but not actually clearing or planting the site until a few years later, when hopefully any neighboring gardeners had left the area. Crops planted. Gardeners reported planting a wide variety of crops in their gardens. In 1989, gardeners reported planting 16 types of crops, while in 1990, 13 were reported (Figure 6-4). Three crops were more frequently planted than other crops: com (99% of gardens in 1989 [n = 150 gardens] and 100% in 1990 [n = 40 gardens]), squash (84% and 63%), and kidney beans (75% 197 3 4 5 6 7 GARDEN DISTANCE (KM) 9 => 10 1989 1990 Figure 6-3. Distribution of gardens by distance from X-Hazil Sitr during 1989 (n = 150 gardens, average distance = 4.0 km) and 1990 (n = 40 gardens, average distance = 3.8 km). and 48%; Figure 6-4). The frequency with which crops were planted in gardens was not significantly different between years (x^ = 11.510, d.f. = 10, P = 0.319). Gardens typically contained several crops. In 1989, the average number of crops planted per garden was 5.9 (SE = 0.25, n = 150 gardens, range = 1-12 crops per garden; Figure 6-5). In 1990, the average number was 3.9 (SE = 0.47, n = 40 gardens, range = 1-11 crops per garden). The number of crops planted per garden was significantly different between years (x' approximation = 13.773, d.f. = 1, P = 0.0002). Crop phenologv. Although planted at the same time, seeds in a particular garden may not germinate at the same time or develop at the same rate. This may affect wildlife use of gardens if wild animals select for or against certain plant species or plant parts (e.g., com cob or sweet potato tuber). 198 KB I M SP CHI F J CHA L GARDEN CROPS 1989 ^ 1990 Figure 6-4. Frequency with which crops were planted in gardens during 1989 and 1990 (C = com, S = squash, KB = kidney beans, I = ibes, M = yam, SP = sweet potato, CHI = chile, F = fruits, Y = yucca, J = jicama, CHA = chaya, L = lentils, and O = other). Crop phenology was studied intensively on the seven selected gardens in 1990. Six of the seven selected gardens were planted during 6/13-6/28, while one was planted about 7/25. The seven selected gardens were surveyed eight times each during 7/90-1/91. The number of reported observations per gardens, however, was less than the number of garden surveys as some gardens were planted later or harvested earlier than others. Gardeners reported that the key factor in deciding when to plant was sufficient rainfall. In 1990, relatively heavy rainfall occurred at the beginning of the planung season on 4/3 (11.2 mm), 4/12 (13.5 mm), 4/27 (40.0 mm), 6/7 (16.0 mm), and 6/8 (20.5 mm; see Appendix A). Many gardens were planted soon after each of these dates. Com, the most abundant crop according to the density of stalks (see below), was present in the gardens for about six months in 1990. During 7/90-8/90, com plants primarily were at the seedling stage (Figure 6-6). In 9/90, about 37% of the plants were seedlings, 29% had flowers, and 34% had 199 4 5 6 7 8 9 NO. CROPS PEK GARDEN 10 11 12 1989 ^ 1990 Figure 6-5. Number of crops per garden during 1989 (n = 150 gardens, average = 5.9 crops per garden) and 1990 (n = 40 gardens, average = 3.9 crops per garden). immature fruits (cobs). During 10/90, about 25% of the com plants had immature cobs, 50% had mature cobs, and 23% had been harvested. In 11/90, 68% of the plants had mature cobs and 32% had been harvested. By the end of 12/90, all of the com had been harvested. Kidney beans, the second most abundant crop, were present in the gardens for about six months in 1990. During 7/90-10/90, kidney bean plants were at the seedling stage (Figure 6-7). In 11/90, about 52% of the plants had flowers, while 39% had been harvested. During 12/90, about 26% of the kidney bean plants were seedlings, about 24% had flowers, and 50% had been harvested. By 1/91, all of the kidney beans had been harvested. Squash, the third most abundant CTop, was present in the gardens for about four months in 1990. During 7/90-9/90, squash plants primarily were at the seedling stage, but 4-6% of the plants had mature fruits during 8/90-9/90 (Figure 6-8). In 11/90, 100% of the plants were harvested, but in 12/90 200 100 90 a < 70 w "^ 50 I ^^ § 30 ^ 20 10 0 zszxzxzszz 7/90 8/90 9/90 10/90 11/90 12/90 1/91 MONTH SEEDLING MATFRUTK W/FIflWERS HARVESTED IMMATFRUrre Figure 6-6. Monthly phenology of com plants in seven selected gardens during 1990 (n = 7 gardens). Plants were categorized as seedlings, with flowers, with immature fruits, with mature fruits, and harvested. additional squash germinated and 17% had mature fruits seedlings. By 1/91 all remaining squash were harvested. Density and mortality of crops. The density of seedlings varied greatly among the seven selected gardens. Garden number 6 had the least number of stalks at 45.9 stalks/m^ (n = 6 surveys), while garden number 10 had the greatest number at 179.0 stalks/m^ (n = 5 surveys; Table 6-1). The average plot had 92.5 stalks/m' and was composed primarily of com (87% of stalks), kidney beans (5%), and squash (3%). Not all of the crop seeds that germinated in the seven selected gardens developed and produced mature fruits that could be harvested. Among the three major crops, squash seedlings experienced the greatest mortality with an average loss of 82% (SE = 4.26, n = 6 gardens), while kidney beans experienced an average loss of 62% (SE = 7.02, n = 7 gardens), and com experienced an average 201 100 90- 80 a o < 70H ^ 50- C/] E 30 ^ 20 10 0 IAAAaAI 7/90 8/90 9/90 10/90 11/90 12/90 1/91 MONTH SEEDLING ^ ff/ FLOWERS m IMMAT FRUTre MAT FRUnS ^3 HARVESTED Figure 6-7. Monthly phenology of kidney beans at seven selected gardens during 1990 (n = 7). See Figure 6-6 for key to abbreviations. loss of 21% (SE = 4.42, n = 7 gardens). Wildlife Use of Gardens and Adjacent Forest Crop predation. Wild animals can use gardens as sites to obtain food or locations where social interactions may take place. This use may be dependent upon either the wildlife species or the developmental stage of the crops in the gardens or both. Fourteen taxa of wild animals were identified by gardeners as the main crop predators. In 1989, wildlife identified as "birds" were the main crop predator at 50% of the gardens (n = 150 gardens; Figure 6-9). Gardeners generally did not identify a specific avian taxon as the main crop predator, but instead usually mentioned a combination of parrots (Psittacidae), jays (Corvidae), 202 100 90 o ^ 50 I 40 S 30 ^° 20- 10- 0 ^ ^. 7/90 8/90 9/90 10/90 11/90 12/90 1/91 MONTH I SEEDUNG ^ W/ FLOWEES MAT FRUnS ^3 HARVESTED IMMAT FRUITS Figure 6-8. Monthly phenology of squash as seven selected gardens during 1990 (n = 7 gardens). See Figure 6-6 for key to abbreviations. blackbirds (Icteridae), and pigeons and doves (Columbidae; see Appendix Z for a list of conunon and scientific names of crop predators). The coati was the main mammalian crop predator at 13% of the gardens in 1989, while crop predation was not a problem at 11% of the gardens. In 1990, the coati was the main crop predator (30% of gardens, n = 40 gardens), but the collared peccary (Tayassu tajacu; 23%), birds (20%), and taxa summarized as "other" (15%) also were reported by gardeners. Actual crop predation in crop phenology study plots was observed on two occasions. On 10/30/90, paca (Agouti paca) tracks and feeding sign on a com plant were observed in garden eight, and on 1/18/91, agouti (Dasyprocta punctata) tracks and feeding sign on a camote plant (tuber) were observed in garden seven. White-tailed deer (Odocoileus virginianus) tracks were observed on 9/10/90 in garden eight, but no crop predation was noted. Other examples of crop predation in gardens were noted, but were not included here as they did not specifically occur on the study plots. 203 Table 6-1. Density of crops (mean number of stalks/m^) planted in the seven selected gardens in 1990. For crop density, the plots were surveyed during 7/90-11/90, beginning before planting and ending when the com harvest began in the garden. Additional surveys were conducted through 1/91 to monitor crop phenology. Garden number and soil type (Mean number of stalks/m^) Mean number of stalks (n = 7 gardens) Crop" 5' 6^ T 8" 9b 10' 11" SE Com 89.8 36.8 43.1 108.4 57.8 156.5 74.0 80.9 15.82 Kidney bean 7.6 1.7 2.1 9.2 0.0 6.3 6.5 4.8 1.31 Squash 0.0 0.4 1.4 5.8 2.4 6.5 3.6 2.9 0.96 Other 0.0 7.0* 4.1« 0.6^ 0.0 2.5« 0.0 2.0 1.02 Sweet potato 0.0 0.0 5.9 0.7 0.0 1.7 3.1 1.6 0.83 Lentil 0.0 0.0 0.0 0.0 0.0 5.4 0.0 0.8 0.77 Yam 0.0 0.0 2.8 0.0 0.0 0.0 0.0 0.4 0.40 Ji'cama 0.0 0.0 1.0 0.0 0.0 0.0 1.3 0.3 0.21 Mean total 97.3 45.9 51.7 124.8 60.2 179.0 88.4 92.5 17.90 No. surveys 5 6 5 6 6 5 5 5.4 Garden size (ha) 1.2 0.6 0.6 1.2 0.6 2.5 1.2 1.1 No. plots 8 5 5 8 5 12 8 7.3 " Includes only crops planted '' K'aiikab soil type. ' Yaax'hom soil type. '' Zacate. ' Watermelon & sugar cane. ' Banana. ^ Lee {Lagenaria siceraria). in seven selected gardens. 204 50- 45- 40- 35- 25- 20- 15- 10- 5-1 0- O &? BIRDS C»An NONE IWDCDON C PBOCjlRr ITT DEER "GUSANDS' PRINCIPLE CROP PREDATOR OTHEK 1989 ^ 1990 Figure 6-9. Frequency with which wild animals were identified by gardeners as the main crop predators during 1989 and 1990 (none = no crop predators, c peccary = collared peccary, wt deer = white-tailed deer, gusanos = various taxa of Gastropods). Wildlife tracks. Wildlife track surveys were used as a tool to complement measurements of crop predation because wild animals could either have entered a garden in order to consume a cultivated plant or to conduct some other activity. Track surveys also complemented the wildlife censuses (see Chapter 4) in determining which species occurred in an area and the frequency with which specific areas of habitat were used by wildlife. A survey of track slicks suggested that there were differences in wildlife use patterns by month, vegetation type, and between gardens and adjacent forest. A total of 96 tracks sets were recorded from 18 track slicks during 7/90-1/91 (Table 6-2). These tracks were made by seven species of game mammals, two species of game birds, and eight species of nongame mammals (see Chapter 3 for a discussion of game and nongame species). Dogs, as 205 Table 6-2. Frequency of occurrence of wildlife track sets in 18 track slicks (1 m by 20 m) along gardens and transects in adjacent forest during 1990. Species' Number of track sets by vegetation types and track slick location Late Secondary Forest without Gardens Late Secondary Forest with Gardens Early Secondary Forest Tran.'' Gar.' Tran. Total Gar. Tran. Total Grand total a) Mammals Opossum Armadillo Pocket gopher* Agouti* Paca* Grey fox CoaU* Tayra Skunk Puma/Cougar Unknown felid Collared peccary* White-tailed deer* Brocket deer* Unknown mammal 3 2 3 4 1 1 7 13 2 2 1 15 13 2 2 2 3 3 1 2 1 15 1 5 3 3 1 2 1 1 2 1 1 32 1 21 7 5 1 1 3 1 1 2 7 4 5 Total mammals 16 28 10 38 21 17 38 92 b) Birds Plain chachalaca* Great curassow* Total birds 1 1 2 2 4 Grand total 16 29 11 40 21 19 40 96 Number taxa 7 8 4 9 8 9 14 17 Number readings'* 48 38 49 135 ' * = Game species. '' Tran. = Track slick along transect. ' Gar. = Track slick along garden. ^ Number of times the track slicks were surveyed for animal tracks. 206 well as humans left tracks on the slicks, but their tracks were not counted. Armadillo (Dasypus novemcinctus; 33% of total) and agouti (22%) tracks were most frequently observed among the 17 taxa reported. The track sets observed in garden and transect track slicks represented wildlife use over a relatively short period of time. Although not specifically measured during this study, the duration of wildlife tracks was estimated by repeated observations of tracks during other phases of the project. Tracks made by mammals in relatively dry soil usually lasted about 1-3 days, while those made in relatively moist soil usually lasted about 5-7 days. Bird tracks usually remained visible for only 1-2 days. Wind, rain, soil depth, soil moisture content, extent of canopy coverage, and amount of ground litter, however, affected the duration of all track sets. Use of gardens versus adjacent forest. Although small sample sizes precluded statistical analyses, at seven sites (three sites in Late Secondary Forest with Gardens and four sites in Early Secondary Forest) it was possible to compare wildlife use of gardens and the adjacent forest simultaneously in a desaiptive manner. For these two vegetation types, the number of track sets observed in garden u^ack slicks was greater than the number of track sets observed in transect track slicks (29 track sets versus 11 track sets in 38 readings for Late Secondary Forest with Gardens and 21 track sets versus 19 track sets in 49 readings for Early Secondary Forest; Table 6-2). Use patterns of garden and transect track slicks by wild animals were not tested statistically, but differed numerically for the two most frequently reported species (Table 6-2). For the armadillo, the number of track sets in garden track slicks was about equal to the number in transect track slicks in both Late Secondary Forest with Gardens (7 track sets versus 8, respectively, for 38 readings) and Early Secondary Forest (7 track sets versus 8, respectively, for 49 readings). For the agouti, however, the number of track sets in garden track slicks was about equal to the number in transect track slicks in Early Secondary Forest (2 track sets versus 3, respectively, for 49 readings), but in Late Secondary Forest with Gardens, 13 u-ack sets were recorded for garden track slicks while 0 track sets were recorded for transect track slicks (38 readings). 207 Variation in wildlife use by vegetation type. Based on a descriptiye analysis of the data, there were differences in the number and kinds of track sets by vegetation type. Among the three vegetation types, track slicks in Late Secondary Forest without Gardens had fewer taxa (7 taxa) and fewer track sets (16 track sets per 48 readings), while track slicks in Early Secondary Forest had the greatest number of taxa (14 taxa per 49 readings, including 7 game taxa; Table 6-2). The number of track sets in Late Secondary Forest with Gardens and Early Secondary Forest were about equal at 40 per 38 readings and 40 per 49 readings, respectively. Use patterns of track slicks in Late Secondary Forest without Gardens, Late Secondary Forest with Gardens, and Early Secondary Forest differed numerically for the two most abundant species (Table 6-2). For the armadillo, the number of track sets in Late Secondary Forest with Gardens and Early Secondary Forest track slicks was about equal at 15 track sets per 38 readings and 15 track sets per 49 readings, respectively, while relatively few track sets were recorded for Late Secondary Forest without Gardens (2 track sets per 48 readings). For the agouti, however, the nimiber of track sets in Late Secondary Forest with Gardens was greatest at 13 track sets per 38 readings, while relatively few track sets were recorded for Late Secondary Forest without Gardens (3 track sets per 48 readings) and Early Secondary Forest (5 track sets per 49 readings). Monthly variations in wildlife use of gardens and adjacent forest. Wildlife species exhibited monthly differences in the frequency with which they used gardens and adjacent forests. Overall, 93% of the track sets were observed in the wet season during 7/90-10/90 (Figure 6-10; data from Table 6- 2). For Late Secondary Forest with Gardens, the peak of use was in the wet season during 7/90-9/90 (75% of track sets; n = 40 track sets during 38 readings), while peak usage in the other two vegetation types was noted over shorter periods of time (81% of the track sets for Late Secondary Forest without Gardens [n = 16 track sets during 48 readings] were observed during 8/90-9/90 and 68% of the track sets for Early Secondary Forest [n = 40 track sets during 49 readings] were observed in the wet season during 7/90-8/90). Only five track sets were observed during 11/90-1/91 as the wet season ended and the dry season began. 208 t4 11 H 10 f 4 A) Na •nvoK wra: lsf w/o a It 14 IS 10 9 4 t 0 m Na TRAOK SET*: t«F W/ 9 7/90 VOQ t/M ia/«0 WM t2/M VM MONTH 7/M a/M t/to n/M 11/tO 11/tO Vti MONTH 10 14 It to 4i 1 0 C) Na TKACK lETa: EOF r/00 o/M o/ao «/w vno s/«o vm MONTH Figure 6-10. Monthly frequency of wildlife track sets in a) Late Secondary Forest without Gardens (LSF W/O G; 16 track sets during 48 readings), b) Late Secondary Forest with Gardens (LSF W/ G; 40 track sets during 38 readings), and c) Early Secondary Forest (ESF; 40 track sets during 49 readings). 209 Discussion From the perspective of this study, gardens are important for two reasons: One, gardens provide a source of food to game animals. In Chapter 5, the frequency and abundance of garden crops in the stomach contents of game animals were quantified. Those results will be discussed below with respect to the information presented in this chapter. Two, gardens provide a place where many game animals are taken by Maya hunters. In Chapter 3, the number and kinds of game hunted in gardens were presented. Those results also will be discussed below with respect to the garden information presented here. First, however, the specific results about gardens will be discussed. Gardeners To the Maya, growing a garden is an ethnic identifier, separating them from all other people (Bums, 1983). While some people consider com farming a peasant occupation, to the Maya, planting a garden is a reaffirmation of what it means to be Maya. The results of this study indicate that in small, rural villages, most adult Maya men still cultivate gardens, but the practice and cultural importance of gardening by the Maya may be decreasing. During the early 1900s, prior to the modem era in the Yucatan Peninsula, essentially every able- bodied, adult male planted a garden (Redfield and Villa Rojas, 1962; Villa Rojas, 1987). Now, however, the proportion of people planting gardens is less than before. For example. Murphy (1990) reported that at Seiior, Quintana Roo (a small Maya village about 50 km north of X-Hazil Sur), only about 83% of the adult men planted gardens. At X-Hazil Sur, about 80-85% of the adult men and widows, called ejidatarios, planted gardens. This decrease from 100% in the proportion of people planting gardens during modem times is consistent with a general decline in planting throughout the state of Quintana Roo as the Maya diversify their repertoire of subsistence activities to include other economic enterprises, such as timber harvest and the growing of citrus fruits for commercial sale (Sullivan, 1987). 210 While these results indicate a reduction in the proportion of people planting gardens, planting gardens continues to be an important cultural activity. The results indicate that Maya gardeners, as they have done for more than 100 y, still frequently work together in teams composed of friends or family members (Murphy, 1990; Villa Rojas, 1987) rather than individually plant a garden. Working together has many benefits because the various activities associated with planting are difficult, time consuming, labor intensive, and expose the gardener to the hot sun and high temperatures of the area. By working together, for example, gardeners finish the work sooner. In addition, if one of the gardeners becomes ill or decides to work elsewhere temporarily, the other partner will be able to continue caring for the garden. Thus, working together enhances the chances that the garden will be harvested in spite of unexpected events that could keep a Maya gardener from tending his crops. The results for X-Hazil Sur also suggest that the Maya gardeners may be changing the number of gardens they plant per household per year. According to other studies in the Yucatan Peninsula, the Maya typically plant two gardens per household per year (Murphy, 1990; Villa Rojas, 1987). The results for X-Hazil Sur indicated that gardeners planted 1.2-1.3 gardens per garden group per year. While a garden group and a household are different units of measurement, they are roughly equivalent for this type of comparison. This results suggest that Maya gardeners are tending fewer fields than as recently as 60 y ago. Gardens Annual horticultural cvcle. Gardening, as practiced at X-Hazil Sur, is similar to gardening practiced by Maya elsewhere in the Yucatan Peninsula (cf.. Murphy, 1990; Redfield and Villa Rojas, 1962; Smith and Cameron, 1977; Villa Rojas, 1987). This is to be expected given the similar soils, topography, and climate of the region (Mosely and Terry, 1980). The single most important factor affecting the annual horticultural cycle is rainfall. In order for gardens to be successful, heavy rains must begin in May or June, when gardens should be planted, and cease in November or December, when com should be harvested. The 1989 growing season was 211 largely a failure at X-Hazil Sur due to rains that were late in starting and less substantial than average (Figure 2-4). As a result, plantings often were delayed or were repeated as fanners tried to adjust to the unexpected conditions and still salvage a harvest. In the end, large numbers of gardeners simply abandoned their fields and the meager crops. The impact of the hot, dry conditions in 1989 on the wildlife is unclear as several factors may have affected wildlife populations. For example, fewer workers in the gardens and unharvested crops abandoned in the gardens may have favored wildlife by reducing the number of potential hunters and increasing the amount of food potentially available to wild animals. On the other hand, a decrease in the amount of natural foods in the forest due to the dry conditions and an increase in the amount of hunting by hunters seeking to compensate for the failed gardens, for example, may have had negative impacts on wildlife populations. The 1989 crop failure, according to X-Hazil Sur residents, was attributable to Hurricane Gilbert, the largest and strongest hurricane ever recorded by modem meteorological instruments (Anon. 1988; Wilder, 1988). This hurricane swept across the Yucatan Peninsula on 14-15 September 1988, making landfall about 175 km north of X-Hazil Sur. Strong winds and high seas caused major damage to both coastal and inland areas as the storm moved west across the peninsula. The following year, the Yucatan Peninsula suffered a serious drought. Local residents reported that extended dry seasons usually occurred after previous hurricanes in the area, but had no explanation for the phenomenon. Given the geographical position of the Yucatan Peninsula and the relatively high frequency of hurricanes and other tropical storms in the area, crop failures of this nature, such as the one after Hurricane Gilbert, probably occur every 10-20 y. The 1990 growing season was considered about average by Maya gardeners at X-Hazil Sur. Heavy rains began in June during the customary period and continued through the growing season. As a result, crops were planted and weeded during the usual time periods. The rains lessened in December (Figure 2-4) and gardeners reported that an average com crop was harvested. 212 Garden type and years of previous use. Most gardens in the Yucatan Peninsula legally can be cleared either from Late Secondary Forest or Early Secondary Forest. At Ejido X-Hazil y Anexos, gardens can not be planted in the permanent forest reserve established by the Plan Estatal Forestal. Given a choice, gardeners prefer to clear sites in Late Secondary Forest because the work is easier. Gardens also vary in the number of years that they can be planted consecutively and still produce an acceptable crop. While gardens in the Yucatan Peninsula rarely are planted more than 3 y in a row, regardless of soil fertility or the forest type of the site (Redfield and Villa Rojas, 1962), gardens planted on sites in Late Secondary Forest reportedly can be planted about 3 y in a row, while sites in Early Secondary Forest usually are changed after 1 y. This difference is due, according to Maya gardeners, to increased soil fertility and reduced weed problems (Noguez-Galvez, 1991). The results of the present study suggest that gardeners at X-Hazil Sur are changing the types of gardens they plant and the number of times that crops are planted in these fields. According to recent studies in the region, gardeners plant two main types of gardens: One, the chacben garden is planted on newly cleared land in Late Secondary Forest. Two, the hubche garden is planted on Early Secondary Forest lands cultivated the previous year. Usually these gardeners have a single field of each type (Murphy, 1990; Edwards, 1986). This use of two garden categories may be confusing as other studies have used additional or different terms. Redfield and Villa Rojas (1962), for example, differentiated between garden types at Chan Kom in 1930-1931 of the basis of the harvest number (e.g., first-, second-, or third-year) and the previous status of the site (e.g., forest or fallow garden). According to this terminology, there were about six types of gardens. Villa Rojas (1987), however, used only three categories based on research at Seiior in 1935-1936. These categories were based solely on the previous status of the site and were named as follows: chacben (planted in virgin lands), zacab (planted in soils used 1-2 y before), and hubche (planted in soils fallowed for 7-8 y). This variety of terms, rather dian suggesting a multitude of garden types, probably represents regional or temporal differences in terminology. 213 Gardeners at X-Hazil Sur differed from those gardeners at Senor or Chan Kom in two aspects from this pattern of planting one old garden and one new garden: One, about 91-98% of the gardens at X-Hazil Sur were on new sites, rather than about 50%, as would be expected when single gardens of each of the two types are planted. Two, about 99% of the gardens at X-Hazil Sur were cleared from Early Secondary Forest and would correspond to the hubche category, rather than about 50%, as would be predicted based on the other studies. These results were roughly similar to those obtained by Noguez-Galvez (1991) during a separate study of soils at X-Hazil Sur. Thus, Maya gardeners at X- Hazil Sur typically are planting a new garden in Early Secondary Forest every year. As mentioned above, the reason for using Early Secondary Forest is not biological, rather it is due to ejido and state forestry rules that proscribe clearing Late Secondary Forest for gardens. Gardeners at X-Hazil Sur also differed from other indigenous people in that they apparently did not actively manage the forest or garden fallow, as has been reported elsewhere in Mexico. For example, traditional Maya practiced several silvicultural techniques, including the pet kot (vegetation enclosed within a wall of stones) and tolche (forested belt around gardens; G6mez-Pompa, 1987a, 1987b; G6mez-Pompa et al., 1987). Another form of forest management called the te'lom has been described for the Huastec Maya in northeastern Mexico (Alcorn, 1984a, 1984b). A te'lom is a group of useful trees situated on steep slopes and ridges. Some of these trees occur naturally, while others are transplanted from other gardens. While only a minimum of weeding occurs, the permanence of the site and the accumulated benefits of weeding and care over several years tend to favor the useful species. Neither Rewald (1989) nor Murphy (1990) observed these silvicultural techniques at Senor. Likewise, none of diese practices was observed at Ejido X-Hazil y Anexos during 1989-1990. Size. Several factors can influence the size of the field that a gardener will plant, including: soil fertility, number of helpers, weather conditions, family size of the gardener, age and health of the gardener, availability of suitable lands, time available to clear and plant, frequency and severity of crop predation by wildlife, and number of domestic animals that also must be fed. Hypothetically, a Maya gardener would plant the largest garden that he could successfully manage as this would provide 214 abundant food to eat or sell. At X-Hazil Sur and generally throughout the Yucatan Peninsula, however, there is little incentive to produce an excess of crops. Crop storage, for example, is a problem due to insects and crop predators (Villa Rojas, 1987). Also, rural Maya people generally do not buy large amounts of food on a regular basis, but instead use their cash to buy clothes and durable goods. While garden size may vary among families, the most important consideration is that a gardener must have a large enough garden to meet the needs of his family. Considering how weather, crop predation, and economic practices can vary over time and from region to region, it is remarkable that garden size apparently has varied little over the past 60 y in the Yucatan Peninsula. Redfield and Villa Rojas (1962), for example, reported average garden sizes of 2.9 ha (1930; n = 52 gardens) and 2.5 ha (1931, n = 46 gardens). Villa Rojas (1987) reported average garden sizes of 1.5-2.3 ha (n = 52 informants, each with 1-3 gardens; combined total of 4.7 ha of gardens per informant), depending on the type of garden. Murphy (1990) reported average garden sizes of 1.9 ha for old gardens and 2.9 ha for new gardens, for a total of 4.8 ha per household (n = 59 households). At X-Hazil Sur, the average garden size was 2.1 ha in 1989 and 2.7 ha in 1990. These values were very similar to the average garden sizes cited above and suggest that garden size varies little from year to year and region to region in the Yucatin Peninsula. Based on the 1989 garden data for X-Hazil Sur, the total amount of land used as gardens on the ejido was about 5.45 km" (ca. 1% of the total study area). This was substantially less than the figure of 33.56 km^ presented in Table 2-3 that was based on information provided by the Plan Estatal Forestal. While the reasons for this difference are not clear, it is possible that the photo-interpretation did not clearly differentiate between active gardens (mlpas) and previously active gardens (acahuales). This would be highly possible given the small size of gardens (1-3 ha), large scale of aerial photographs (1:37,000), and the time of year when the aerial photographs were taken. According to the information on the aerial photographs, the flights were conducted in February, about 3-4 months prior to the planting season. This likely contributed to an overestimate of the area in gardens because 215 at that time old and new gardens would have a similar appearance and would be hard to differentiate without resorting to complex laboratory analyses. Distance from X-Hazil Sur. In the Yucatan Peninsula, most Maya live in small rural villages and plant their gardens in forests on the outskirts of the village (Villa Rojas, 1987). As villages have increased in size, however, the amount of unoccupied land near settlements and available for horticulture has decreased. As a result, Maya gardeners have had to clear gardens at increasingly greater distances from the village. The distance that a gardener must travel to reach the field site is important because of the high frequency of visits he must make to manage the garden. In addition, the gardener must transport his crops from the garden to his home. Gardeners today usually walk or ride a bicycle on their almost- daily visits to the field, but as recently as the 1960s gardeners at X-Hazil Sur used horses and mules to meet their daily transportation needs (A. Foot Yam, pers. comm.). Few Maya gardeners have access to trucks for daily use. Under optimal conditions, a gardener on bicycle following a dirt trail through the forest might take 20-30 min to cover 3 km, but travel time can increase substantially when it is muddy or a load must be carried. Thus, it is highly advantageous for a gardener to minimize the distance to the field. At X-Hazil Sur, the majority of gardens were near the village, but gardens also were distributed in other areas of the ejido, except in the permanent forest reserve east of Mexico Highway 307. While the average garden distances at X-Hazil Sur were 4.0 km in 1989 and 3.8 km in 1990, about 50% of the gardens were at 1.0-2.9 km from the village, but gardeners also planted fields up to 16.6 km away. Perhaps the main reason for this widespread distribution of gardens is that the ejido is well-supplied with roads and trails that greatly facilitate access to garden sites (Figure 2-5). For example, Mexican Highway 307 crosses the ejido from north to south and the Pemex roads promote travel in the southern part of the area. As can be observed from Figure 2-5, the vast majority of gardens are within 2 km of these roads. These roads and trails greatly ease the work associated with reaching the gardens. 216 Crops planted. Among the several studies of Maya gardening practices, a common observation is that the Maya plant a wide variety of crops. Redfield and Villa Rojas (1962), for example, reported that gardeners in Chan Kom mainly planted com, but also cultivated beans, squash, jicamas, sweet potatoes, and chile, for a total of six crops. Villa Rojas (1987) indicated that gardeners at Senor planted 11 types of crops in their fields, and that com, beans, and squash were most frequently cultivated. Murphy (1990) reported nine crops at Senor in a later study; again, com, beans, and squash were most frequently planted. The number of types of crops planted by gardeners at X-Hazil Sur was greater than reported for other Maya gardeners. In 1989, gardeners at X-Hazil Sur reported planting 16 types of crops, while in 1990, 13 types were reported. Cora (99% of gardens in 1989 and 100% in 1990), kidney beans (48% and 75%), and squash (63% and 84%), as reported for other Maya gardens and by Noguez-Galvez (1991) in a separate study at X-Hazil Sur, were still the most frequently planted crops. The additional crops in gardens planted at X-Hazil Sur probably do not represent an expansion of the Maya diet because these crops traditionally have been cultivated by the Maya elsewhere in the Yucatan Peninsula (Nations and Nigh, 1980; Smith and Cameron, 1977). Rather, the appearance of additional crops in gardens probably represents an improvement in reporting practices by researchers. Previously, certain crops that were planted in small quantities or that required special attention were cultivated only in special plots called pet hot (G6mez-Pompa et al., 1987) and ka'anche (Barrera, 1980; Vargas Rivero, 1983). These specialized studies may have overlooked the fact that gardeners planted these crops in their regular gardens, too. While 16 types of crops were available to gardeners at X-Hazil Sur, not all gardeners planted all of these potential crops in their gardens every growing season. The average number of crops per garden was 5.9 in 1989 and 3.9 in 1990, but up to 11 crops were planted in a single garden. This suggests that gardens can be extremely diverse with respect to crops. These averages and ranges were substantially higher than reported elsewhere for Maya gardeners in the Yucatan Peninsula. In the only other study where specific crops were indicated for individual gardens, Redfield and Villa Rojas (1962) 217 reported that the average number of crops per garden was 2.0 (n = 53 gardens, SE = 0. 16, range = 1-5 crops per garden). One possible explanation for this difference is that gardens at X-Hazil Sur were culturally different from other Maya and traditionally planted a more diversified garden. A better explanation, however, is that the use of pet hot and ka'anche plots now is declining (G6mez-Pompa et al., 1987) and that certain types of crops that formerly were planted only in those plots now also are being planted in gardens. Crop phenology. Based on 1990 data, crops at X-Hazil Sur were present in gardens only about five months during the year, July-November. Gardens contained primarily seedlings during the first part of the growing season. Com, for example, was at the seedling stage during July-August (Figure 6-6), while kidney beans were at the seedling stage well during July-October (Figure 6-7). Most crops matured rapidly, however, and by the end of December, all had been harvested. Densitv and mortality of crops. Although the density of CTops varied greatly between gardens at X-Hazil Sur, a substantial amount of crops were potentially available to wildlife. The average density of crops varied greatly between gardens (45.8-179.0 stalks/m^; Table 6-1). Gardens primarily were composed of com (86% of stalks), beans (5%), and squash (3%). Seedling mortality was relatively high for squash (82% of seedlings) and kidney beans (62%), however, and compared with com (21% mortality) few mature plants survived to be harvested. All three of these crops were eaten by wildlife (see Chapter 5). Wildlife Use of Gardens and Adjacent Forest Crop predation. In this study, crop predation was used as one measure to quantify the degree and seasonality of wildlife use of gardens and aops. Given that hunters also hunt in their gardens to obtain game and to protect crops, crop predation might also explain some of the variation in game harvest pattems. Fourteen taxa of wild animals were identified as crop predators for gardens at X-Hazil Sur. Birds (e.g., Psittacidae, Corvidae, Icteridae, and Columbidae) were the main crop predators in 1989 (50% of 218 gardens), while in 1990, the coati (30%) and collared peccary (23%) were identified as the main crop predators. In one of the few studies that specifically mentions crop predation in the Yucatan Peninsula, Murphy (1990) reported that white-tailed deer and paca ate crops at the village of Senor. Wildlife tracks. Few track sets were recorded for the track slicks along gardens and transects. This corresponded with the low sighting rate for animal censuses (see Chapter 4). Some of the 17 taxa of wild animals represented on the slicks also were taken by hunters (see Chapter 3) while other taxa were observed only during animal censuses (see Chapter 4). Based on the number of track sets recorded, the armadillo (a nongame species, 32 track sets) and the agouti (a game species, 21 track sets) were the most abundant animals. Neither of these two taxa, however, were reported as crop predators (see Figure 6-9). Use of gardens versus adjacent forest. Although the paired comparisons of track sets in slicks along gardens and transects yielded only limited results, a few conclusions can be made. In Late Secondary Forest with Gardens, the total number of track sets (29 versus 11), as well as the number of taxa (8 versus 4) observed were greater in garden track slicks than in transect track slicks, respectively, by a ratio of more than two to one (Table 6-2). In Early Secondary Forest, however, the total number of track sets (21 versus 19) and number of taxa (8 versus 9) observed were about equal between garden and transect track slicks. This suggests that in Late Secondary Forest with Gardens, there was a substantial difference in wildlife use of gardens versus transect, but that in Early Secondary Forest wildlife use of gardens and transects was similar. Variation in wildlife use by vegetation type. The relative abundance of tracks sets suggests differences in animal abundance among vegetation types. Late Secondary Forest with Gardens had the highest ratio of track sets per reading at 1.05 (40 track sets per 38 readings), followed by Early Secondary Forest at 0.82, and Late Secondary Forest without Gardens at 0.33 (Table 6-2). One possible explanation of this is that the mixture of gardens and forests provides more food and cover for wildlife than either forests alone or the combination of gardens, fruit and vegetable plots, former gardens, catde corrals, and pastures. 219 Monthly variation in wildlife use of gardens and adjacent forest. The monthly frequency of track sets along gardens and transects were similar in that tracks were relatively frequent in the wet season during 7/90-10/90 (93% of track sets) and infrequent as the wet season ended and the dry season began during 11/90-1/91 (Figure 6-10). One possible explanation is that wildlife use of gardens was greater during the period between planting and harvest because of the availability of crops as food. After harvest there was no little or no food available in the gardens to attract the wildlife. A second explanation about the monthly differences in track set frequency is that juvenile animals accompanied their mother during part of the year. This would inflate population densities as two animals were occupying an area that one normally would hold. A third explanation is that wild animals may have restricted their activities in response to the increased number of adult men tapping s^odilla trees. While chicle is tapped during July-March at the study area, the majority of chicle is harvested during September- January (Barrera de Jorgenson, 1993). During 1989 at Ejido X-Hazil y Anexos, 311 chicleros (chicle tappers) harvested this product, while in 1990, 333 chicleros were active on the ejido. Most chicleros conmiuted daily between their home and work site. Since the sapodilla tree primarily is located in Late Secondary Forest, the density of chicleros was about 0.66 chiclerosDam?- Late Secondary Forest (322 chicleroslA%9Ai km^; Table 2-3). Given the intensity and duration of these disturbances associated with chicle taping, many species of wildlife may either have sought refuge in isolated areas without sapodilla trees or severely limited their activity during 10/90-1/91, for example, foraging during the night when chicleros were not working. Game harvest in gardens. Although areas categorized as Plots & Gardens composed only 6.1% of the study area, the number of wild animals harvested in those areas by Maya hunters was proportionally higher than expected for 10 of 12 game species (Table 3-3). The top five game taxa most frequently taken in Plots & Gardens were the white-tailed deer (62.5% of individuals taken were harvested in Plots & Gardens, n = 26 deer kills), ocellated turkey (Agriocharis ocellata; 50.0%, n = 6 kills), agouti (48.6%, n = 35 kills), coati (46.7%, n = 167 kills), and pocket gopher {Orthogeomys hispidus; 41.5%, n = 53 kills). Only the white-lipped peccary (three individuals taken during a single 220 outing) and the thicket tinamou {Crypturellus cinnamomeus; not known to use gardens or eat crops) were not taken in gardens. Indigenous people throughout South America exphcitly recognize that a function of gardens is to attract game for hunting (Dufour, 1990). Crops and the surrounding vegetation are especially attractive to deer, peccary, and rodents. Some indigenous groups, such as the Ka'apor and the Kuikuru of Brazil, respond to crop predation by planting more crops dian necessary for their personal consumption (Balee, 1984; Balee and Gely, 1989; Cameiro, 1983). Other groups, such as the Kay^o (Parker et al., 1983; Posey, 1982, 1983, 1984), disperse their gardens throughout the area. Indigenous groups in Mesoamerica also recognize that gardens can attract game. In southern Mexico, for example, Lacandon Maya farmers leave a portion of their com harvest standing in the field (Nations and Nigh, 1980). This enables hunters to harvest wild animals such as the deer, squirrels, pacas, coatis, and peccaries attracted to the com. In Nicaragua, wild animals such as pacas, agoutis, white-tailed deer, collared peccaries, and Baird's tapir {Tapirus bairdii) are major crop pests in Miskito Indian gardens (Nietschmann, 1973). Paca and white-tailed deer are frequently harvested in gardens and plantations by Miskito hunters. Wildlife densities in three forest successional stages. Due to small sample sizes, wildlife population density estimates could not be determined for all species and vegetation types. Instead, sighting frequencies were calculated for three forest successional stages: Late Secondary Forest without Gardens, Late Secondary Forest with Gardens, and Early Secondary Forest. Statistical comparisons could be made for four species. Significant differences were detected for plain chachalaca {Ortalis vetula) sighting frequencies between the three stages (x^ ^proximation = 8.4900, d.f. = 2, P = 0.0143; Table 4-1). No significant differences in sighting frequencies were detected between forest successional stages for squirrels (Sciurus spp.; P = 0.1596), coatis (P = 0.0877), or kinkajous (Potos flavus; P = 0.0665). When summarized by game versus nongame taxa, significant differences were detected for game bird sighting frequencies between the three forest stages (x^ approximation = 7.2947, d.f. = 2, P = 0.0261; Table 4-1). The greatest average sighting frequency was 5.8 sightings 221 per 10,000 km in Early Secondary Forest. No significant differences in sighting frequencies were detected between forest successional stages for game mammals (P = 0.1972), nongame mammals (P = 0.1752), or nongame birds (P = 0.1994). While it may be premature to speculate on these data, it appears that there was little difference in sighting frequencies of wildlife between three forest successional stages at Ejido X-Hazil y Anexos. Although sample sizes were small, it appears that the wildlife was using each of the three forest successional stages and not selecting for or against any of the three stages. This suggests that all three stages collectively or any one stage individually could be sufficient to ensure the survival of the wildlife on the study area. Crops and game food habits. Of the several crops planted in gardens at Ejido X-Hazil y Anexos, only six taxa were recorded as being consumed by game species. Com was most frequently taken by wild animals and occurred in 25.4% (frequency of occurrence) of 284 game stomachs that were analyzed during June 1989-October 1990 (Table 5-2). Squash was the second most frequently taken crop (6.0% of stomachs), while sweet potato, kidney bean, yuca, and the category other/zacate occurred in ^ 3.2% of stomachs each. Of die 12 game species, 10 consumed crops at Ejido X-Hazil y Anexos; crops were not recorded for die white-lipped peccary (three stomachs analyzed), and no pocket gopher stomachs were analyzed (contents could not be identified visually). By frequency of occurrence (each species considered separately), the ocellated turkey most frequently consumed crops (80.0% of stomachs; Table 5-2). The frequency of crop consumption for die remaining nine game species (excluding the pocket gopher), in declining rank order, was as follows: agouti (50.0% of stomachs), paca (41.7%), coati (40.3%), diicket tinamou (33.3%), great curassow {Crax rubra; 30.8%), collared peccary (24.1%), brocket deer {Mazama americana; 12.5%), white-tailed deer (9.1%), and plain chachalaca (4.8%). These results suggest that, by frequency of occurrence in game stomachs, most crops were of litUe or no importance to game diets. Com, however, was relatively important. Com also was the most frequendy planted crop. While die relative importance of com in game diets may suggest selection by 222 these game species for nutritional reasons, it could also mean that game animals are simply consuming com because it is relatively easy to find in gardens. These results also indicate that game species vary in their consumption of crops. Some species, such as the ocellated turkey and agouti, frequently consumed crops, while others either infrequently consumed crops or consumed none at all. While it is possible that many game animals were shot in gardens before they had the opportunity to consume any crops, it also is possible that they were consuming noncrops while in the gardens (see Chapter 5 for additional information about game diets). CHAPTER 7 SYNTHESIS AND CONCLUSIONS In this dissertation I have presented information about subsistence hunting by Maya Indians at Ejido X-Hazil y Anexos, Quintana Roo, Mexico. These Indians have hunted and planted gardens in the Yucatan Peninsula for several thousand years. That both of these practices continue today, after 50-100 y of acculturation, indicates that hunting and gardening are important elements of Maya culture. Maya subsistence hunting was studied to test the hypothesis of garden hunting. Linares (1976) described garden hunting as a game-procurement system where hunters specialized in certain species of terrestrial mammals whose population density was greater in the vicinity of gardens than in forested areas without gardens. According to Linares (1976), the densities of these species were greater because they ate garden crops and they were tolerant of human disturbances. As a consequence of their eating crops, these terrestrial mammals tended to occur through out die year in the vicinity of gardens planted by the hunters. This behavior resulted in a higher biomass of these species in the vicinity of the gardens than in the adjacent forest. In view of this higher density of game in the vicinity of gardens than in the forest, hunters shifted their hunting practices from sites in tropical forests to garden sites in order to focus primarily on those species of terrestrial mammals that occurred in the vicinity of gardens. Garden hunting, according to Linares (1976), resulted in the substitution of a naturally occurring wildlife community present in the area on a seasonal basis for a culturally created community of terrestrial mammals present in the area throughout the year. Garden hunting, according to Linares (1976) was based on several premises: one, wild animals taken as game by hunters occurred in gardens and the garden-forest ecotone; two, hunters focussed their harvest of game on those species of wild animals that used gardens; Uiree, the densities of these game species were greater in the vicinity of gardens tlian in forests without gardens; and four, these wild animals consumed crops from gardens. This study was designed to test each of these premises. 223 224 Game Harvest and Wildlife Use of Gardens Maya hunters at X-Hazil Sur harvested eight species of game mammals and four species of game birds. Game inventories were compiled by living at X-Hazil Sur and interviewing hunters immediately after they returned from a hunt (see Chapter 3). This encouraged hunters to report the game they harvested and facilitated accurate measurements of the game taken. Among the game harvested, the coati (Nasua nasua, n = 167 individuals taken during June 1989- October 1990), pocket gopher (Orthogeomys hispidus, n = 53), and paca {Agouti paca, n = 47) were the most frequently taken mammals, whereas the plain chachalaca (Ortalis vetula, n = 167) was the most frequently taken bird (Table 3-1). Hunters did not take several other potential game species, including insects, reptiles, and amphibians, which are widely eaten by other indigenous people in forested Neotropical areas, or other mammals such as the tapir (Tapirus bairdii), armadillo (Dasypus novemcifictus), or primates {Ateles geojfroyi and Alouatta palliata). This indicates that Maya hunters at X-Hazil Sur are taking a limited number of taxa from all of the species found in the area. Except for the white-lipped peccary (three individuals taken during a single outing), all game birds and mammals at Ejido X-Hazil y Anexos were taken at least some of the time in areas categorized as Combined/Early Secondary Forest. These areas were composed of gardens, former gardens, and fallowed areas that were reverting to forest. Eight game taxa were taken ^ 50% of the time in these areas (Table 3-7): pocket gopher (94.3% of kills in Combined/Early Secondary Forest, n = 53 kills), plain chachalaca (94.0%, n = 167), white-tailed deer (83.3%, n = 26), agouti (68.6%, n = 35), ocellated turkey (66.7%, n = 6), paca (65.9%, n = 47), coati (55.1%, n = 167), and brocket deer (50.0%, n = 16). These high proportions of game taken in areas categorized as Combined/Early Secondary Forest show that most of the game birds and mammals harvested by Maya hunters at X- Hazil Sur used gardens and the disturbed areas in the vicinity of gardens. The high proportion of game taken in the vicinity of gardens at Ejido X-Hazil y Anexos also has been noted in other studies of hunters/horticulturalists. In a study of the Runa Indians in Ecuador, Irvine (1987) determined that hunters concentrated on game species that occurred in gardens and 225 adjacent fallows and infrequently harvested game that occurred in the forest. In a study of Agta hunters and Palanan fanners in the Philippines, Peterson (1981, 1982) also determined that hunters concentrated on game species that consumed garden crops. In a study of Efe hunters and Lese horticulturalists in the Ituri Forest of Zaire, Wilkie (1989) also reported that regrowth forest surrounding horticulturalists' villages can and do provide substantial quantities of game. There were two key elements to each of these studies: One, the farmers/gardeners continually cleared new garden sites from the surrounding forest and abandoned old garden sites. Two, the vegetation of the areas was a mixture of crop fields, forest edges, and broken cover. Peterson (1981:20) concluded that farming expansion, at the level encountered in the Philippines, appeared to be supporting a high density of game rather than causing environmental destruction. Wildlife Densities Linares' (1976) third premise for garden hunting was that the densities of game animals were greater in the vicinity of gardens than in forests without gardens. Data on wildlife densities were obtained by conducting censuses along transects in three successional stages of forest; Late Secondary without Gardens, Late Secondary Forest with Gardens, and Early Secondary Forest (see Chapter 4). Censuses were conducted during March-November 1990. This period included one growing season, but not a complete year. Relatively few animals were sighted during the 121 censuses (ca. 240 km and 244 h, Appendix J). A total of 240 sightings were made (150 mammals and 90 birds). More than 50% of the taxa were sighted by census number 22, where as no new taxa were sighted after census number 75. This suggests that the censuses recorded most of the potential game birds and mammals at the study area. Among the birds and maimnals sighted, seven mammalian taxa and three avian taxa were game species. However, only the coati and plain chachalaca, among game species, were sufficiently abundant to calculate densities and compare sighting frequencies between forest successional stages. The coati had a population density of 1.0 individuals/km^ in Late Secondary Forest without Gardens 226 and 3.9 individuals/km^ in Late Secondary Forest with Gardens (Table 4-2). No coatis were sighted in Combined/Early Secondary Forest. The plain chachalaca had a population density of 8.7 individuals/km^ in Late Secondary Forest wiUiout Gardens, 2.4 individuals/km" in Late Secondary Forest with Gardens, and 20.3 individuals/km^ in Combined/Early Secondary Forest (Table 4-2). Due to small sample sizes, it was not possible to compare densities statistically, however, it was possible to compare average sighting frequencies. Average sighting frequencies were significantly different among three forest successional stages for the coati (P = 0.0877) and the plain chachalaca (P = 0.0143; Table 4-1). Coatis were more frequently sighted in Late Secondary Forest with Gardens, whereas plain chachalacas were more frequently sighted in Combined/Early Secondary Forest. Given the nature of these data, animal taxa were lumped into four categories of game versus nongame birds and mammals in order to generalize about wildlife densities. For game mammals, nongame mammals, and nongame birds, there were no significant differences in average sighting frequencies among the three forest successional stages (P > 0.05). Only those taxa categorized as game birds, primarily plain chachalacas, exhibited significant differences in average sighting frequencies among the three forest successional stages (P = 0.0261; Table 4-1). This indicates that even for broad groups of animals, there generally were no significant differences detected in average sighting frequencies among forest successional stages. This general absence of significant differences in average sighting frequencies among forest successional stages for bird and mammals species, taxonomic groups, and game and nongame birds and mammals may be due to small sample sizes (e.g., too few censuses to conducted statistical analyses) and experimental design problems (e.g., the transects may have been too close together). However, it is highly possible that in fact there were no differences in wildlife densities between forest areas in the vicinity of gardens and forest areas not in the vicinity of gardens, as had been proposed by Linares (1976). This might be expected given the low human population density (ca. 3.0 people/km"), high degree of habitat interspersion, and ecological characteristics of the wild animals that occur at the study area (see Chapter 2). 227 The lack of significant differences in game population densities among forest successional stages determined at Ejido X-Hazil y Anexos also has been noted in other areas. Wilkie and Finn (1990) studied densities for 19 species of terrestrial mammals at the Ituri Forest and determined that (1) only three species were less dense in post-agricultural forest and (2) small duikers {Cephalophus spp. , Bovidae Family), although an important source of food for humans, seemed unaffected by forest clearing for subsistence farming. Wilkie and Finn (1990) indicated, however, that habitat loss or hunting pressure had reduced the abundance of larger ungulates near villages. This would suggest that game species at X-Hazil could sustain limited levels of horticulture and hunting without experiencing differences in their population densities among forest successional stages, but that increased levels could negatively affect the larger species. Consumption of Garden Crops by Game Species Linares' (1976) fourth premise was that the game species taken by hunters would consume crops from the gardens. At X-Hazil Sur, 16 taxa of crops were planted, but not all gardens had all 16 crops. Com, kidney beans, and squash were the three most frequently planted crops. Com was planted in 99% of the gardens during 1989 (n = 150 gardens) and 100% of the gardens in 1990 (n = 40 gardens; Figure 6-4). Squash and kidney beans also were planted in most gardens during 1989 and 1990. Crops generally were available in gardens between May, the beginning of plant season, and January, the end of the harvest season (Figure 6-1). Six crops were frequently consumed by game species. Com was most frequently consumed (25.4% of 284 stomach samples), but squash (6.0%), other/zacate (3.2%), sweet potato (2.8%), kidney bean (1.4%), and yuca (0.3%; Table 5-2) also were eaten by game taxa. Com, squash, and other/zacate, including crops from the currents and previous growing seasons, were consumed throughout the year by game species, whereas the other three crop species were consumed only during January- April, when gardens were being cleared and bumed (Table 5-3). Thus, whereas com was the crop taxa most frequently consumed by game species, 6 of the 16 crops planted in gardens were eaten. 228 Four game species were the primary consumers of crops: paca, agouti, coati, and collared peccary (Tables 5-4 and 5-5). These four taxa composed four of the top six game species harvested by number of individuals (rank order 1, 4, 5, and 6) and by total body weight (rank order 2, 3, 4, and 6; Table 3-1). This indicates that at Ejido X-Hazil y Anexos some of the most frequently taken game species were major consumers of crops and supports one of Linares' (1976) assumptions. In conclusion, these data suggest that Maya Indians at Ejido X-Hazil y Anexos did not practice garden hunting, where the biomass of game, it is proposed, is increased and able to offset the loss due to hunting. Rather, the data suggest that gardens are acting as a sink (cf., Pulliam, 1988; PuUiam and Danielson, 1991) and that Maya hunters are taking as game those species that occasionally feed on crops. In order to help develop wildlife management plans for the area, it would be important to calculate wildlife production and potential harvest for the main game species (Redford, 1992). Productivity is measured by calculating population density and reproductive rate (cf . , Eisenberg and Seidensticker, 1976; Eisenberg et al., 1973). Robinson and Redford (1991c) proposed a model to calculate production using three elements: (1) population density at carrying capacity, (2) the maximum rate of population increase, and (3) the density that produces the maximum sustainable yield. Unfortunately, the production and potential harvest values calculated by Robinson and Redford (1991c) are not directly applicable to the present study because the average densities they used were at least 10 times greater ~) than those at Ejido X-Hazil y Anexos. Additional census and reproductive data are necessary to determine potential harvest levels. Of Linares' (1976) four premises, three were supported: one, wild animals clearly used gardens; two, hunters concentrated their hunting on those animal species that occurred in gardens; and three, game species consumed crops found in gardens. The X-Hazil data, however, did not support the model of garden hunting with respect to wildlife densities. Linares (1976) predicted that game species taken by garden hunting would have greater densities in the vicinity of gardens than in forests without gardens. Hunters at X-Hazil Sur commonly reported that many game species were more abundant in 229 the vicinity of a garden than at forested sites away from gardens. This study, however, generally did not find differences in the density of wildlife in the vicinity of gardens as compared with their densities in forest areas without gardens. Closing Comments Given the nature and extent of subsistence hunting by Maya Indians at Ejido X-Hazil y Anexos, the harvest of some game species likely can continue for the indefinite future. Population levels of the coati, plain chachalaca, pocket gopher, paca, collared peccary, and agouti appear to be stable. For the white-tailed deer, brocket deer, thicket tinamou, great curassow, and ocellated turkey, population levels do not appear to be stable, and hunters frequently complained that these species had noticeably declined in number during the past 20-30 y. Hunters also reported that previously they used to hunt primates (Ateles and Alouatta) and the crested guan (Penelope purpurascens), but that these three species now were so hard to find that they ahnost never were hunted. Presently there are no effective enviroiunental education or wildlife management programs underway at the study area. As a result, hunters harvest game without regard to age or sex of the prey. If the local wildlife populations are to remain viable, effective education and management programs must be implemented. Recognizing the independent nature of many Maya Indians with respect to govenmient regulations and prohibitions, this may be difficult. However, education and management programs are presently being used by local officials and state forestry managers to regulate the harvest of timber and chicle, the latex of the s^odilla tree (Manilkara zapota; cf., Barrera de Jorgenson, 1993). I would strongly suggest that Mexican wildlife officials establish a wildlife management program diat is similar to the forestry program and is designed to accommodate the special situation of Mexican ejido s. Such an arrangement could include bag limits, established seasons, and areas on the ejido were hunting would be prohibited. At the same time, conservationists and education officials could institute activities in the local schools and villages. These educational activities should be based on species familiar to the Maya. Such a plan likely could assure the continued survival of the 230 game birds and mammals at Ejido X-Hazil y Anexos as well as the cultural traditions of the Maya Indians that have subsisted on these species for millennia. APPENDIX A TEMPERATURE (°C) AND PRECIPITATION (MM) AT X-HAZIL SUR, QUINTANA ROO, MEXICO, DURING JULY 1989-DECEMBER 1990 July 1989 !\ugust 1989 September 1989 Date TMin* TMax" Precip' TMin TMax Precip TMin TMax Precip 1 22 34 9.0 23 35 M 2 22 37 5.0 24 36 M 3 23 32 2.0 23 37 4.4 4 23 35 M^ 23 34 0.0 23 35 0.0 5 23 36 0.0 23 31 0.0 23 36 0.5 6 24 34 M 19 31 11.0 23 34 0.5 7 24 34 1.0 22 35 0.0 24 37 1.0 8 22 36 3.2 23 36 0.0 24 37 0.1 9 21 35 0.2 23 35 4.0 24 33 7.0 10 21 33 0.3 22 35 0.0 24 35 0.2 11 25 34 0.0 24 36 0.0 24 36 80.0 12 24 37 0.0 24 35 4.8 23 33 4.0 13 24 36 0.0 23 35 0.0 23 34 18.5 14 22 37 0.0 23 33 8.5 24 33 3.6 15 22 37 0.0 23 33 0.0 25 30 23.0 16 23 38 0.0 24 34 4.8 24 32 6.5 17 24 37 0.0 24 34 0.2 24 33 22.6 18 24 37 0.0 24 35 24.0 24 35 0.3 19 24 38 0.0 24 36 12.5 24 33 0.1 20 24 38 0.0 23 35 0.3 24 34 1.0 21 24 38 0.0 24 34 6.0 24 29 4.1 22 23 38 0.0 25 34 0.0 25 31 0.1 23 23 37 0.0 24 35 M 24 32 4.0 24 24 38 M 24 35 M 25 31 47.0 25 23 34 40.0 23 35 10.5 24 36 0.0 26 23 34 40.0 23 30 15.0 24 33 0.0 27 22 31 24.5 24 34 0.7 24 32 2.8 28 23 30 82.0 24 36 0.3 24 30 33.0 29 23 30 20.5 24 35 0.0 24 32 2.2 30 24 31 19.0 24 35 0.0 24 32 4.0 31 24 34 0.0 23 36 0.0 Average 23.2 35.3 23.2 34.4 23.9 33.5 Total 230.7 118.6 270.1 ' TMin = Daily minimum temperature. •" TMax = Daily maximum temperature. ' Precip = Precipitation. " M = Mist (< 0 1 ml). 232 233 ( 3ctober 1989 November 1989 December 1989 Date TMin" TMax" Precip' TMin TMax Precip TMin TMax Precip 1 24 33 3.2 19 31 0.0 20 32 10.0 2 23 33 7.0 20 31 0.0 20 28 1.5 3 24 32 17.5 23 32 4.3 20 26 0.3 4 24 33 4.6 21 28 23.0 19 32 0.0 5 23 33 0.0 23 30 0.1 16 28 0.0 6 23 33 0.0 19 31 0.0 16 29 1.6 7 23 33 2.2 21 31 0.9 21 30 3.5 8 24 32 0.9 21 32 0.0 21 31 15.0 9 23 33 3.0 22 32 11.0 20 24 0.4 10 23 31 1.0 21 29 16.0 14 24 0.1 11 23 31 1.0 22 28 0.5 14 30 0.0 12 23 33 0.0 22 28 0.5 19 23 1.8 13 23 32 0.0 22 32 0.0 21 25 0.0 14 22 31 0.0 22 31 0.0 19 28 1.6 15 23 33 0.0 21 30 0.0 19 29 5.0 16 23 32 0.0 22 31 4.0 21 30 0.3 17 23 34 0.0 21 26 13.5 19 30 0.0 18 23 34 7.0 21 29 8.0 19 31 0.0 19 22 24 0.2 24 30 0.1 19 30 0.0 20 22 22 0.0 23 30 0.7 19 30 0.0 21 21 23 0.0 22 31 0.0 20 30 0.0 22 19 30 0.0 20 30 0.0 18 30 0.0 23 18 29 0.0 20 31 9.0 19 20 0.5 24 19 30 0.0 21 30 0.5 11 17 0.0 25 19 32 2.8 21 32 0.7 12 23 0.0 26 19 32 0.0 22 30 3.4 12 23 0.0 27 19 32 0.0 24 31 0.0 9 26 0.0 28 19 32 0.0 21 31 11.0 14 27 M 29 18 32 0.0 21 29 2.0 13 29 0.7 30 18 32 0.0 22 23 0.2 19 31 0.1 31 18 32 0.0 20 30 0.0 Average 21.4 31.2 21.5 30.3 17.5 27.4 Total 50.4 109.4 42.4 ' TMin = Daily minimum temperature. " TMax = Daily maximum temperature. ' Precip = Precipitation. " M = Mist (< 0.1 ml). 234 January 1990 February 199C ) March 1990 Date TMin' TMax" Precip' TMin TMax Precip TMin TMax Precip 1 19 31 0.5 20 30 0.0 20 31 0.7 2 19 29 M 20 30 0.0 17 31 0.0 3 17 29 17.0 19 30 M 20 31 0.5 4 21 28 0.8 19 32 0.0 20 30 0.0 5 21 30 5.2 18 30 0.9 18 32 0.0 6 20 30 0.5 22 29 0.9 18 32 M 7 20 30 0.0 22 31 0.0 20 32 6.0 8 20 29 0.0 19 31 0.0 20 31 0.4 9 19 30 0.5 19 31 0.0 19 31 15.5 10 19 30 0.0 19 33 0.0 20 30 0.4 11 19 29 2.0 21 31 0.0 23 30 4.0 12 20 27 2.2 17 29 0.4 22 31 0.0 13 20 27 21.0 19 31 0.0 19 32 0.0 14 19 26 0.4 19 32 0.0 19 31 0.3 15 18 27 4.0 21 34 0.0 25 33 0.0 16 20 29 0.9 22 33 0.0 25 32 67.0 17 20 29 0.0 21 33 0.0 25 32 67.0 18 20 29 0.7 20 32 0.0 19 28 0.7 19 21 29 3.0 23 32 0.0 20 32 16.5 20 20 29 1.6 17 31 0.0 21 28 8.5 21 20 30 0.3 19 33 0.0 17 22 0.2 22 21 30 0.0 21 32 0.0 13 31 0.0 23 20 28 1.2 21 39 0.0 17 30 0.0 24 17 29 0.0 16 27 3.8 20 31 0.5 25 18 30 7.0 16 21 0.0 19 31 1.3 26 20 25 7.3 17 29 0.0 23 31 4.6 27 19 30 0.0 18 31 4.6 23 31 0.0 28 19 27 0.0 20 30 0.6 19 32 0.0 29 19 31 0.0 19 32 0.0 30 18 30 0.0 23 34 0.0 31 20 29 0.0 23 33 0.0 Average 19.5 28.9 19.5 31.0 20.2 30.9 Total 76.1 11.2 127.1 ' TMin = Daily minimum temperature. " TMax = Daily maximum temperature. ' Precip = Precipitation. " M = Mist (< 0.1 ml). 235 April 1989 May 1989 June 1989 Date TMin" TMax" Precip" TMin TMax Precip TMin TMax Precip 1 22 35 0.0 21 34 0.0 26 36 0.0 2 21 34 0.0 21 34 0.0 24 36 4.0 3 22 37 11.2 21 34 0.0 24 36 0.0 4 23 35 00. 21 34 0.0 23 37 0.0 5 21 35 0.0 21 34 0.0 23 37 0.0 6 21 35 0.0 20 36 0.0 24 34 M 7 22 30 7.5 21 36 0.0 23 33 16.0 8 22 33 0.5 20 34 0.0 23 27 20.5 9 22 34 0.0 24 37 0.0 23 30 1.0 10 22 33 0.0 24 38 0.0 23 34 0.0 11 21 33 0.0 23 36 0.0 24 34 0.0 12 21 34 13.5 23 36 0.0 23 35 0.0 13 22 32 0.0 22 37 0.0 24 35 0.0 14 21 34 0.0 22 37 0.0 23 38 0.0 15 20 36 0.7 23 36 0.0 24 36 9.0 16 22 36 0.0 22 35 1.8 23 35 0.0 17 21 35 0.0 23 37 0.0 23 37 0.0 18 16 35 0.0 23 37 0.0 24 36 0.0 19 19 33 4.0 23 36 4.4 24 34 3.0 20 19 33 0.2 24 35 0.0 23 37 0.0 21 21 33 0.8 25 35 0.5 25 32 1.6 22 21 33 2.8 29 30 2.2 25 35 0.2 23 21 32 0.0 24 35 3.4 25 37 0.0 24 18 34 0.0 24 36 0.0 24 36 0.0 25 20 34 0.0 24 38 7.9 25 37 0.0 26 20 33 0.0 25 37 0.1 24 37 0.2 27 24 34 40.0 24 34 0.2 23 33 16.0 28 22 32 0.0 25 39 0.0 23 26 21.0 29 24 35 0.0 26 39 0.0 24 30 0.6 30 22 35 0.0 26 38 0.0 25 33 0.0 31 26 36 0.0 Average 21.1 33.9 23.2 35.8 23.8 34.4 Total 81.2 20.5 103.1 ' TMin = Daily minimum temperature. *" TMax = Daily maximum temperature. ' Precip = Precipitation. " M = Mist (< 0.1 ml). 236 July 1990 fSiugust 1990 September 1990 Date TMin' TMax" Precip' TMin TMax Precip TMin TMax Precip 1 23 35 0.0 23 37 0.0 22 33 17.5 2 22 35 0.0 24 36 0.0 23 29 3.0 3 24 34 1.0 25 35 2.4 23 29 5.0 4 24 35 0.0 24 36 3.0 22 32 24.5 5 25 33 9.0 24 26 60.2 22 31 2.2 6 24 35 4.6 23 28 8.0 22 32 2.4 7 24 34 1.2 22 34 0.0 23 35 0.2 8 23 33 27.5 24 34 6.0 22 34 0.3 9 23 33 0.5 23 30 0.0 24 34 10.0 10 22 35 2.4 22 34 4.0 23 33 0.0 11 22 35 0.4 23 35 1.0 24 36 0.0 12 23 34 0.4 24 36 0.0 24 36 0.0 13 23 34 1.0 24 35 0.0 23 35 0.2 14 23 36 0.0 23 33 0.0 23 35 6.0 15 23 36 4.4 23 35 5.0 22 32 7.5 16 22 33 10.5 23 34 2.2 23 33 1.4 17 23 35 5.8 24 34 0.5 24 34 0.4 18 22 35 1.0 23 36 0.0 24 34 0.0 19 23 33 13.0 22 37 0.0 24 34 1.6 20 22 32 35.0 22 36 0.0 24 34 0.0 21 22 34 0.5 23 36 4.0 23 33 17.5 22 23 34 5.0 23 35 0.0 22 34 0.0 23 24 35 0.0 24 35 0.5 22 35 0.0 24 23 36 0.0 24 35 1.5 22 35 0.0 25 24 36 0.0 23 37 0.0 22 35 135.9 26 23 38 1.0 23 36 5.0 22 34 0.0 27 22 34 0.0 23 35 0.0 23 34 4.4 28 22 36 0.0 24 34 1.0 23 33 3.8 29 23 38 0.0 23 36 0.0 23 30 0.0 30 22 37 0.0 20 36 0.0 22 33 0.0 31 23 36 0.0 21 37 0.0 Average 23.0 34.8 23.1 34.0 22.9 33.4 Total 124.2 104.3 243.8 ' TMin = Daily minimum temperature. '' TMax = Daily maximum temperature. ' Precip = Precipitation. " M = Mist (< 0.1 ml). 237 1 October 1990 November 1990 December 1990 Date TMin' TMax" Precip' TMin TMax Precip TMin TMax Precip 1 23 29 42.0 21 30 4.0 22 25 11.5 2 22 34 17.0 21 31 0.0 22 27 10.0 3 24 32 0.3 20 32 0.5 22 29 1.8 4 22 34 8.0 23 30 17.5 22 28 0.0 5 24 33 0.0 19 27 3.0 22 24 17.0 6 23 32 30.0 24 26 15.0 21 26 2.6 7 23 34 0.0 24 31 0.0 21 28 0.5 8 23 33 0.0 20 31 0.0 19 25 0.0 9 23 34 0.0 22 33 0.0 13 24 0.0 10 22 34 3.0 18 28 0.0 12 25 0.0 11 22 33 11.0 17 27 0.0 14 27 0.0 12 24 31 0.0 19 26 5.5 13 13 19 32 0.0 18 23 27.5 14 18 33 0.0 18 26 46.0 15 20 34 0.0 23 28 15.0 16 21 34 0.0 18 30 18.0 17 24 32 10.0 23 33 0.0 18 23 33 0.0 21 27 0.0 19 19 33 0.0 18 29 0.0 20 21 33 1.0 17 30 0.0 21 22 33 0.0 16 29 0.0 22 23 34 0.0 18 30 0.0 23 23 34 5.0 17 30 0.0 24 23 33 0.0 17 30 0.0 25 24 31 0.0 17 29 0.0 26 19 29 0.0 16 29 0.0 27 16 29 0.0 20 31 0.0 28 18 29 0.0 20 31 3.5 29 17 31 0.0 23 30 36.0 30 19 31 1.6 23 27 20.0 31 20 32 2.0 Average 21.5 32.4 19.7 29.1 18.6 26.2 Total 130.9 211.5 43.4 " TMin = Daily minimum temperature. "" TMax = Daily maximum temperature. ' Precip = Precipitation. ^ M = Mist (< 0.1 ml). APPENDIX B MEAN BODY MASS (G), DIET CLASSIFICATION, MEAN POPULATION DENSITY (NUMBER OF INDIVIDUALS/KM^), AND GAME STATUS FOR SELECTED MAMMALS IN QUINT ANA ROO ,300 FO 13.4 N ,041 FO 55.3 N N N N 400 lO 62.9 N 239 Mean body Diet Mean population Game Taxa' mass (g)'' class' density" status"" a) Marsupials Caluromys derbianus Didelphis marsupialis Didelphis virginiana Marmosa canescens Mamiosa mexicana Philander opossum b) Primates Alouatta pigra 6,500 FH 12.0 N Ateles geoffroyi 7,500 FH 18.0 N c) Edentates Dasypus novemcinctus 3,544 lO 21.9 N Tamandua mexicana 4,210 MY 5.2 N d) Lagomorphs Sylvilagus floridanus 1,025 HG 35.0 N e) Rodents Agouti paca 8,227 FG 27.5 G Coendou mexicanus N Dasyprocta punctata 3,600 FG 19.7 G Onhogeomys hispidus 400 G Sciurus deppei 225 N Sciurus yucatanensis 400 N f) Carnivores Bassariscus sumichrasti Conepatus semistriatus Eira Barbara Felis concolor Felis pardalis Felis wiedii N 1,700 FO 13.8 N 3,980 FO 1.0 N 37,000 CA 0.1 N 10,460 CA 0.8 N CA N 5,000 CA 0.5 N 2.910 CA 2.4 N N 3,880 FO 15.1 G 68,750 CA 0.1 N 2,490 FO 24.4 N 8,850 FO 10.0 N 1,750 N 240 Mean body Diet Mean population Game Taxa" mass (g)' class' density" status'' Felis yagouaroundi Galictis vittata Mustela frenata Nasua nasua Panthera onca Potos flavus Procyon lotor Urocyon cinereoargenteus g) Perissodactyls Tapirus bairdii 275.000 FH 0.5 N h) Artiodactyls Tayassu pecari Tayassu tajacu Mazama americana Odocoileus virginianus 40.000 Iffi 2^8 G ' Excludes bats, insectivores, small rodents, marine mammals, and endemic species located on off-shore islands. " Based in part on Navarro L. et al. (1990) and Robinson and Redford (1986, 1989). Additional information obtained from animals collected during this study (see Chapter 3 for additional information about weights of game species). ' Diet class (adapted from Eisenberg, 1981): CA = carnivore, FG = frugivore-granivore, FH = frugivore-herbivore, FO = frugivore-omnivore, HB = herbivore-browser, HG = herbivore-grazer, lO = insectivore-omnivore, and MY = myrmecophage. ■* Game status was based on information provided by Maya hunters at Ejido X-Hazil y Anexos: G = game species and N = nongame species (see Chapter 3 for additional information about how game species were defined). 28,550 FH 4.9 G 17,520 FH 11.9 G 26,100 FH 10.5 G APPENDIX C MEAN BODY MASS (G), DIET CLASSIFICATION, MEAN POPULATION DENSITY (NUMBER OF INDIVIDUALS/KM'), AND GAME STATUS FOR SELECTED BIRDS IN QUINTANA ROO" mass (g)' Diet class' Game status'' 432-485 FG NO 350-500 FG G FG NO 900-1200 FG NO Taxa" a) Tinamous Crypturellus boucardi Crypturellus cinnamomeus Crypturellus soui Tinamus major b) Chachalacas. Guans. and Curassows Crax rubra 3100-4270 FH G Penelope purpurascens 1620-2430 FH NG Ortalis vetula 397-527 FH G c) Quails Odontophorus thoracicus 170-266 10 NG d) Turkeys Agriocharis ocellatus 2600-3100 FO G e) Pigeons and Doves Columba flavirostris 241-336 FG NG Columba speciosa 254-287 FG NG Geotrygon montana 110-145 FG NG Leptotila spp. 145-205 FG NG Zenaida asiatica 130-196 FG NG e) Parakeets and Parrots Aratinga astec FG NG Amazona autumnalis FG NG ' Based in part on Leopold (1972), L6pez Omat (1990), Paynter (1955a, 1955b), and Terborgii et al., 1990. Additional information obtained from animals collected during this study (see Chapter 3 for additional information about weights of game species). '' Potential game taxa only. ' Based in part on Terborgh et al. (1990). '' Based in part on Leopold (1972, 1977), Paynter (1955a, 1955b), and information obtained from Maya hunters during this study. 242 APPENDIX D DATA FORMS USED DURING HUNTER INTERVIEWS. FORM A INCLUDES SPECIFIC QUESTIONS ABOUT THE OUTING (ONE FORM PER OUTING), WHEREAS FORM B INCLUDES QUESTIONS ABOUT THE GAME SPECIES TAKEN (ONE FORM PER PREY ITEM). BOTH FORMS ORIGINALLY APPEARED IN SPANISH Form A: JORGENSON AND CO., HUNTER INTERVIEW, Q. ROO, MEXICO, 1989-1990 1 . Record number? Field number? 2. Interviewer? 1-JPJ, 3-ABdeJ, 7-RUCh, 8-ABX, 9-Other 3. Interviewee? 4. Interview date? / / (month/day/year) 5. Kill date? / / 6. (Question not used) 7. (Question not used) 8. Time hunting outing began (departure from house)? (0000-2359; 0000 = midnight, 1200 = noon. 1800 = 6 pm) 9. Time of game kill? 10. Time hunting outing ended (returned to house)? 1 1 . Total duration of hunting outing? 12. Hunter group size? 13. Group members (names)? 14. Number of shotguns in group? 15. Number of rifles in group? 16. Number of other types of weapons? (See 32.2-32.20 for dogs) 17. What types of weapons? 18. Type of hunting? 1 -Tracking/Stalking, 2-Stand/Platform, 3-Trapping, 9-Other 19. If tracking/stalking, distance covered during the hunting outing? m 20. Location of the game kill site (direction and distance from the hunter's house or other landmark) ? 21. If the game kill site is a garden, the name of the garden owner? 22. Map coordinates of tlie game kill site? Other animals noted during the hunting outing: 23. Species? 24. Type of observation? 1-Visual, 2-Sign (feces, tracks, nest, etc.), 3-Sound/Call, 9-Other 25. Number of individuals sighted/observed? 26. Vegetation type where animal was sighted/observed? 244 245 1-Vegetable plot, 2-Garden [2.1-Newly cleared garden, 2.2-Newly planted garden, 2.3- Harvested garden, 2.4-Old garden], S-Orchard, 4-Early secondary forest, 5-Late secondary forest, 5.1-"Bumed" late secondary forest, 5. 2- "Partially cleared" late secondary forest, 6- "Low height" late secondary forest, 7-SeasonaI wetlands, 9-Other Result of sighting/observation: 27. Animal shot at? 1-Yes, 2-No, 3-Not applicable 28. Firearm/ weapon used? 10-20 gauge shotgun, 11-16 gauge shotgun, 12-12 gauge shotgun, 20-22 caliber rifle, 21-Other caliber rifle, 30-Trap/Snare, 40-Not applicable, 99-Other 29. Name of hunter? 30. Result of sighting/observation? 1 -Animal hunted/collected/trapped, 2- Animal escaped (no shot), 3-Shot missed the animal, 4- No shot attempted, 5-Wounded, 9-Other 30. 1 Total number of shots fired? 31. Field number of hunted/collected animal? JPJ- 32. If the outing was specifically to hunt, why did the hunter go to this specific site?_ 32.1 Main activity of the outing? 1-Hunting, 2-Work in the garden, 3-T^ chicle, 4-Logging, 9-Other 32.2 Did dogs accompany the hunter? 1-Yes, 2-No 32.3 How many dogs accompanied the hunter? 32.4 Name of dog? 32.5 Sex of dog? 1-Male, 2-Female, 3-Unknown 32.20 Name of the dog that first located or killed the prey? 33. Notes (for example, moon, weather, fresh logging or garden burning, ripe fruits or vegetables) : 34. Map of the hunting outing and game kill site (note distances, directions, and landmarks): 246 Form B: JORGENSON AND CO., ANIMAL COLLECTION, Q. ROO, MEXICO, 1989-1990 1. Record number? 2. Field number? JPJ- Date (month/date/year) 3. Kill date? / / 4. Interview date? / / 5. Hunter/collector? 6. (Question not used) 7. Scientific name of animal? 8. Conunon name in Maya/Spanish? 9. Country? Mexico 10. State? Quintana Roo 11. Municipio? 1 -Felipe Carrillo Puerto, 9-Other 12. Name of the general locality? 1-Rancho Las Palmas, 2-Ejido X-Hazil y Anexos, 9-Other 13. Location of the game kill site (direction and distance from the hunter's house or other landmark) ? 14. Latitude? ° ' North 15. Longitude? ° 'West 16. Elevation? m 17. Map coordinates of the game kill site? 18. Vegetation type where animal was sighted/observed? 1-Vegetable plot, 2-Garden [2.1-Newly cleared garden, 2.2-Newly planted garden, 2.3- Harvested garden, 2.4-Old garden], 3-Orchard, 4-Early secondary forest, 5-Late secondary forest, 5.1-"Bumed" late secondary forest, 5. 2- "Partially cleared" late secondary forest, 6- "Low height" late secondary forest, 7-Seasonal wetlands, 9-Other 19. Age of the game kill site? y 19.1 Name in Maya for this type of vegetation? 20. Weight/measurements of the animal measured (1) or estimated (2)? 21. Total length of animal (from nose/bill to tip of tail)? mm 21.1 Length of the folded wing (birds only)? nam 22. Tail length of animal? mm 23. Hind foot length of animal (tarsus for birds; right/left)? mm 247 24. Ear length of animal (bill for birds; right/left)? mm 25. Total weight of animal? g 25.1 Net weight of animal (without guts etc.)? g 26. Total weight of animal estimated by the hunter? g 27. Sex of animal? 1-Male, 2-Female, 3-Unknown 28. Reproductive condition? Females: 1-Lactating, 2- With embryos/eggs, 3-With a litter, 4-Not lactating/no embryos/no litter. 5-Unknown Males (condition of the testes): 6-Abdominal, 7-Scrotal, 8-Unknown Sex unknown: 9-Unknown 29. Age of the animal estimated by the hunter? 1-Aduh, 2-Subadult/young, 3-Newbom, 4-Unknown 29.1 Behavior of the animal at the time when hunted/collected? 1-Resting, 2-Eating, 3- Walking/running/flying, 9-Other 30. Time of game kill? ((X)00-2359; (WOO = midnight, 1200 = noon, 1800 = 6 pm) 30.1 Weights/measurements from a dead (1) or live (2) animal? Sample Collected Form Preparation 30.2 Complete animal? 30.3 30.4 Feathers? 30.5 30.6 Feet? 30.7 31. Skull? 32. 33. Skin? 34. 35. External parasites? 36. 37. Internal parasites? 38. 39. Internal organs? 40. 41. Stomach contents? 42. Number Items 30.31 30.51 30.71 32.1 34.1 36.1 38.1 40.1 42.1 Form of preparation: 1-Dry, 2-Salt, 3-Com meal, 4-15% Formalin, 5-70% Ethyl alcohol, 9-Other 43. Total volume of the stomach contents (incl. 43.1 & 43.3)? ml For birds (43.1-43.4): 43.1 Subtotal in gizzard? ml 43.2 Number of containers? 43 . 3 Subtotal in mouth/esophagus/aop? ml 248 43.4 Number of containers? 44. Preliminary identification of the stomach contents, according to the hunter? 45. Type of outing? 1-Active hunting/fishing/trapping, 2-Opportunistic hunting/fishing/trapping, 3-Road kill, 9- Other 46. Purpose of the collection? 1-Personal consumption (including limited sale or exchange), 2-Personal pet, 3-Pet to be sold commercially, 4-Kill a noxious animal (not for consumption), 5-Scientific collection, 9-Other 47. Interviewer? . 1-JPJ, 3-ABdeJ, 7-RUCh, 8-ABX, 9-Other 48. Notes: APPENDIX E GAME HARVEST INVENTORIES FOR THE SEVEN MAIN HUNTERS (315 PREY WITH WEIGHTS) AT X-HAZIL SUR COMPARED WITH THE GAME TAKEN BY THE REMAINING 79 HUNTERS (261 PREY WITH WEIGHTS) THAT REPORTED TAKING GAME (ARRANGED BY GRAND TOTAL NUMBER OF PREY ITEMS) Hunter number & age (y) All Game Species #35 (37) #6 (35) # 1 (17) # 18 (17) #9 (21) #62 (23) #98 (16) other hunters (n = 79) n a) Mammals Pocket gopher 6 2 3 8 34 53 Paca 10 5 6 1 25 47 Agouti 4 1 8 1 1 1 19 35 Coati 58 29 13 6 2 1 58 167 White-lipped peccary 3 Collared 1 7 4 1 1 26 40 peccary Brocket deer 1 1 14 16 White-tailed deer 4 1 19 24 Total mammals" 73 50 38 10 12 195 385 b) Birds Thicket tinamou Great curassow Plain chachalaca 12 28 24 13 23 5 13 8 13 55 167 Ocellated turkey 1 5 6 Total birds' 12 11 6 30 26 16 25 73 199 Grand total' 85 61 44 40 31 28 27 268 584 Total weight (kg)' 274.5 475.5 165.5 64.1 24.3 45.0 35.8 1615.5 2700.1 Mean prey weight (kg) '' 3.2 7.8 3.8 1.6 0.8 1.7 1.3 6.2 4.7 SE" 0.28 1.42 0,68 0.61 0.25 0.75 0.74 0.60 0.33 n' 85 61 44 40 31 27 27 261 576 ' Includes ^ Includes specimens with and without weight measurements, only specimens with weight measurements. 250 APPENDIX F CATALOGUE OF GAME AND NONGAME ANIMALS COLLECTED AT EJIDO X-HAZIL Y ANEXOS DURING 1989-1990, INCLUDING DONATIONS BY LOCAL RESIDENTS, ANIMALS FOUND ALONG THE ROAD, AND GAME TAKEN BY HUNTERS Scientific Name Common Name Quantity A) Mammals Didelphis spp. Chiroptera (8 spp.) Ateles geojfroyi Tamandua mexicana Dasypus novemcinctus Sciurus deppei Sciurus yucatanensis Sciurus sp. Orthogeomys hispidus Agouti paca Dasyprocta punctata Urocyon cinereoargenteus Procyon lotor Nasua nasua Potos flavus Eira barbara Conepatus semistriatus Felis concolor Felis onca Felis pardalis Felis wiedii Felis yagouaroundi Tayassu pecari Tayassu tajacu Odocoileus virginianus Mazama americana Subtotal Mammals Opossum' 27 Bats 18 Spider monkey 1 Tamandua 2 Annadillo 8 Squirrel"'" 6 Squirrel"'' 4 Squirrel"'' 3 Pocket gopher"'' 52 Paca" 47 Agouti*" 35 Grey fox 2 Raccoon 4 Coatimundi'' 167 Kinkajou 4 Tayra 1 Skunk 1 Puma 1 Jaguar" 1 Ocelot" 4 Margay" 6 Yagouaroundi 1 White-lipped peccary" 3 Collared peccary" 40 White-tailed deer" 24 Brocket deer" 16 478 252 253 Scientific Name Common Name Quantity B) Birds Crypturellus cinnamomeus Buteo nitidus Ortalis vetula Crax rubra Agriocharis ocellata Columba flavirostris Columba speciosa Zenaida asiatica Leptotila verreaiixi Aratinga astec Amazona albifrons Piaya cayana Chordeiles minor Ciccaba virgata Ramphastos sulfuratus Campephilus giiatemalensis Subtotal Birds C) Fish Cichlasoma urophthalmus Subtotal Fish D) Reptiles Bothrops sp. Micrurus sp. Unidentified Subtotal Reptiles Thicket tinamou 13 Gray hawk 1 Plain chachalaca"'' 168 Great curassow'' 13 Ocellated turkey'' 6 Red-billed pigeon"'" 22 Scaled pigeon''' 1 White-winged dove''' 1 White-tipped dove*'' 1 Aztec parakeet' 2 White-fronted parrot' 7 Squirrel cuckoo 1 Common nighthawk 1 Mottled owl 1 Keel-billed toucan 2 Pale-billed woodpecker 1 Unidentified Fer de lance Coral snake Unidentified lizard 241 8 8 1 1 1 3_ 730 Grand total ' Animal for which data were not collected during the entire study period. '' Game and nongame animals taken at the village of X-Hazil Sur, Quintana Roo, Mexico, during April 1989-November 1990. APPENDIX G PERCENT FREQUENCY OF THE GAME KILL SITES IN THE FOUR MAIN VEGETATION TYPES FOR THE GAME TAKEN BY HUNTERS AT X-HAZIL SUR Percentage of game kill sites by vegetation type Game species Plots & Gardens (6.1%)' Rarly Secondary Forest (5.2%) Late Secondary Forest (88.5%) Other (0.2%) n a) Manunals Pocket gopher 41.5 52.8 5.7 0.0 53 Paca 34.0 31.9 34.0 0.0 47 Agouti 48.6 20.0 31.4 0.0 35 Coati 46.7 8.4 44.9 0.0 167 White-lipped peccary 0.0 0.0 100.0 0.0 3 Collared peccary 25.0 17.5 57.5 0.0 40 Brocket deer 37.5 12.5 50.0 0.0 16 White-tailed deer 62.5 20.8 12.5 4.2 26 % Mammals combined 42.6 20.3 36.9 0.3 385 b) Birds Thicket tinamou 0.0 23.1 76.9 0.0 13 Great curassow 7.7 0.0 92.3 0.0 13 Plain chachalaca 7.2 86.8 6.0 0.0 167 Ocellated turkey 50.0 16.7 33.3 0.0 6 % Birds combined 8.0 74.9 17.1 0.0 199 % All game species combined 30.8 38.9 30.1 0.2 584 ' Values in ( ) indicate the percentage of the study area in 552.95 km-. each vegetation type. Total area = 255 APPENDIX H SUMMARY OF CENSUS RESULTS AND CHARACTERISTICS OF 12 TRANSECTS IN THREE SUCCESSIONAL STAGES OF FOREST Transect length Transect succes- sional Number of censuses Total duration censuses Average census walking rate Number of sightings Transect number (m) stage" Sunrise Sunset (h) (km/h) Mammals Birds 1^ 1600 LSF w/o G 3 3 9.42 1.02 3 1 2 1960 LSF w/oG 5 6 21.20 1.02 14 9 3 2000 LSF w/o G 5 5 19.23 1.04 14 2 4 2000 LSF w/oG 5 5 20.33 0.98 8 5 5 2000 C/ESF 5 5 20.93 0.96 11 14 6 2000 LSF w/G 5 6 22.42 0.98 12 1 7 2000 LSF w/G 5 6 23.00 0.96 33 0 8 2000 LSF w/G 5 6 21.97 1.00 15 8 9 2000 C/ESF 5 6 24.23 0.91 7 13 10 2000 C/ESF 5 5 20.53 0.97 6 16 11 2130 C/ESF 5 5 20.75 1.03 9 13 12 2000 LSF w/oG 5 5 20.25 0.99 18 8 Total 58 63 244.27 150 90 Forest successional stages; LSF w/o G = Late Secondary Forest without Gardens (41 censuses; total distance censused = 81.56 km); LSF w/ G = Late Secondary Forest with Gardens (39 censuses; total distance censused = 75.60 km); C/ESF = Combined/Early Secondary Forest (41 censuses; total distance censused = 83.30 km). Censuses along Transect # 1 were discontinued after six censuses due to problems with a local resident. 257 APPENDIX I CENSUS DATA FORM JORGENSON AND CO., ANIMAL CENSUS DATA FORM 1. Date (month/day /year)? / / 2. Census number? 3 . Transect number (1-12) ? 4. Transect type? 5. Transect length? 6. Direction of census? 1--West to east 2~East to west 7. Principal observer (data collector)? 1-JPJ, 2-AMV, 3-ABdeJ, 4- , 5- 8. Guide/field assistant? 1-JPA, 2-RUCh, 3-JHBX, 4-ABdeJ, 5-JPJ, 6-DSCh, 8-GGP, 9 Other 9. Time census ended? h 12. Time of return? h 10. Time census started? h 13. Time of departure? h 11. Duration of census? h:m 14. Total time spent? h:m Weather conditions at the beginning of the census: 15. Temperature? 16. Wind? l--none/little, 2--moderate, 3--strong (difficult to hear animal noises) 17. Direction of wind (0°-360°)? 18. Sky? (l-clear, 2"moderately cloudy, 3~cloudy 19. Rain during census? (l~none, 2--light/moderate, 3--strong) Condition of the moon during census: 20. Phase? [0-14 (days in the sky), 0=new moon and 14=full moon] 21. Brighmess? (1-none/little, 2-moderate, 3-very bright) 22. If transect originates at a garden, condition of the garden? 2-milpa (kol) [2.1-new clearing (tumbem kol), 2.1-planting (pa'a kal), 2.3~post harvest {sin ho chik), 2.4-fallow (sa ka), 4-early secondary forest (cleared) (hub che), 5-late secondary forest (not cleared) {sak al che), 9-other 23. Number (quantity) of animals observed? 23. 1 . Record number of observed animals Speed of census: 24. m/min: m/ min= m/min 24.1. m/h: h/min x 60 = m/h 259 260 25. Number (sighting number) of animal observed? 26. Transect number (li) 26.1. Perpendicular distance from the transect to the animal observed? (xi)? 27. Distance between observer and animal (ri)? ^m 28. Angle between observer and animal (oi, 0 < x < 180)? 29. Direction of the animal? 30. Direction of transect? 3 1 . Time animal was observed? h 32. Position of observation of animal within the transect (0 < x < 2130 m)? 33. Type of vegetation where animal was observed? l~vegetable garden (kan che), 2--milpa (kol) [2.1--new clearing (tumbem kol), 2.2--Planting (pa'akal), 2. 3. -post harvest, (sin ho chik), 2.4-fallow {sa ka), 3-family garden (-), 4-early secondary forest {hub che), 5--late secondary forest (no kux kax, ka na kax), 5.1--bumed (to ko che), 5.2~low late secondary forest [somewhat cleared] (ka bu kax), 6-low late secondary forest [not cleared] (sak al che), 7--wetland (hak kan), 9-other 34. Scientific name of animal observed? 35. Common Spanish/Maya name of animal observed? 36. Sex of animal observed? (l--male, 2--female, 3--unknown) 37. Age of animal observed? (1-adult, 2--subadultyyoung, 3--new bom/offspring, 9--other 38. Behavior of animal observed during observation? (l--resting, 2~eating, 3-walking/running/flying, 9--other_ If eating: 39. Scientific name of plant? 40. Spanish/Maya common name of plant? 41. Height of animal observed (0 = on ground)? m If not on the ground: 42. Scientific name of plant where animal was observed? 43. Spanish/Maya conunon name? 44. Type of observation (initial factor that attracted observer's attention)? (l~visual, 2~signs (feces, tracks, nest, etc.) Notes: APPENDIX J MEAN SIGHTING DISTANCES OF WILDLIFE (M), SUMMARIZED BY SPECIES, GROUPS, AND GAME AND NONGAME BIRDS AND MAMMALS CENSUSED ALONG 12 TRANSECTS IN THREE SUCCESSION AL STAGES OF FOREST AT EJIDO X-HAZIL Y ANEXOS, QUINT AN A ROO, MEXICO, DURING 1990 (D.F. = 2 THROUGHOUT) Successional stages x' approxi- mation Late Secondary Forest without Gardens Late Secondary Forest with Gardens Combined/ Early Secondary Forest Taxa" x' SE X SE jc SE P a) Species Squirrels 13.1 (1.2) 11.9 (1.3) 16.1 (3.5) 0.2866 0.8665 Coati*' 12.0 (6.0) 25.6 (2.9) ... (...) 2.6305 0.1048 Kinkajou 17.5 (2.4) 14.5 (1.9) 11.2 (1.2) 1.5109 0.4698 Plain chachaiaca* 11.3 (2.5) 13.9 (3.1) 12.7 (0.8) 1.4101 0.4941 b) Groups Marsupials & 8.8 (1.6) 10.8 (2.9) 10.6 (3.8) 0.1079 0.9475 Edentates Rodents 13.2 (1.0) 14.7 (1.4) 15.8 (2.8) 0.2731 0.8723 Carnivores 17.2 (2.1) 20.2 (2.0) 16.1 (3.0) 1.3694 0.5042 Artiodactyls 29.1 (10.8) 15.6 (11.6) 17.2 (2.5) 1.3849 0.5004 Birds 14.2 (2.8) 13.2 (2.2) 12.8 (0.7) 0.4826 0.7856 c) Game versus nongame species' Game mammals 16.0 (3.0) 22.0 (2.0) 16.2 (1.6) 3.3406 0.1882 Nongame 14.5 (1.2) 12.6 (1.1) 14.8 (2.1) 1.3309 0.5140 mammals Game birds 11.0 (1.9) 12.8 (2.5) 13.2 (0.8) 1.9881 0.3701 Nongame birds 36.4 (13.6) 16.8 (...) 10.5 (1.4) 3.5857 0. 1665 " See Appendix L for common and scientific names. '' X = The constant 10,000 multiplied by the average sighting distance for the four replicate transects. ' * = Game species. See Chapter 3 for additional information about game species. 262 APPENDIX K NUMBER OF SIGHTINGS/100 KM AND TOTAL NUMBER OF SIGHTINGS (N), SUMMARIZED BY SPECIES, GROUPS, AND GAME AND NONGAME BIRDS AND MAMM.\LS Taxa' Successional stages Late Secondary Forest without Gardens Late Secondary Forest with Gardens Combined/ Early Secondary Forest MAMMALS a) Marsupials & Edentates Opossum Tamandua Nine-banded armadillo 2.5 (2)" 2.6 (2) 0.0 (0) 4 0.0 (0) 0.0 (0) 1.2 (1) 1 2.5 (2) 7.9 (6) 6.0 (5) 13 Subtotal 4.9 (4) 10.6 (8) 7.2 (6) 18 b) Rodents Squirrels' Pocket gopher*'* Paca* Agouti* 27.0 (22) 17.2 (13) 14.4 (12) 47 0.0 (0) 0.0 (0) 1.2 (1) 1 2.5 (2) 7.9 (6) 0.0 (0) 8 6.1 (5) 6.6 (5) 2.4 (2) 12 Subtotal 35.6 (29) 31.7 (24) 18.0(15) 68 c) Carnivores Gray fox Coati* Kinkajou Mustelids' 0.0 (0) 1.3 (1) 3.6 (3) 4 2.5 (2) 19.8 (15) 0.0 (0) 17 9.6 (16) 15.9(12) 3.6 (3) 31 1.2 (1) 1.3 (1) 1.2 (1) 3 Subtotal 23.3 (19) 38.4 (29) 8.4 (7) 55 d) Artiodactyls Collared peccary* 0.0 (0) 0.0 (0) 2.4 (2)f 2 Brocket 1.2 (1) 1.3 (1) 2.4 (2) 4 deer* White-tailed deer* 1.2 (1) 1.3 (1) 1.2 (1) Subtotal 2.5 (2) 2.6 (2) 6.0 (5) Total mammals 66.2 (54) 83.3 (63) 39.6 (33) 150 264 265 Successional stages Taxa^ Late Secondary Forest without Gardens Late Secondary Forest with Gardens Combined/ Early Secondary Forest n BIRDS Tinamou** 6.1 (5) 4.0 (3) 4.8 (4) 12 Plain chachalaca* 19.6 (16) 6.6 (5) 51.6 (43) 64 Ocellated turkey* 0.0 (0) 1.3 (1) 1.2 (1) 2 Columbids'' 3.7 (3) 0.0 (0) 9.6 (8) 11 White-fronted parrot 0.0 (0) 1.3 (1) 0.0 (0) 1 Total birds 29.4 (24) 13.2 (10) 67.2 (56) 90 GAME VERSUS NONGM-IE SPECIES" Game mammals 13.5 (11) 37.0 (28) 9.6 (8) 47 Nongame mammals 52.7 (43) 46.3 (35) 30.0 (25) 103 Game birds 25.7 (21) 11.9 (9) 57.6 (48) 78 Nongame birds 3.7 (3) 1.3 (1) 9.6 (8) 12 Total (78) (73) (89) 240 See Appendix L for common and scientific names. (n) = Number of sightings. Composed of Sciurus deppei, S. yucatanensis, and an unidentified squirrel (not a new taxon). * = Game species. See Chapter 3 for additional information about game species. Tayra and skunk. One sighting of a single individual and one sighting of nine individuals in a single group. Slaty-breasted and thicket tinamou. Scaled pigeon and white-tipped dove. APPENDIX L COMMON AND SCIENTIFIC NAMES OF BIRDS AND MAMMALS SIGHTED AT EJIDO X- HAZIL Y ANEXOS DURING 121 CENSUSES IN 1990 Scientific name' Family English common name a) Mammals Didelphis spp. Tamandua mexicana Dasypus novemcinctus Sciurus deppei Sciurus yucatanensis Orthogeomys hispidus*^ Agouti paca* Dasyprocta punctata* Urocyon cinereoargenteus Nasua nasua* Potos flavus Eira barbara Conepatus semistriatus Tayassu tajacu* Mazama americana* Odocoileus virginianus* Didelphidae Myrmecophagidae Dasypodidae Sciuridae Sciuridae Geomyidae Agoutidae Dasyproctidae Canidae Procyonidae Procyonidae Mustelidae Mustelidae Tayassuidae Cervidae Cervidae Opossum Tamandua Nine-banded armadillo Squirrel Squirrel Pocket gopher Paca Agouti Gray fox Coaa Kinkajou Tayra Skunk Collared peccary Brocket deer White-tailed deer b) Birds Crypturellus boucardi Crypturellus cinnamomeus* Ortalis vetula* Agriocharis ocellata* Columba speciosa Leptotila verreauxi Amazona albifrons Tinamidae Tinamidae Cracidae Meleagrididae Columbidae Columbidae Psittacidae Slaty-breasted tinamou Thicket tinamou Plain chachalaca Ocellated turkey Scaled pigeon White-tipped dove White-fronted parrot Primary sources for scientific names: Leopold (1977) and Peterson and Chalif (1973). * = Game species. See Chapter 3 for additional information about game species. 267 APPENDIX M DATA FORM: STOMACH CONTENTS ANALYSIS STOMACH CONTENTS ANALYSIS JORGENSON AND COMPANY: 1990 Field number? Maya common name of animal? Dale of collection (month/date/year) ? Scientific name of animal? Item Maya common name Item scientific name Plant/animal part consumed (ml) no. Fruits & seeds Leaves Animal matter^ Other" Total 1 2 3 4 5 6 7 8 9 Total volume (ml) Total volume of the original sample (ml) Original identification by? Date of identification (month/date/year)? / / Confirmation by? Date of confirmation (month/date/year)? / / ° Animal matter includes hair, snail shells, and insect antennae. ^ Other includes tubers, roots, stems, plastic, rocks, and soil. 269 APPENDIX N PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF PACA FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 36 STOMACHS (TOTAL VOLUME ANALYZED = 1886.5 ML) COLLECTED FROM PACAS TAKEN BY MAYA HUNTERS AT X-HAZIL SUR, QUINTANA ROO, MEXICO, DURING JUNE 1989-OCTOBER 1990 Food item" Family Local names'" % Occ. % Vol. Manilkara zapota (L.) v. Royen Sapotaceae Ya Chicozapote 47.2 26.6 Zea mays L. * Gramineae Nail Mail Com 25.0 15.4 Dendropanax arboreus (L.) Dene. & Planch. Araliaceae Sac chacd 22.2 15.3 Cucurbita moschata Duch.* Cucurbitaceae Ku'um Calabaza Squash 22.2 14.9 Byrsonima CTassifolia (L.) HBK. Malpighiaceae Chi Nance duke 22.2 1.6 Unidentified plant — ... 19.4 < 0.0 Coccoloba acapulcensis Standi. Polygonaceae To 'yub 16.7 8.6 Misc. mammal tissue (skin & hair) Class Mammalia ... 16.7 < 0.0 Cissus sicyoides L. Vitaceae Tacant 11.1 8.4 Pithecellobium sp. Leguminosae Choc che 11.1 0.6 Misc. mammal tissue (feces) Class Mammalia ... 11.1 0.1 Diospyros sp. Ebenaceae sun & Uh chu che 8.3 0.1 Brosimum alicastrum Swartz Moraceae Ox Ramon 8.3 < 0.0 Psychotria sp. Rubiaceae Kanan verde 8.3 < 0.0 Dipholis salicifolia (L.) A. DC. Sapotaceae Ik che <4 Tzi tzi yah 5.6 0.3 Unidentified plant ... Chit ku 'uk 5.6 0.1 Cydista potosina (Schum. & Loes.) Loes. Bignoniaceae E ki xil 5.6 < 0.0 271 272 Food item' Family Local names'" % Occ. % Vol. Unidentified insect Order Orthoptera Grillo 5.6 < 0.0 Manihot esculenta Crantz.* Euphorbiaceae Dzi'im Yuca Yuca 2.8 2.9 iDomoea batatas (L.) Lam.* Convolvulaceae Is Camote 2.8 2.7 Byrsonima bucidaefolia Standi. Malpighiaceae Sac paa Nance agrio 2.8 2.4 Pouteria unilocularis (Donn. Smith) Baehni Sapotaceae Chac yd 2.8 0.1 Unidentified terrestrial worm Class Oligochaeta Lu kum Lombriz 2.8 0.1 Sabal vara C. Wright ex Beccari Palmae Bom & Xa 'an Huano 2.8 < 0.0 Coccoloba SD. Polygonaceae Chich bob 2.8 < 0.0 Unidentified plant Euphorbiaceae Naap che 2.8 < 0.0 Unidentified slug Class Gastropoda Gusano 2.8 < 0.0 Unidentified plant Unidentified Zu uc Zacale' 2.8 < 0.0 ' * = Crop species. '' Maya names precede Spanish names. English names included for crops only. ' Zu uc/Zacate (unidentified species) apparently is not the same as Ak nom/Zacate fScleria lithospermum (L.) Swartz]. APPENDIX O PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF AGOUTI FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 26 STOMACHS (TOTAL VOLUME ANALYZED = 1167.5 ML) Food item" Family Local names'" % Occ. % Vol. Manilkara zaoota (L.) v. Royen Sapotaceae Ya Chicozapote 38.5 8.7 Byrsonima crassifolia (L.) HBK. Malpighiaceae Chi Nance dulce 30.8 10.2 Zea mavs L. * Gramineae Nail Mail Com 23.1 25.5 Ipomoea batatas (L.) Lam.* Convolvulaceae Is Camote Sweet potato 15.4 20.2 Cucurbita moschata Duch.* Cucurbitaceae Ku um Calabaza Squash 15.4 13.6 Brosimum alicastrum Swartz Moraceae Ox Ramon 11.5 < 0.0 Leguminosae' Leguminosae Chok Che 7.7 8.7 Coccoloba sp. Polygonaceae Chich bob 7.7 3.9 DiosDvros sp. Ebenaceae Uh chu che 7.7 0.3 Misc. mammal tissue (skin &hair) Class Mammalia ... 7.7 < 0.0 Unidentified plant ... ... 7.7 < 0.0 Sabal vapa C. Wright ex Beccari Palmae Bom & Xa 'an Huano 7.7 < 0.0 Celtis iauanaea Qacq.) Sarg. Ulmaceae Chich mu uk 3.8 4.3 Byrsonima bucidaefolia Standi. Malpighiaceae Sac paa Nance agrio 3.8 2.8 Passiflora sp. Passifloraceae Ton ton tzimini 3.8 1.4 Piper auritum HBK. Piperaceae Ma ku lam 3.8 0.3 Unidentified slug Class Gastropoda Gusano 3.8 0.1 274 275 % % Food item" Family Local names"' Occ. Vol. Scleria lithosDerma (L.) Cyperaceae Ak nom 3.8 < 0.0 Swartz*'' Zacate Coccoloba acaoulcensis Polygonaceae To 'yub 3.8 < 0.0 Standi. — Nectaiidra saneuinea Rottb. Lauraceae Ho choc 3.8 < 0.0 Dendropanax arboreus (L.) Araliaceae Sac chacd 3.8 < 0.0 Dene. & Planch. — ' * = Crop species. *■ Maya names precede Spanish names. English names included for crops only. ' Either Caesalpinia sp. or Pithecellobium sp. '' Ak nom/Zacate apparendy is not the same as Zu uc/Zacate (unidentified species). APPENDIX P PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF COATI FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 129 STOMACHS (TOTAL VOLUME ANALYZED = 4678.5 ML) Food item' Family Local names'" % Occ. % Vol. Unidentified slug Class Gastropoda Gusano 50.4 3.8 Unidentified insect Order Orthoptera Grillo 44.2 1.1 Zea mavs L.* Gramineae Nail Maiz Com 40.3 53.5 Manilkara zaoota (L.) v. Royen Sapotaceae Ya Chicozapote 28.7 19.0 Unidentified snail Class Gastropoda Caracal 20.2 < 0.0 Vitex gaumeri Greenm. Verbenaceae Yaxnic 14.0 4.7 Coccoloba acaoulcensis Standi. Polygonaceae To'yub 9.3 4.0 Brosimum alicastrum Swartz Moraceae Ox Ramon 7.8 5.8 Unidentified slug Class Gastropoda Nokol Gusano 7.0 0.5 Unidentified terrestrial worm Class Oligochaeta Lu kum Lombriz 7.0 0.1 Unidentified insect Class Orthoptera Mas Grillo 6.2 0.2 Unidentified plant Unidentified — 6.2 < 0.0 Unidentified insect Class Orthoptera Kon choch Grillo 5.4 0.4 Unidentified plant Unidentified Chit ku'uk 5.4 < 0.0 Misc. mammal tissue (skin &hair) Class Mammalia — 4.7 1.6 Unidentified insect Class Orthoptera Xo Cucaracha 3.9 0.3 Pouteria unilocularis (Donn. Smith.) Baehni Sapotaceae Chak yd 3.1 1.0 Bvrsonima bucidaefolia Standi. Malpighiaceae Sac paa Nance agrio 3.1 0.4 277 278 Food item' Family Local names'" Occ. Vol. Unidentified insects (various) Class Orthoptera Unidentified plant Unidentified reptile Unidentified Class Reptilia Dipholis salicifolia (L.) A. Sapotaceae DC. Kuklin (?) Hierba mora Largartillo Ik die & Tzi tzi yah 3.1 0.1 2.3 0.7 2.3 0.1 1.6 0.8 Coccoloba sp. Polygonaceae Chich bob 1.6 0.4 Unidentified plant Unidentified reptile Sabal yapa C. Wright ex Beccari Unidentified mushroom Unidentified insect Lysiloma latisiliqua (L.) Benth. Leguminosae' Diospyros sp. Unidentified tarantula Unidentified amphibian Carica papaya L. Unidentified Class Reptilia Palmae Class Basidiomycetes Class Odonata Celtis iguanaea (Jacq.) Sarg. Ulmaceae Leguminosae Leguminosae Ebenaceae Class Arachnoidea Class Amphibia Cariacaceae Dendropanax arboreus (L. ) Araliaceae Dene. & Planch. Chico low 1.6 0.2 Kan 1.6 0.2 Culebra Bom & Xa 'an 1.6 0.1 Hiiano ... 1.6 0.1 Hongo Turis 1.6 < 0.0 Libelula Chich mu uk 1.6 < 0.0 Tzalam 1.6 < 0.0 Chok che 1.6 < 0.0 sua & 0.8 0.8 Uh chu che 0.8 < 0.0 Tarantula ... 0.8 < 0.0 Rana/Sapo Pu chich & Puut 0.8 < 0.0 Sac chaca 0.8 < 0.0 279 Food item' Family Local names'" % Occ. % Vol. Unidentified plant Unidentified Can pe tzu 0.8 < 0.0 Dalbergia glabra (Miller) Standi. Leguminosae Muk 0.8 < 0.0 Bvrsonima crassifolia (L.) HBK. Malpighiaceae Chi Nance dulce 0.8 < 0.0 Metooium brownei (Jaca.) Urban Anacardiaceae Chechem 0.8 < 0.0 Unidentified plant Sapindaceae Tzool 0.8 < 0.0 Cvdista potosina (Schum. & Loes.) Loes. Bignoniaceae E ki xil 0.8 < 0.0 Ficus SD. Moraceae Ha huay & Sac ha huay 0.8 < 0.0 Acacia glomerosa Benth. Leguminosae Ju pich 0.8 < 0.0 Unidentified scorpion Class Arachnoidea Alacran 0.8 < 0.0 Unidentified fly Order Diptera Tabano 0.8 < 0.0 Unidentified slug Order Gastropoda Piix Gusano 0.8 < 0.0 Unidentified bird Class Aves Pdjaro 0.8 < 0.0 Plastic, Rocks, & Soil — ... 0.8 < 0.0 ' * = Crop species. '' Maya names precede Spanish names. English names included for crops only. ' Either Caesalpinia sp. or Pithecellobium sp. APPENDIX Q PERCE^^^ occurrence (% oco and volume (% vol.) of white-lipped peccary FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF THREE STOMACHS (TOTAL VOLUME ANALYZED = 208.0 ML) Food item' Family Local names'" % Occ. % Vol. Brosimum alicastrum Swartz Moraceae Ox Ramon 100.0 &7J Unidentified plant Unidentified Can pe tzu 100.0 4.6 Psvchotria sp. Rubiaceae Kanan verde 100.0 1.0 Acoelorrhaphe wrightii (Griseb. & Wendi.) Wendl. ex Beccari Palmae Taxiste 66.7 3.8 Carica papava L. Cariacaceae Pu chich & Puut Papaya 66.7 1.4 Cissus sicvoides L. Vitaceae Ta cani 66.7 1.0 Unidentified plant Unidentified ... 66.7 < 0.0 Malmea depressa (Baillon) R.E. Fries Annonaceae Elemuy 33.3 0.5 Vitex eaumeri Greenm. Verbenaceae Yaxtiic 33.3 < 0.0 Manilkara zapota (L.) v. Roven Sapotaceae Ya Chicozapote 33.3 < 0.0 Dalbergia glabra (Miller) Standi. Leguminosae Muk 33.3 < 0.0 Hamelia patens Jacq. Rubiaceae Scanam Kanan rojo 33.3 < 0.0 Trophis racemosa (L.) Urban Moraceae Sak ox 33.3 < 0.0 Unidentified mushroom Class Basidiomycetes Hongo 33.3 < 0.0 Unidentified insect Order Odonata Turis Libelula 33.3 < 0.0 Unidentified slug Class Gastropoda Gusano 33.3 < 0.0 Misc. mammal tissue (skin & hair) Class Mammalia ;;; 33.3 < 0.0 " * = Crop species. '' Maya names precede Spanish names. English names included for crops only. 281 APPENDIX R PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF COLLARED PECCARY FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 29 STOMACHS (TOTAL VOLUME ANALYZED = 1829.0 ML) Food item^ Family Local names'' % Occ. % Vol. Psvchotria sp. Rubiaceae Brosimum alicastrum Swartz Moraceae Carica papaya L. Cariacaceae Misc. mammal tissue (skin & Class Mammalia hair) Unidentified plant Bvrsonima bucidaefolia Standi. Cissus sicyoides L. Metopium brownei (Jacq.) Urban Vitex gaumeri Greenm . Diospyros sp. Unidentified Malpighiaceae Vitaceae Anacardiaceae Verbenaceae Cucurbita moschata Duch.* Cucurbitaceae Ebenaceae — 79.3 23.8 Kanan verde Ox 75.9 15.2 Ramon Pu chich & Pmt 48.3 11.2 Papaya ... 48.3 < 0.0 ... 31.0 0.2 Sac paa 27.6 11.7 Nance agrio Ta cani 27.6 0.9 Chechem 20.7 1.4 Yamic 20.7 0.4 Ku 'urn 13.8 5.9 Calabaza Squash Silil & Uh chuche 13.8 3.1 Unidentified plant Zeamays L.* Unidentified Gramineae Ipomoea batatas (L.) Lam.* Convolvulaceae Can pe tzu Nail Mail Com Is Camote Sweet potato Desmoncus auasillarius H. Bartlett Palmae Bayal Unidentified slug Order Gastropoda Giisano Thevetia eaumeri Hemsl. Apocynaceae Akitz 13.8 10.3 10.3 3.1 12.8 6.6 10.3 2.8 10.3 < 0.0 10.3 < 0.0 283 284 Food item" Family Local names'' % Occ. % Vol. Dalbergia glabra (Miller) Standi. Leguminosae Muk 10.3 < 0.0 Malmea deoressa (Baillon) R.E. Fires Annonaceae. Elemuy 10.3 < 0.0 TroDhis racemosa (L.) Urban Moraceae Sak ox 6.9 < 0.0 Unidentified plant Unidentified Chico low 6.9 < 0.0 loomoea triloba L. "^ Convolvulaceae Is a kil 6.9 < 0.0 Sarcostemma bilobum H. & A. Asclepiadaceae Tzotz ca bi 3.4 0.3 Scleria lithosperma (L.) Swartz*' Cyperaceae Ak nom Zacate 3.4 0.2 Sabal vapa C. Wright ex Beccari Palmae Bom & Xa'an Huano 3.4 0.1 Bauhinia divaricata L. Leguminosae Tzu run toe 3.4 < 0.0 Dipholis salicifolia (L.) A. DC. Sapotaceae Ik Che & Tzi tzi yah 3.4 < 0.0 Bvrsonima CTassifolia (L.) HBK. Malpighiaceae Chi Nance duke 3.4 < 0.0 Dendropanax arboreus (L.) Dene. & Planch. Araliaceae Sac chacd 3.4 < 0.0 Nectandra sp. Lauraceae Ho choc 3.4 < 0.0 Unidentified plant Class Basidiomycetes Hongo 3.4 < 0.0 Unidentified plant Unidentified Ni te 3.4 < 0.0 Caesalpinia vucatanensis Greenm. Leguminosae Tah kin che 3.4 < 0.0 Coccoloba acapulcensis Standi. Polygonaceae To 'yub 3.4 < 0.0 Unidentified insect Class Orthoptera Kuklin (?) 3.4 < 0.0 285 % % Food item' Family Local names'" Occ. Vol. Unidentified reptile Class Reptilia Tortuga 3.4 < 0.0 Unidentified reptile Class Reptilia Kan Culebra 3.4 < 0.0 Plastic, Rocks, & Soil — ... 3.4 < 0.0 * * = Crop species. *■ Maya names precede Spanish names. English names included for crops only. ' Ak nom/Zacate apparently is not the same as Zu uc/Zacate (unidentified species). APPENDIX S PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF BROCKET DEER FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF EIGHT STOMACHS (TOTAL VOLUME ANALYZED = 446.0 ML) Food item" Family Local names'" Occ. % Vol. Psychotria sp. Brosimum alicastrum Swartz Eugenia sp. Unidentified plant Trophis racemosa (L.) Urban Manilkara zapota (L.) v. Royen Sarcostemma bilobum H. & A. Galactia striata (Jacq.) Urban Sapindaceae Bauhinia divaricata L. Carica papaya L. Thevetia gaumeri Hemsl. Unidentified plant Trichilia hirta L. Acoelorrhaphe wrightii (Griseb. & Palmae Wendl.) Wendl. ex Beccari Swartzia cubensis (Britton & Wils.) Leguminosae Standi. Byrsonima bucidaefolia Standi. Rubiaceae Kanan verde 87.5 58.3 Moraceae Ox Ramon 62.5 1.6 Myrtaceae Sip che 50.0 0.4 Unidentified — 50.0 0.3 Moraceae Sak ox 37.5 1.1 Sapotaceae Ya Chicozapote 37.5 0.7 Class Mammalia ... 37.5 < 0.0 Asclepiadaceae Tzotz cabi 25.0 9.2 Leguminosae Ka xa yuk 25.0 8.1 Sapindaceae Tzool 25.0 1.6 Leguminosae Tzu run tok 25.0 0.9 Cariacaceae Pu chich & Puut Papaya 25.0 0.2 Apocynaceae Akitz 25.0 < 0.0 Class ... 25.0 < 0.0 Basidiomycetes Hongo Meliaceae Chili cha han 12.5 11.7 Palmae Taxiste 12.5 2.2 Leguminosae Ka taal ox 12.5 0.9 Malpighiaceae Sac paa Nance agrio 12.5 0.7 287 288 Food item' Family Local names'" % Occ. % Vol. Unidentified plant Unidentified Chit ku 'uk 12.5 0.4 Unidentified plant Unidentified Chico low 12.5 0.4 Acanthocereus oentagonus (L.) Britton & Rose Cactaceae Pi ta ha ya 12.5 0.4 Colubrina sp. Rhamnaceae Bu klun che 12.5 0.3 Metopium brownei (Jacq.) Urban Anacardiaceae Chechem 12.5 0.2 Scleria hthosperma (L.) Swartz*' Cyperaceae Ak nom Zacate 12.5 0.2 Nectandra sanguinea Rottb. Lauraceae Ho choc 12.5 < 0.0 Vitex gaumeri Greenm. Verbenaceae Yaxnic 12.5 < 0.0 Unidentified plant' Unidentified Zu uc Zacate 12.5 < 0.0 Unidentified plant Unidentified Can pe tzu 12.5 < 0.0 Cvdista potosina (Schum. & Loes.) Loes. Bignoniaceae E ki xil 12.5 < 0.0 Ipomoea triloba L. Convolvulaceae Is a kil 12.5 < 0.0 Lvsiloma latisiliaua (L.) Benth. Leguminosae Tzalam 12.5 < 0.0 Plastic, Rocks, & Soil ... ... 12.5 < 0.0 ' * = Crop species. *■ Maya names precede Spanish names. English names included for crops only. ' Ak nom/Zacate apparently is not the same as Zu uc/Zacate (unidentified species). APPENDIX T PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF WHITE-TAILED DEER FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 11 STOMACHS (TOTAL VOLUME ANALYZED = 601.0 ML) % % Vol. Food item' Family Local names'' Occ. Psvchotria sd. Rubiaceae Kalian verde 72.7 42.7 Unidentified plant Unidenufied ... 54.5 0.7 Trophis racemosa (L.) Urban Moraceae Sak ox 45.5 7.7 Brosimum alicastrum Swartz Moraceae Ox Ramon 45.5 1.0 Galactia striata (Jacq.) Urban Leguminosae Ka xa yuk 36.4 5.6 Ipomoea triloba L. Convolvulaceae Is a kil 36.4 0.7 Eugenia sp. Myrtaceae Sip che 36.4 0.6 Misc. mammal tissue (skin & Class Manunalia 36.4 < 0.0 hair) Euphorbia heterophvlla L. Euphorbiaceae Jobon kak 27.3 9.4 Sarcostenuna bilobum H. & A. Asclepiadaceae Tzotz ca bi 27.3 6.8 Momordica charantia L. Cucurbitaceae Yoch way 27.3 0.8 Hamelia patens Jacq. Rubiaceae Scanam Kanan rojo 27.3 0.4 Dalbergia glabra (Miller) Leguminosae Muk 27.3 < 0.0 Standi. ... ChrvsophvUum mexicanum Sapotaceae Chi keh 18.2 < 0.0 Brand, ex Standi. ... Lvsiloma latisiliaua (L.) Benth. Leguminosae Tzalam 18.2 < 0.0 Unidentified plant' Unidentified Zu uc Zacate 9.1 9.4 Phaseolus vulgaris L.* Leguminosae Buul Frijol Kidney bean 9.1 9.0 Byrsonima bucidaefolia Standi. Malpighiaceae Sac paa Nance agrio 9.1 1.7 290 291 % % Vol. Food item' Family Local names'" Occ. Bvrsonima crassifolia (L.) Malpighiaceae Chi 9.1 1.5 HBK. Nance duke Protium copal (Schlecht. & Burseraceae Po on 9.1 0.5 Cham.) Engl. Copal Nyctaginaceae Nyctaginaceae Tatzi 9.1 0.4 Unidentified plant Unidentified Chi lar 9.1 0.3 Ficus sp. Moraceae Ha huay & Sac ha huay 9.1 0.2 Croton glabellus L. Euphorbiaceae Peres cutz Oregono wech 9.1 0.2 Manilkara zapota (L.) v. Roven Sapotaceae Ya Chicozapote 9.1 0.2 Exothea diphvUa (Standi.) Sapindaceae Wa yun kox 9.1 0.1 Lundell ... Unidentified plant Unidentified Can pe tzu 9.1 0.1 Smilax mexicana Griseb. Smilacaceae Koke 9.1 < 0.0 Vitex gaumeri Greenm. Verbenaceae Yaxnic 9.1 < 0.0 Thevetia gaumeri Hemsl. Apocynaceae Akitz 9.1 < 0.0 Carica papaya L. Cariacaceae Pu chich & Puut Papaya 9.1 < 0.0 Diospvros sp. Ebenaceae sun & Uh chu che 9.1 < 0.0 Sapindaceae Sapindaceae Tzool 9.1 < 0.0 Petrea arborea HBK. Verbenaceae Yoch op chimin 9.1 < 0.0 Celtis iguanaea (Jacq.) Sarg. Ulmaceae Chich mu uk 9.1 < 0.0 Unidentified plant Unidentified Chit ku 'uk 9.1 < 0.0 292 % % Vol. Food item" Family Local names'' Occ. Paullinia pinnata L. Sapindaceae Chac tzacan 9.1 < 0.0 Cvdista DOtosina (Schum. & Bignoniaceae E ki xil 9.1 < 0.0 Loes.) Loes. — Mimosa oigra L. Leguminosae Mu tzil 9.1 < 0.0 Amphilophium peniculatus (?) Bignoniaceae Sit wech 9.1 < 0.0 Forchhammeria trifoliata Radlk. Capparidaceae 9.1 < 0.0 Tres marias Unidentified animal (bone) Phylum Chordata 9.1 < 0.0 Plastic, Rocks, & Soil 9.1 < 0.0 " * = Crop species. *■ Maya names precede Spanish names. English names included for crops only. ' Zu uc/Zacate apparently is not the same as Ak nom/Zacate (Scleria lithosperma). APPENDIX U PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF THICKET TINAMOU FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF THREE STOMACHS (ESOPHAGUS, CROP, PROVENTRICULUS, AND GIZZARD; TOTAL VOLUME ANALYZED = 13.0 ML) Food item' Family Local names' % Occ. % Vol. Unidentified plant Unidentified ... 100.0 69.2 ChrvsoDhvllum mexicanum Brand, ex Standi. S^wtaceae Chi keh 66.7 26.9 Galactia striata (Jaca.) Urban Leguminosae Ka xa yuk 66.7 < 0.0 Unidentified insect Class Insecta Kuklin (?) 33.3 3.8 Unidentified plant Unidentified Xeret 33.3 < 0.0 Scleria lithosperma (L.) Swartz*' Cyperaceae Ak nom Zacate 33.3 < 0.0 Sarcostemma bilobum H. & A. Asclepiadaceae Tzotz ca bi 33.3 < 0.0 " * = Crop species. '' Maya names precede Spanish names. English names included for crops only. ' Ak nom/Zacate apparently is not the same as Zu uc/Zacate (unidentified species). 294 APPENDIX V PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF GREAT CURASSOW FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 13 STOMACHS (ESOPHAGUS, CROP, PROVENTRICULUS, AND GIZZARD; TOTAL VOLUME ANALYZED = 694.0 ML) Food item" Family Local names'" % Occ. % Vol. Plastic, Rocks, & Soil ... — 61.5 2.5 Diospvros SD. Ebenaceae sun & Uh chu che 53.8 12.5 Manilkara zaDota (L.) v. Roven Sapotaceae Ya Chicozapote 53.8 5.6 Brosimum alicastrum Swartz Moraceae Ox Ramon 46.2 34.5 Unidentified plant Unidentified — 38.5 1.1 Dipholis salicifolia (L.) A. DC. Sapotaceae Ik Che & Tzi tzi yah 38.5 0.5 Pouteria unilocularis (Donn. Sniitb) Baehni Sapotaceae Cliak ya 23.1 14.9 Leguminosae' Leguminosae Chok che 23.1 3.7 Galactia striata (Jaca.) Urban Leguminosae Ka xa yuk 23.1 0.1 Coccoloba acapulcensis Standi. Polygonaceae To 'yub 15.4 7.9 Phaseolus vulgaris L.* Leguminosae Buul Frijol Kidney bean 15.4 4.2 Sabal vapa C. Wright ex Beccari Palmae Bom & Xa'an Huano 15.4 3.9 Psvchotria sp. Rubiaceae Kanan verde 15.4 0.9 Thevetia gaiuneri Hemsl. Apocynaceae Akitz 15.4 0.2 Metopium brownei (Jacq.) Urban Anacardiaceae Chechem 7.7 2.9 Unidentified plant Unidentified Ko lop 7.7 1.4 Sarcostemma bilobum H. & A. Asclepiadaceae Tzotz ca bi 7.7 1.3 Coccoloba sp. Polygonaceae Chick bob 7.7 0.9 296 297 Food item' Family Local names'' % Occ. % Vol. Unidentified plant Unidentified Ka ni macal 7.7 0.7 Unidentified plant"* Unidentified Zu uc Zacate 7.7 0.1 Momordica charantia L. Cucurbitaceae Yoch way 7.7 0.1 Cucurbita moschata Duch.* Cucurbitaceae Ku'um Calabaza Squash 7.7 < 0.0 Allophylus cominia (L.) Swartz Sapindaceae Ik batch 7.7 < 0.0 Scleria lithosoerma (L.) Swartz*" Cyperaceae Ak nom Zacate 7.7 < 0.0 Unidentified bird Class Aves ... 7.7 < 0.0 Misc. mammal tissue (skin & hair) Class Mammalia ... 7.7 < 0.0 ' * = Crop species. '' Maya names precede Spanish names. English names included for crops only. J Either Caesalpinia sp. or Pithecellobium sp. '^ Zu uc/Zacate (unidentified species) apparently is not the same as Ak nom/Zacate (Scleria lithosperma). APPENDIX W PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF PLAIN CHACHALACA FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF 21 STOMACHS (ESOPHAGUS, CROP, PROVENTRICULUS, AND GIZZARD; TOTAL VOLUME ANALYZED = 329.5 ML) Food item" Family Local names"' % Occ. % Vol. Chrvsophyllum mexicanum Brand, ex Standi. S^otaceae Metopium brownei (Jacq.) Urban Anacardiaceae Sabal vapa C. Wright ex Beccari Palmae Allophylus cominia (L.) Swartz Sapindaceae Trema miCTantha (L.) Blume Unidentified plant Psvchotria sp. Thevetia wrightii Hemsl. Exothea diphylla (Standi.) Lundell Celtis iguanaea (Jacq.) Sarg. Cecropia obtusifolia Bert. Mikania sp. Hamelia patens Jacq . Cestrum sp. (?) Astronium graveolens Jacq . Ficus sp. Eugenia mavana Standi. Ulmaceae Unidentified Rubiaceae Apocynaceae Sapindaceae Scleria lithosperma (L.) Swartz*' Cyperaceae Ulmaceae Moraceae Compositae Rubiaceae Solanaceae Anacardiaceae Moraceae Myrtaceae Chi keh Chechem Bom & Xa'an Huano Ik batch Payux Xeret Kanan verde Akitz Wa yun kox Ak nom Zacate Chich mu uk Guarunio Okin sitkuu Scanam Kanan rojo A ka xiu Ku lin che Ha huay & Sac ha huay Sac lob che 23.8 18.1 23.8 9.4 19.0 17.9 14.3 11.8 14.3 1.8 9.5 5.8 9.5 5.5 9.5 1.2 9.5 0.2 4.8 6.7 4.8 6.1 4.8 4.9 4.8 3.9 4.8 3.0 4.8 2.4 4.8 1.2 4.8 0.2 4.8 < 0.0 299 300 % % Food item" Family Local names'" Occ. Vol. Unidentified plant Unidentified Tzitz 4.8 < 0.0 Unidentified plant Unidentified — 4.8 < 0.0 ' * = Crop species. '' Maya names precede Spanish names. English names included for crops only. ' AJc nom/Zacate apparently is not the same as Zu uc/Zacate (unidentified species). APPENDIX X PERCENT OCCURRENCE (% OCC.) AND VOLUME (% VOL.) OF OCELLATED TURKEY FOOD ITEMS BASED ON AN ANALYSIS OF THE CONTENTS OF FIVE STOMACHS (ESOPHAGUS, CROP, PROVENTRICULUS, AND GIZZAPUD; TOTAL VOLUME ANALYZED = 211.5 ML) Food item' Family Local names'" % Occ. % Vol. Unidentified plant Ficus sp. Cyperaceae Rubiaceae Scleria lithosperma (L.) Swartz*' Psychotria sp. Plastic, Rocks. & Soil Unidentified plant Zea mays L.* Manilkara zaoota (L.) v. Royen Unidentified plant Phaseolus yulgaris L.* Allophvlus cominia (L.) Swartz Byrsonima bucidaefolia Standi. Unidentified plant Unidentified snail Galactia striata (Jacg.) Urban Leguminosae Unidentified Moraceae Ak nom Zacate Zacate Kana verde Unidentified — Gramineae Nail Maiz Com Sapotaceae Ya Chicozapote Unidentified Tzitz Leguminosae Bind Frijol Kidney bean Sapindaceae Ik batch Malpighiaceae Sac paa Nance agrio Unidentified Hierba mora Class Gastropoda Caracal Leguminosae Ka xa vuk Can pe tzu Ha huay & Sac ha huay 60.0 60.0 60.0 60.0 40.0 7.6 5.0 3.3 3.3 24.6 40.0 12.5 40.0 0.5 20.0 18.9 20.0 8.0 20.0 6.4 20.0 4.3 20.0 3.8 20.0 0.7 20.0 0.5 20.0 0.5 Trema micrantha (L.) Blume Ulmaceae Payux 20.0 0.2 Sabal yapa C. Wright ex Beccari Palmae Bom & Xa'an Huano 20.0 < 0.0 • * = Crop species. *■ Maya names precede Spanish names. English names included for crops only. ' Ak norrT/Zacate apparently is not the same as Zu uc/Zacate (unidentified species). 302 APPENDIX Y GARDEN INTERVIEW FORM 1. Record number? 2. Garden number? 3. Date of the interview (m/d/y)? / / 4. Interviewer? 5. Interviewee? 5.3. Did the interviewee plant a garden (1 = yes, 2 = no)? 6. Garden size (number of "mecates")? 7. Number of years that the site has been used as a garden immediately prior to this planting? 7.1. Date that garden clearing began (m/d/y)? / / 7.2. Date that the garden was burned (m/d/y)? / / 8. Date that planting began (m/d/y)? / /_ For crops listed in 9-21, crops planted in garden (1 = yes, planted; 2 = no, not planted; and 3 = yes will plant, but later in the season): 9. Squash? 16. Jicama? 10. Sweet potato? 17. Lentejas? 11. Chaya? 18. Yam? 12. Chile pepper? 19. Com? 13. Kidney bean? 20. _ 21. Yuca? 14. Fruits (assorted)? Other? 15. Ibes? For garden sides indicated in 35-38, the type of vegetation around the garden (1 = hortaliza, 2 = milpa, 3 = huerto familiar, 4 = huamil/hubche, 5 = monte alto, 6 = monte bajo, 1 = sabana, 9 = otro: 35. North side? 37. South side? 36. East side? 38. West side? 39. Direction and distance of the garden from X-Hazil Sur? 39.1. Name of the garden site?_ 304 305 40. Were there habitat disturbances in the vicinity of the garden during the growing season (1 = yes, 2 = no)? 40.1. If yes, what type of disturbances (1 = fire, 2 = timber harvest)? For garden sides indicated in 41-44, owners of gardens within 500 m (first and last names)? 41. North side? 43. South side? 42. East side? 44. West side? 45. Does the gardener work alone (1) or in a group (2)? 46. If in a group, how many people assist the gardener?_ 47. If in a group, the names and relationship of the assistants to the gardener? 48. What species of animal is the most serious CTOp predator in the garden (1 = armadillo, 2 = cabrito, 3 = jabali, 4 = loros/pdjaros, 5 = mapaclie, 6 = tejon, 7 = tepescuintle, 8 = venado cola blanca, 9 = sereke, 10 = otro)? 49. Will you need to replant the garden due to bad weather (1 = yes, 2 = no, 3 = "depends on the rains")? 50. Frequency wiUi which die gardener hunts in the garden (1 = "a lot" [> 4 times per month], 2 = "once in a while" [2-3 times per mondi], 3 = "rarely or never" [0-1 times per month])? 51. Interviewer notes. 52. Site map (indicate adjacent gardens and direction, distance, and surrounding vegetation for this specific garden). APPENDIX Z GAME SPECIES AND MAIN CROP SPECIES AT EJIDO X-HAZIL Y ANEXOS, QUINTANA ROO, MEXICO, DURING 1989-1990 Scientific name Class or Order/Family' Common name a) Game species' Orthogeomys hispiduf Mammalia/Geomyidae Pocket gopher Agouti paca Mammalia/ Agoutirtae Paca Dasyprocta punctata Mammalia/Dasyproctidae Agouti Nasua nasua Mammalia/Procyonidae Coati Tayassu pecari Mammalia/Tayassuidae White-lipped peccary Tayassu tajacu Mammalia/Tayassuidae Collared peccary Mazama americana Mammalia/Cervidae Brocket deer Odocoileus virginianus Mammalia/Cervidae White-tailed deer Crypturellus cinnamomeus Aves/Tinamidae Thicket tinamou Crax rubra Aves/Cracidae Great curassow Ortalis vetula Aves/Cracidae Plain chachalaca Agriocharis ocellata Aves/Meleagrididae Ocellated turkey b) Main crop species'* Ipomoea batatas Tubiflorales/ Convolvulaceae Sweet potato Cucurbita moschata Campanulales/ Cucurbitaceae Squash Dioscorea alata' Liliales/Dioscoreaceae Yam Cnidoscolus chayamansa Geraniales/ Euphorbiaceae Chaya Manihot esculenta Geraniales/ Euphorbiaceae Yuca Cajanus cajan Resales/ Fabaceae (Leguminosae) Lentil Zea mays Graminales/Gramineae Com Pachyrrhizus erosus var. Rosales/Leguminosae Jicama palmatilobus Phaseolus lunatus Rosales/Leguminosae lb bean Phaseolus vulgaris Rosales/Leguminosae Kidney bean 307 308 Scientific name Class or Order/Family^ Common name Capsicum annum Tubiflorales/ Chili Solanaceae Other' Fruits* ' Class and Family for game species (sorted taxonomically). Order and Family for main crop species (sorted alphabetically by Family). '' Game species based on hunter survey during 1989-1990 (see Ch^ter 3). ' Pocket gophers were excluded from food item analyses because stomach contents were not readily available or identifiable. '' Main crop species based on gardener survey during 1989-1990 (see Chi^ter 6). ' The term "yam" refers primarily to Dioscorea alata, but X-Hazil Sur residents also cultivated another macal that was not identified, but may be either Colocasia esculenta or Xanthosoma yucatanense (Arales/Araceae). ' The term "other" refers primarily to zacate (Scleria lithosperma [Graminales/Cyperaceae]), but also includes an unidentified bean (Phaseolus sp.), an unidentified squash {Cucurbita sp.), and a an unidentified cucumber {Cucumis sp.). * The term "fruits" was used by X-Hazil Sur residents to refer to several species of plants. For analytical purposes, these species were combined, but included the following: Anacardiaceae, mango {Mangifera indica); Bromeliaceae, pineapple {Ananas sativa); Cariacaceae, papaya (Carica papaya); Cucurbitaceae, watermelon (Citrullus vulgaris); Gramineae, sugarcane (Saccharum officinarum); Lauraceae, avocado (Persea americana); Leguminosae, peanut (Arachis hypodaea); Musaceae, banana (Musa sp.); Rutaceae, lime (Citrus aurantifolia) and orange (C. sinensis); and Sapotaceae {Pouteria mammosa). LITERATURE CITED Adams, R. E. W. (ed.). 1977. The origins of Maya civilization. University of New Mexico Press, Albuquerque, 465 pp. Adams, R. E. W. 1991. Prehistoric Mesoamerica (revised edition). University of Oklahoma Press, Norman, 454 pp. Aguilera, H. N. 1959. Suelos. Pp. 117-212 in Los recursos naturales del sureste y su aprovechamiento (E. Beltran, ed.). Instituto Mexicano de Recursos Naturales Renovables (IMERNAR, A. C), Mexico, 326 pp. Alcorn, J. B. 1984a. Huastec Maya ethnobotany. University of Texas Press, Austin, 982 pp. Alcorn, J. B. 1984b. Development policy, forests, and peasant farms: reflections on Huastec-managed forests' contributions to commercial production and resource conservation. Economic Botany 38:389-406. Alvard, M., and H. Kaplan. 1991. Procurement technology and prey mortality among indigenous Neotropical hunters. Pp. 79-104 in Human predators and prey mortality (M. C. Stiner, ed.). Westview Press, Boulder, 276 pp. Amadon, D. 1983. Crax rubra (Pavon Grande, Pav6n, Granadera, Great Curassow). Pp. 569-570 in Costa Rican natural history (D. H. Janzen, ed.). University of Chicago Press, Cicago, 816 pp. Andrews, A. P. 1991. America's ancient mariners. Natural History 10/91:72-75. Andrews, E. W. 1942. Yucatan: architecture. Carnegie Institution of Washington. Year Book, Number 41, Washington, DC, 257-263 pp. Andrews, E. W. 1960. Excavations at Dzibilchaltiin, northern Yucatan. Proceedings of the American Philosophical Society 104(3):254-265. Andrews, E. W. 1965. Archaeology and prehistory in the northern Maya lowlands: an introduction. Pp. 288-330 in Handbook of Middle American Indians, volume 2 (G. R. Willey, ed.). University of Texas Press, Austin, 560 pp. Andrews. E. W. 1973. The development of Maya civilization after abandonment of the southern cities (2d revised edition). Pp. 243-265 in The classic Maya collapse (T. P. Culbert, ed.). University of New Mexico Press, Albuquerque, 549 pp. Anonymous. 1988. The monster that stalked the Gulf of Mexico. U.S. News & World Report 105(12):9-10. 309 310 Aranda Sanchez, J. M. 1981. Rastros de los mamiferos silvestres de Mexico. Institute Nacional de Investigaciones sobre Recursos Bi6ticos, Xalapa, Veracruz. Ediciones Macci6, 198 pp. Ayres, J. M., and C. Ayres. 1979. Aspectos da ca9a no alto rio Aripuana. Acta Amazonica 9(2):287-298. Ayres, J. M., D. de Magalhaes Lima, E. de Souza Martins, and J. L. K. Barreiros. 1991. On the track of the road: changes in subsistence hunting in a Brazilian Amazonian village. Pp. 82-92 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Back, W., and J. M. Lesser. 1981. Chemical constraints of ground water management in the Yucatan Peninsula, Mexico. Journal of Hydrology 51:119-130. Bailey, R. C, G. Head, M. Jenike, B. Owen, R. Rechtman, and E. Zechenter. 1989. Hunting and gathering in tropical rain forest: is it possible? American Anthropologist 91(l):59-82. Baker, R. H. 1974. Records of mammals from Ecuador. Publications of the Museum, Michigan State University, Biological Series 5(7): 129-146. Balee, W. 1984. The persistence of Ka'apor culture. Ph.D. dissert., Columbia University, New York, 290 pp. Bal^e, W. 1985. Ka'apor ritual hunUng. Human Ecology 13(4):485-5 10. Balee, W. 1989. The culture of Amazonian forests. Advances in Economic Botany 7:1-21. Balde, W., and A. G6ly. 1989. Managed forest succession in Amazonia: the Ka'poor case. Advances in Economic Botany 7:129-158. Barrera, A. 1964. La Peninsula de Yucatan como provincia bi6tica. Universidad de Yucat^, Centro de Estudios Mayas, and Institute Nacional de Antropologia e Historia, Merida, 44 pp. Barrera, A. 1980. Sobre la unidad de habitaci6n tradicional campesina y el manejo de recursos bi6ticos en el area maya yucatanense. Bi6tica 5(3): 1 15-129. Barrera, A., A. G6mez-Pompa, and C. Vazquez- Yanes. 1977. El manejo de las selvas por los mayas: sus implicaciones silvicolas y agricolas. Bi6tica 2(2):47-61. Barrera de Jorgenson, A. 1993. Chicle extraction and conservation in Quintana Roo, Mexico. M.S. thesis. University of Florida, Gainesville, 122 pp. Beckerman, S. 1979. The abundance of protein in Amazonia: a reply to Gross. American Anthropologist 81(3):533-560. Beckerman, S. 1980. Fishing and hunting by the Bari of Colombia. Working Papers on South American Indians 2:67-111. Beckerman, S. 1983. Optimal foraging group size for a human population; the case of Bari fishing. American Zoologist 23:283-290. 311 Beissinger, S. R., and N. F. R. Snyder (eds.). 1991. New world parrots in crisis. Smithsonian Institution Press, Washington, DC, 304 pp. Bennett, C. F., Jr. 1962. The Bayano Cuna Indians, Panama: an ecological study of livelihood and diet. Annals of the Association of American Geographers 52(l):32-50. Bigler, W. J. 1974. Seasonal movements and activity patterns of the collared peccary. Journal of Mammalogy 55:851-855. Bimey, E. C, J. B. Bowles, R. M. Timm, and S. L. Williams. 1974. Mammalian distributional records in Yucatan and Quintana Roo, with comments on reproduction, structure, and status of Peninsular populations. Occasional Papers, Bell Museum of Natural History, 13:1-25. Bisbal E., F. J. 1986. Food habits of some neotropical carnivores in Venezuela (Mammalia, Camivora). Mammalia 50(3):329-339. Bodmer, R. E. 1989. Ungulate biomass in relation to feeding strategy within Amazonian forests. Oecologia 81:547-550. Bodmer, R. E. 1990. Ungulate frugivores and the browser-grazer continuum. Oikos 57:319-325. Bodmer, R. E. 1991. Strategies of seed dispersal and seed predation in Amazonian ungulates. Biotropica 23(3):255-261. Bodmer, R. E., T. G. Fang, and L. Moya Ibaiiez. 1988a. Primates and ungulates: a comparison of susceptibility to hunting. Primate Conservation 9:79-83. Bodmer, R. E., T. G. Fang, and L. Moya Ibanez. 1988b. Ungulate management and conservation in the Peruvian Amazon. Biological Conservation 45:303-310. Branan, W. V., M. C. M. Werkhoven, and R. L. Marchington. 1985. Food habits of brocket and white-tailed deer in Suriname. Journal of Wildlife Management 49(4):972-976. Broad, S. 1984. The peccary skin trade. TRAFFIC Bulletin 6:27-28. Bumham, K. P., D. R. Anderson, and J. L. Laake. 1980. Estimation of density from line transect sampling of biological populations. Wildlife Monographs No. 72, 202 pp. Bums, A. F. 1973. Patterns in Yucatec Mayan native performance. Ph.D. dissert.. University of Washington, Seattle, 430 pp. Bums, A. F. 1977. The Caste War in the 1970's: present-day accounts from village Quintana Roo. Pp. 259-273 in Anthropology and history in Yucatan (G. D. Jones, ed.). University of Texas Press, Austin, 344 pp. Bums, A. F. 1980. Interactive features in Yucatec Mayan narratives. Language in Society 9:307- 319. Bums, A. F. 1983. An epoch of miracles: oral literature of the Yucatec Maya. University of Texas Press, Austin, 266 pp. Caballero, J. 1992. Maya homegardens: past, present and future. Etnoecol6gica l(l):35-54. 312 Cabrera Cano, E., M. Sousa, and O. Tellez. 1982. Imagenes de la flora quintanarroense. Centre de Investigaciones de Quintana Roo and Universidad Nacional Aut6noma de Mexico, Puerto Morelos, 224 pp. Cajal, J. L. 1988. The lesser rhea in the Argentine puna region: present situation. Biological Conservation 45:81-91. Camarena-Luhrs, T., and S. I. Salazar-Vallejo (eds.). 1991. Estudios ecol6gicos preliminares de la zona sur de Quintana Roo. Centro de Investigaciones de Quintana Roo, Chetumal, 231 pp. Cant, J. G. H. 1977. A census of the agouti (Dasyprocta punctata) in seasonally dry forest at Tikal, Guatemala, with some comments on strip censusing. Journal of Manmialogy 58(4):688-690. Careaga Viliesid, L. (compiler). 1990. Quintana Roo: textos de su historia. 2 vols. Institute de Investigaciones, Mexico, 393 -I- 433 pp. Cameiro, R. L. 1970. The transition from hunting to horticulture in the Amazon Basin. Proceedings of the VIII International Congress of Anthropological and Ethnological Sciences, Tokyo-Kyoto, Vol. 3, pp. 244-248. Cameiro, R. L. 1983. The cultivation of manioc among the Kuikuru of the Upper Xingu. Pp. 65-111 in Adaptive responses of native Amazonians (R. B. Hames and W. T. Vickers, eds.). Academic Press, New York, 516 pp. Carter, W. E. 1969. New lands and old traditions. Latin American Monographs-Second Series, Number 6. University of Florida Press, Gainesville, 153 pp. Castellanos, H. G. 1983. Aspectos de la organizaci6n social del baquiro de collar, Tayassu tajacu L., en el estado Guarico- Venezuela. Acta Biol6gica Venezuelica 11(4): 127-143. Castro, N., J. Revilla, and M. Neville. 1975-76. Came de monte como una fuente de proteinas en 1976 Iquitos, con referenda especial a monos. Revista Forestal de Peru 6:19-32. Charles-Dominique, P., M. Atramentowicz, M. Charles-Dominique, H. Gerard, A. Hladik, C. M. Hladik, and M. F. Prevost. 1981. Les mammiferes frugivores arboricoles noctumes d'une foret guyanaise: inter-relations plantes-animaux. Revue d'ecologie: la terre et la vie 35:341- 435. Chavez Le6n, G. 1980. La fauna silvestre de Quintana Roo, una riqueza mal aprovechada. Revista de Ciencia Forestal 5(27):3-16. Chavez Le6n, G. 1983. Determinaci6n de las relaciones hombre-fauna silvestre en una zona rural de Quintana Roo. SARH, Subseaetaria Forestal, Institute Nacional de Investigaciones Forestales, Mexico, Boletfn No. 94, 59 pp. Chemas, A., and V. Rico-Cjray. 1991. Apiculture and management of associated vegetation by the Maya of Tixcacaltuyub, Yucatan, Mexico. Agroforestry Systems 13:13-25. Chemela, J. M. 1982. An indigenous system of forest and fish management in the Uaupes Basin of Brazil. Cultural Survival Quarterly 6(2): 17-18. 313 Chemela, J. M. 1985. Indigenous fishing in the Neotropics: the Tukanoan Uanano of the blackwater Vaupds River Basin in Brazil and Colombia. Interciencia 10(2):78-86. Chicch6n, A. 1992. Chimane resource use and market involvement in the Beni Biosphere Reserve, Bolivia. Ph.D. dissert., University of Florida, Gainesville, 271 pp. Clendinnen, I. 1987. Ambivalent conquests: Maya and Spaniard in Yucatan, 1517-1570. Cambridge University Press, Cambridge, 243 pp. CoUett, S. F. 1981. Population characteristics of A^omH' paca (Rodentia) in Colombia. Publications of the Museum, Michigan State University, Biological Series 5(7);485-602. Conklin, H. C. 1961. The study of shifting cultivation. Current Anthropology 2(1):27-61. Cornelius, S. E., M. Alvarado UUoa, J. C. Castro, M. Mata del Valle, and D. C. Robinson. 1991. Management of olive ridley sea turtles {Lepidochelys olivacea) at Playas Nancite and Ostional, Costa Rica. Pp. 111-135 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Cook, S. F., and W. Borah. 1974. The population of Yucatan, 1517-1960. Pp. 1-179 in Essays in population history: Mexico and the Caribbean, vol. 2 (S. F. Cook and W. Borah, eds.). University of California Press, Berkeley, 472 pp. Cowgill, U. M. 1962. An agricultural study of the southern Maya lowlands. American Anthropologist 64:273-286. Crawshaw, P. G., Jr. 1991. Effects of hunting on the reproduction of the Paraguayan caiman (Caiman yacare) in the Pantanal of Mato Grosso, Brazil. Pp. 145-153 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Culbert, T. P., and D. S. Rice (eds.). 1990. Precolumbian population history in the Maya lowlands. University of New Mexico Press, Albuquerque, 395 pp. Dachary, A. C, and S. M. Amaiz Bume. 1983. Estudios socioeconomicos preliminares de Quintana Roo: sector agropecuario y forestal (1902-1980). Centro de Investigaciones de Quintana Roo, A. C, Puerto Morelos, Q.R., 370 pp. Dachary, A. C, and S. M. Amaiz Bume. 1984. Estudios socioeconomicos preliminares de Quintana Roo: el territorio y la poblaci6n (1902-1983). Centro de Investigaciones de Quintana Roo, A.C., Puerto Morelos, Q.R., 294 pp. Dachary, A. C, and S. M. Amaiz Bume. 1989. Sian Ka'an, el hombre y su economia. Centro de Investigaciones de Quintana Roo, Chetumal, 139 pp. Dallmeier, F. 1991. Whistling-ducks as a manageable and sustainable resource in Venezuela: balancing economic costs and benefits. Pp. 266-287 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. 314 Day, G. I. 1986. Javelina: research and management in Arizona. Arizona Game and Fish Department, Phoenix, 122 pp. Deevey, E.S., D.S. Rice, H. Vaughan, M. Brenner, and M. Flannery. 1979. Maya urbanism; impact on a tropical karst environment. Science 206:298-306. Delacour, J., and D. Amandon. 1973. Curassows and related birds. American Museum of Natural History, New York, 247 pp. Denevan, W. M. 1970. Aboriginal drained-field cultivation in the Americas. Science 169:647-654. Denevan, W. M., J. M. Treacy, J. B. Alcorn, C. Padoch, J. Denslow, and S. Flores Paitan. 1984. Indigenous agroforestry in the Peruvian Amazon: Bora Indian management of swidden fallows. Interciencia 9(6): 346-357. Di Castri, F., M. Hadley, and J. Damlamian. 1981. MAB: The Man and the Biosphere Program as an evolving system. Ambio 10(2-3): 52-57. Donkin, R. A. 1985. The peccary: with observations on the introduction of pigs to the New World. Transactions of the American Philosophical Society 75(5): 1-152. Dourojeanni, M. J. 1985. Over-exploited and under-used animals in the Amazon region. Pp. 419-422 in Key environments: Amazonia (G. T. Prance and T. E. Lovejoy, eds.). Pergamon Press, New York, 442 pp. Dowler, R. C, and M. D. Engstrom. 1988. Distributional records of mammals from the southwestern Yucatan Peninsula of Mexico. Annals of Carnegie Museum 57(7): 159-166. Dufour, D. L. 1981. Household variation in energy flow in a population of tropical forest horticulturists. Ph.D. dissert., State University of New York, Binghamton, 401 pp. Dufour, D. L. 1983. Nutrition in the northwest Amazon: household dietary intake and time-energy expenditure. Pp. 329-355 in Adaptive responses of native Amazonians (R. B. Hames and W. T. Vickers, eds.). Academic Press, New York, 516 pp. Dufour, D. L. 1987. Insects as food: a case study from the northwest Amazon. American Anthropologist 89(2):383-397. Dufour, D. L. 1990. Use of tropical rainforests by native Amazonians. BioScience 40(9): 65 2-659. Dumond, D. E. 1977. Independent Maya of the late nineteenth century: chiefdoms and power politics. Pp. 102-138 in Anthropology and history in Yucatan (G. D. Jones, ed.). University of Texas Press, Austin, 344 pp. Dunning, N. P. 1989. Archaeological investigations at Sayil, Yucatan, Mexico: intersite reconnaissance and soil studies during the 1987 season. University of Pittsburgh Anthropological Papers, Number 2, 82 pp. Edwards, C. R. 1986. The human impact on the forest in Quintana Roo, Mexico. Journal of Forest History 30(3): 120-127. 315 Eisenberg, J. F. 1979. Habitat, economy, and society: some correlations and hypotheses for Neotropical primates. Pp. 215-262 in Primate ecology and human origins: ecological influences on social organization (I. S. Bernstein and E. 0. Smith, eds.). Garland STPM Press, New York, 362 pp. Eisenberg, J. F. 1981. The mammalian radiations. University of Chicago Press, Chicago, 610 pp. Eisenberg, J. F., M. A. O'Connell, and P. V. August. 1979. Density, productivity, and distribution of mammals in two Venezuelan habitats. Pp. 187-207 in Vertebrate ecology in the northern Neotropics (J. F. Eisenberg, ed.). Smithsonian Institution Press, Washington, DC, 271 pp. Eisenberg, J. F., and J. Seidensticker. 1976. Ungulates in southern Asia: a consideration of biomass estimates for selected habitats. Biological Conservation 10:293-308. Eisenberg, J. F., and R. W. Thorington, Jr. 1973. A preliminary analysis of a Neotropical mammal fauna. Biotropica 5(3): 150-161. Elmendorf, M. 1976. Nine Mayan women. Schenkman Publishing Company, New York, 159 pp. Emmons, L. H. 1984. Geographic variation in densities and diversities of non-flying mammals in Amazonia. Biotropica 16(3):210-222. Emmons, L. H. 1987. Comparative feeding ecology of felids in a Neotropical rainforest. Behavioral Ecology and Sociobiology 20:271-283. Escobar Nava, A. 1986. Geografia general del Estado de Quintana Roo. Fondo de Fomento Editorial del Gobiemo del Estado de Quintana Roo, Chetumal, 140 pp. Estrada-Loera, E. 1991. Phytogeographic relationships of the Yucatin Peninsula. Journal of Biogeography 18:687-697. Ewell, P. T., and D. Merrill-Sands. 1987. Milpa in Yucatan: a long fallow system and its alternatives in the Maya peasant economy. Pp. 95-129 in Comparative farming systems (B. Turner and S. Brush, eds.). Guilford Press, New York, 428 pp. Fitzgerald, S. 1989. International wildlife trade: whose business is it? World Wildlife Fund, Washington, DC, 459 pp. Flannery, K. V. (ed.). 1982. Maya subsistence: studies in memory of Dennis E. Puleston. Academic Press, New York, 386 pp. Flores, J. S. 1984. Algunas fonnas de caza y pesca usadas en Mesoamerica. Cuademos de Divulgaci6n (Instituto Nacional de Investigaciones sobre Recursos Bi6ticos, Xalapa, Veracruz) 16:1-41. Flores-Villela, O., and P. Gerez Fernandez. 1989. Mexico's living endowment: an overview of biological diversity. Conservation International, Washington, DC, 51 pp. Flowers, N. M. 1983. Seasonal factors in subsistence, nutrition, and child growth in a central Brazilian Indian community. Pp. 357-390 in Adaptive responses of native Amazonians (R. B. Hames and W. T. Vickers, eds.). Academic Press, New York, 516 pp. 316 Forrest, D. W. 1991. The solares of Tixkokob: changing economic roles of homegardens in Yucatan, Mexico. M.A. thesis. University of Florida, Gainesville, 26 pp. Foster, G. M. 1967. The dyadic contract: a model for the social structure of a Mexican peasant village. Pp. 213-230 in Peasant society: a reader (J. M. Potter, M. N. Diaz, and G. M. Foster, eds.). Little, Brown, and Company, Boston, 453 pp. Fragoso, J. M. V. 1991. The effect of hunting on tapirs in Belize. Pp. 154-162 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Fragoso, J. M. V. 1992. The effect of selective logging on Baird's tapir. Pp. 295-304 m Topics in Latin American mammalian biology: their ecology, conservation, and education (M. A. Mares and D. J. Schmidly, eds.). University of Oklahoma Press, Norman, 468 pp. Freese, C. H., P. G. Helme, N. Castro R., and G. Whitesides. 1982. Patterns and determinants on monkey densities in Peru and Bolivia, with notes on distribution. International Journal of Primatology 3:53-90. Frost, M. D. 1974. Man-wildlife relationships in a tropical lagoon area: the case of Tortuguero (Costa Rica). Revista Geografica 81 : 155-164. Frost, M. D. 1977. Wildlife management in Belize: program status and problems. Wildlife Society BuUeUn 5(2):48-51. Galletti, H. A. n.d. Aprovechamientos actuales de la fauna silvestre en la Selva Lacandona y bases para un aprovechamiento mas sostenido. Pp. 55-77 in Acuerdo sobre planificaci6n de aprovechamiento y utilizaci6n de areas forestales tropicales: Mexico-Alemania. Gamboa-Perez, H. C. 1991. Ictiofauna dulceacuicola en la zona sur de Quintana Roo. Pp. 186-198 in Estudios ecol6gicos preliminares de la zona sur de Quintana Roo (T. Camarena-Luhrs and S. 1. Salazar-Vallejo, eds.), Centro de Investigaciones de Quintana Roo, 231 pp. Gann, T. W. F. 1918. The Maya Indians of southern Yucatan and northern British Honduras. Smithsonian Institution, Bureau of American Ethnology, Bulletin 64, 146 pp. Gann, T. W. F. 1926. Ancient cities and modem tribes: exploration and adventure in Maya lands. C. Scribner's Sons, New York, 256 pp. Gann, T. W. F. 1935. Tzibanche, Quintana Roo, Mexico. Maya Research 2:155-166. Garcia, E. 1973. Modificaciones al sistema de clasificaci6n climatica de Koppen (para adaptarlo a las condiciones de la Repiiblica Mexicana), second edition. Universidad Nacional Aut6noma de Mexico, Ciudad Universitaria, Mexico, D.F., 246 pp. Gaumer, G. F. 1917. Monograffa de los mamiferos de Yucatan, Mexico. Departamento de Talleres Graficos de la Secretari'a de Fomento, Mexico, 331 pp. Genoways, H. H., and J. K. Jones, Jr. 1975. Annotated checklist of mammals of the Yucatan Peninsula, Mexico. IV. Camivora, Sirenia, Perissodactyla, and Artiodactyla. Occasional Papers of die Museum, Texas Tech University, 26:1-22. 317 Gilmore, R. M. 1950. Fauna and ethnozoology of South America. Pp. 345-464 in Handbook of South American Indians, Vol. 6 (J. Steward, ed.). Bulletin of the Bureau of American Ethnology 143:1-715. Glanz, W. E. 1982. The terrestrial mammal fauna of Barro Colorado Island: censuses and long-term changes. Pp. 465-468 in The ecology of a tropical forest: seasonal rhythms and long-term changes (E. G. Leigh, Jr., A. S. Rand, and D. M. Windsor, eds.). Smithsonian Institution Press, Washington, DC, 486 pp. Glanz, W. E. 1990. Neotropical mammal densities: how unusual is the conmiunity on Barro Colorado Island, Panama? Pp. 287-313 in Four Neotropical rainforests (A. H. Gentry, ed.). Yale University Press, New Haven, 627 pp. Glanz, W. E. 1991. Mammalian densities at protected versus hunted sites in central Panama. Pp. 163-172 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. G6mez-Pompa, A. 1987a. On Maya silviculture. Mexican Studies/Estudios Mexicanos 3(1): 1-17. G6mez-Pompa, A. 1987b. Tropical deforestation and Maya silviculture: an ecological paradox. Tulane Studies in Zoology and Botany 26(1): 19-37. G6mez-Pompa, A., J. S. Flores, and V. Sosa. 1987. The "pet kot": a man-made forest of the Maya. Interciencia 12(1): 10-15. G6ngora-Arones, E. 1987. Etnozoologia lacandona: la herpetofauna de Lacanja-Chansayab. Cuademos de Divulgacidn (Instituto Nacional de Investigaciones sobre Recursos Bi6ticos, Xalapa, Veracruz, Mexico), No. 31, 31 pp. Gordillo, G. 1988. Campesinos al asalto del cielo: de la expropiaci6n estatal a la apropiaci6n campesina. Siglo XXI Editores, Mexico, DF, 282 pp. Graham, A. 1993. Historical factors and biological diversity in Mexico. Pp. 109-127 in Biological diversity of Mexico: origins and distribution (T. P. Ramamoorthy, R. Bye, A. Lot, and J. Fa, eds.). Oxford University Press, New York, 812 pp. Greenberg, L. S. Z. 1992. Garden hunting among the Yucatec Maya: a coevolutionary history of wildlife and culture. Emoecol6gica l(I):23-33. Grimwood, I. R. 1969. Notes on the distribution and status of some Peruvian mammals (1968). Special Publication, American Committee for International Wildlife Protection and New York Zoological Society, Bronx, New York, 21:1-86. Griscom, L. 1926. The ornithological results of the Mason-Spinden expedition to Yucatan. Part 1. Introduction: birds of the mainland of eastern Yucatan. American Museum Novitates 235:1- 19. Gross, D. 1975. Protein capture and cultural development in the Amazon Basin. American Anthropologist 77:526-549. Haemig, P. D. 1978. Aztec emperor Auitzoti and the great-tailed grackle. Biotropica 10:11-17. 318 Haffer, J. 1974. Avian speciation in tropical South America. Publication of tlie Nuttal Ornithological Club 14:1-390. Halffter, G. 1981. The Mapimi Biosphere Reserve; local participation in conservation and development. Ambio 10(2-3): 93-96. Hall, E. R. 1981. The mammals of North America. Second edition. John Wiley and Sons, New York, 1:1-600 + 90, 2:601-1181 + 90. Hall, E. R., and W. W. Dalquest. 1963. The mammals of Veracruz. University of Kansas Publications, Museum of Natural History, 14(14): 165-362. Hallwachs, W. 1986. Agoutis (Dasyprocta punctata): the inheritors of guapinol (Hymenaea courbaril: Leguminosae). Pp. 285-304 in Frugivores and seed dispersal (A. Estrada and T. H. Fleming, eds). Dr. W. Junk Publishers, Dordrecht, 392 pp. Hamblin, N. L. 1984. Animal use by the Cozumel Maya. University of Arizona Press, Tucson, 206 pp. Hamblin, N. L. 1985. The role of marine resources in the Maya economy: a case study from Cozumel, Mexico. Pp. 159-173 in Prehistoric lowland Maya environment and subsistence economy (M. Pohl, ed.). Papers of the Peabody Museum of Archaeology and Ethnology, Volume 77, Harvard University, Cambridge, 209 pp. Hamblin, N. L., and A. M. Rea. 1985. Isla Cozumel archaeological avifauna. Pp. 175-192 in Prehistoric lowland Maya environment and subsistence economy (M. Pohl, ed.). Papers of the Peabody Museum of Archaeology and Ethnology, Volume 77, Harvard University, Cambridge, 209 pp. Hames, R. B. 1979. A comparison of the efficiencies of the shotgun and the bow in Neotropical forest hunting. Human Ecology 7(3):219-252. Hames, R. B. 1980. Game depletion and hunting zone rotation among the Ye'kwana and Yanomamo of Amazonas, Venezuela. Pp. 24-62 in Working Papers on South American Indians, vol. 2 (R. Hames, ed.). Bennington College, Bennington, VT. Hames, R. B., and W. T. Vickers. 1982. Optimal foraging theory as a model to explain variability in Amazonian hunting. American Ethnologist 9(2): 358-378. Hames, R. B., and W. T. Vickers (eds.). 1983. Adaptive responses of native Amazonians. Academic Press, New York, 516 pp. Hammond, N. 1982a. Ancient Maya civilization. Rutgers University Press, New Brunswick, NJ, 337 pp. Hammond, N. 1982b. The exploration of the Maya world. American Scientist 70:482-495. Hammond, N. 1986. New light on the most ancient Maya. Man (N.S.) 21:399-413. Hanks, W. F. 1990. Referential practice: language and lived space among the Maya. University of Chicago Press, Chicago, 572 pp. 319 Hansen, R. D. 1991. The road to Nakbe. Natural History 5/91:8-14. Harlow, G. E. 1991. Hard rock. Natural History 8/91:4-10. Harris, M. 1974. Cows, pigs, wars, and witches: the riddles of culture. Random House, New York, 276 pp. Hayne, D. W. 1949. An examination of the strip census method for estimating animal populations. Journal of Wildlife Management 13:145-157. Haynes, A. M. 1987. Human exploitation of seabirds in Jamaica. Biological Conservation 41:99-124. Hendrichs, H. 1977. Untersuchungen zur Saugetierfauna in einem palaecotropischen und einem neotropischen Bergregenwaldgebiet. Saugetierkundliche Mitteilungen 25:214-225. Hill, K., and K. Hawkes. 1983. Neotropical hunting among the Ache of eastern Paraguay. Pp. 139- 188 in Adaptive responses of native Amazonians (R. B. Hames and W. T. Vickers, eds.). Academic Press, New York, 516 pp. Hohnberg, A. R. 1969. Nomads of the long bow: the Sirion6 of eastern Bolivia. Waveland Press, Prospect Heights, 294 pp. Hurtado, A. M., K. Hawkes, K. Hill, and H. Kapplan. 1985. Female subsistence strategies among Ache hunter-gatherers of eastern Paraguay. Human Ecology 13(1): 1-28. Hvidberg-Hansen, H. 1970. Utilization of the collared peccary {Tayassu tajacu) in Peru. FAO Forestry Research and Training Project UNDP/SF No. 116, unpublished report. Instituto Nacional de Estadi'stica Geografica e Informatica. 1985. [Aerial photographs; 1:37 000; D. F. 153.43 L526 E16-A36 No. 004-008, D. F. 153.43 L527 E16-A36 No. 003-007, D. F. 153.43 L527 E16-A46 No. 001, D. F. 153.43 L528 E16-A36 No. 004-008, D. F. 153.43 L528 E16-A46 No. 001, and D. F. 153.43 L529 E16-A37 No. 002-005. Direcci6n General de Geografia, Mexico, 21 aerial photogr^hs. Instituto Nacional de Estadistica Geogrffica e Informatica. 1987. [Topographic m^s; Carta topogrdfica (1:50 000; Petcacab E16A36, E16A37, E16A46, and E16A47). Direcci6n General de Geografia, Mexico, 4 sheets. Iftigo-Elias, E. E., and M. A. Ramos. 1991. The psittacine trade in Mexico. Pp. 380-392 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Irvine, D. 1987. Resource management by the Runa Indians of the Ecuadorian Amazon. Ph.D. dissert., Stanford University, Stanford, CA, 306 pp. Janson, C. H., and L. H. Emmons. 1990. Ecological structure of the nonflying mammal community at cocha Cashu Biological Station, Manu National Park, Peru. Pp. 314-338 in Four Neotropical rainforests (A. H. Gentry, ed.). Yale University Press, New Haven, 627 pp. Janzen, D. H. 1974. Tropical blackwaters, animals, and mast fruiting by the Dipterocarpaceae. Biotropica 6:69-103. 320 Johns, A. D. 1986. Effects of habitat disturbance on rainforest wildlife in Brazilian Amazonia. Final report, World Wildlife Fund U.S. Project US-302. World Wildlife Fund, Washington, DC, 111 pp. Johns, A. D. 1988. Economic development and wildlife conservation in Brazilian Amazonia. Ambio 17(5):302-306. Jones, G. D. (ed.). 1977. Anthropology and history in Yucatan. University of Texas Press, Austin, 344 pp. Jones, J. K., Jr., H. H. Genoways, and T. E. Lawlor. 1974a. Annotated checklist of mammals of the Yucatan Peninsula, Mexico. 11. Rodentia. Occasional Papers of the Museum, Texas Tech University, 22:1-24. Jones, J. K., Jr., H. H. Genoways, and J. D. Smith. 1974b. Annotated checklist of mammals of the Yucatan Peninsula, Mexico. III. Marsupialia, Insectivora, Primates, Edentata, and Lagomorpha. Occasional Papers of the Museum, Texas Tech University, 23:1-12. Jorgenson, A. B. 1992. El chicle, un recurso natural fundamental para la economia de Quintana Roo. Amigos de Sian Ka'an 10:3-5. Jorgenson, A., and J. B. Thomsen. 1987. Neotropical parrots imported by the United States, 1981 to 1985. TRAFFIC(USA) 7:3-8. Jorgenson, J. P. 1990. La caceri'a de subsistencia entre los mayas de Quintana Roo. Amigos de Sian Ka'an 7:11-13. Jorgenson, J. P., and A. B. Jorgenson. 1991. Imports of CITES-regulated mammals into the United Sates from Latin America: 1982-1984. Pp. 322-335 in Topics in Latin American manmialian biology: their ecology, conservation, and education (M. A. Mares and D. J. Schmidly, eds.). University of Oklahoma Press, Norman, 468 pp. Karr, J. R., S. K. Robinson, J. G. Blake, and R. O. Bierregaard, Jr. 1990. Birds of four Neotropical forests. Pp. 237-269 in Four Neotropical forests (A. H. Gentry, ed.). Yale University Press, New Haven, 627 pp. Kaufmann, J. H. 1962. Ecology and social behavior of the coati, Nasua narica, on Barro Colorado Island, Panama. University of California Publications in Zoology 60:95-222. Kaufmann, J. H. 1983. Nasua narica (Pizote, Coati). Pp. 478-480 in Costa Rican natural history (D. H. Janzen, ed.). University of Chicago Press, Chicago, 816 pp. Kaufmann, J. H., D. V. Lanning, and S. E. Poole. 1976. Current status and distribution of the coati in the United States. Journal of Mammalogy 57(4):621-637. KiUion, T. W., J. A. Sabloff, G. Tourtellot, and N. P. Dunning. 1989. Intensive surface collection of residential clusters at terminal classic Sayil, Yucatan, Mexico. Journal of Field Archaeology 16(3):273-294. Kiltie, R. A. 1981. Stomach contents of rain forest peccaries (Tayassu tajacu and T. pecari). Biotropica 13(3):234-236. 321 Kiltie, R. A., and J. Terborgh. 1983. Observations on the behavior of rain forest peccaries in Peru: why do white-lipped peccaries form herds? Zeitschrift fiir Tierpsychologie 62(3):24I-255. Konrad, H. W. 1988. De la subsistencia forestal tropical a la produccion para la exportacion: la industria chiclera y la transfonnacion de la economia maya de Quintana Roo de 1890 a 1935. Pp. 161-182 in 45° Congreso Intemacional de Americanistas: etnohistoria e historia de la Americas (E. Reichel D., ed.). Ediciones Uniandes, Bogota, Colombia. Lagueux, C. J. 1991. Economic analysis of sea turtle eggs in a coastal community on the Pacific coast of Honduras. Pp. 136-144 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Lancaster, D. A. 1983. Crypturellus cinnamomeus (Tinamii Canelo, Gongolona, Gallina de Monte, Rufescent Tinamou). Pp. 572-573 in Costa Rican natural history (D. H. Janzen, ed.). University of Chicago Press, Chicago, 816 pp. Landa, F. D. de. 1978. Yucatan before and after the conquest. Translated with notes by William Gates. Dover Publications, New York, 162 pp. (First published as Pubhcation 20 of the Maya Society, Baltimore, in 1937). Lazcano-Barrero, M. A., O. A. Flores-Villela, M. Benabid-Nisenbaum, J. A. Hemandez-G6mez, M. P. Chavez-Pe6n, and A. Cabrera- Aldave. 1988. Estudio y conservaci6n de los anfibios y reptiles de Mexico: una propuesta. Cuademos de Divulgaci6n Inireb No. 25, Instituto Nacional de Investigaciones sobre Recursos Bi6ticos, Xalapa, Veracruz, Mexico, 44 pp. Lee, J. C. 1980. An ecogeographic analysis of the herpetofauna of the Yucatan Peninsula. Miscellaneous Publication, University of Kansas, Museum of Natural History, 67:1-75. Leopold, A. S. 1950. Vegetation zones of Mexico. Ecology 31:507-518. Leopold, A. S. 1972. Wildlife of Mexico: the game birds and mammals. University of California Press, Berkeley, 568 pp. Leopold, A. S. 1977. Fauna silvestre de Mexico: aves y mamiferos de caza. Instituto Mexicano de Recursos Naturales Renovables, Mexico, 673 pp. Linares, O. F. 1976. "Garden hunting" in the American tropics. Human Ecology 4(4):331-349. Lipske, M. 1990. Wonder holes. International Wildlife 20(1):46-51. Lizardi Ramos, C. 1939. Exploraciones arqueol6gicas en Quintana Roo. Revista Mexicana de Estudios Anthropol6gicos 3:46-53. L6pez Omat, A. 1991. Avifauna de la Reserva de la Bi6sfera de Sian Ka'an. Pp. 331-371 in Diversidad biol6gica en la Reserva de la Bi6sfera de Sian Ka'an, Quintana Roo, Mexico (D. Navarro L. and J. G. Robinson, eds.). Centre de Investigaciones de Quintana Roo, Chetumal, and Program for Studies in Tropical Conservation, University of Florida, Gainesville, 471 pp. L6pez Omat, A., J. Lynch, and B. M. de Montes. 1989. New and noteworthy records of birds from the eastern Yucatan Peninsula. Wilson Bulletin 101(3):390-409. 322 L6pez-Portillo, J., M. R. Keyes, A. Gonzalez, E. Cabrera C, and O. Sanchez. 1990. Los incendios de Quintana Roo: ^catistrofe ecol6gica o evento peri6dico? Ciencia y Desarrollo 16(91):43- 57. Lothrop, S. K. 1924. Tulum, an archaeological study of the east coast of Yucatan. Carnegie Institution of Washington, Publication No. 335, 179 pp. Lugo, A. E., J. J. Ewel, Susanna B. Hecht, Peter G. Murphy, Christine Padoch, Marianne C. Schmink, and Donald Stone (eds.). 1987. People and the tropical forest; a research report from the United States Man and the Biosphere Program. U.S. Department of State, Man and the Biosphere Program, Tropical and Subtropical Forests Directorate, Washington, DC, 75 pp. Lundell, C. L. 1933. The agriculture of the Mayas. Southwest Review 19:65-77. Lundell, C. L. 1934. Preliminary sketch of the phytogeography of the Yucatan Peninsula. Carnegie Institution of Washington, Publication 436:257-321. Lundell, C. L. 1937. The vegetation of Peten. Carnegie Institution of Washington, Publication 478:1-244. Luxmoore, R. A. 1988. The caiman lizard. TRAFFIC Bulletin 9-18. Lynch, J. F. 1991. Effects of Hurricane Gilbert on birds in a dry tropical forest in the Yucatan Peninsula. Biotropica 23(4a):488-496. MacKinnon Vda. de Montes, B. 1989. 100 common birds of the Yucatan Peninsula. Amigos de Sian Ka'an, Cancun, Quintana Roo, Mexico, 220 pp. Malcolm, J. R. 1990. Estimation of mammalian densities in continuous forest north of Manaus. Pp. 339-357 in Four Neotropical rainforests (A. H. Gentry, ed.). Yale University Press, New Haven, 627 pp. Mandujano, S., and V. Rico-Gray. 1991. Hunting, use, and knowledge of the biology of the white- tailed deer (Odocoileus virgiiiianus Hays) by the Maya of central Yucatan, Mexico. Journal of Ethnobiology 11(2): 175-183. March M., I. J. 1987. Los Lacandones de Mexico y su relaci6n con los mamiferos silvestres: un estudio etnozool6gico. Bi6tica 12(l):43-56. Marchington, R. L., and D. H. Hirth. 1984. Behavior. Pp. 129-168 in White-tailed deer ecology and management (L. K. Halls, ed.). Stackpole Books, Harrisburg, 870 pp. Marcus, J. 1991. First dates. Natural History 4/91:26-29. Markgraf, V. 1985. Late Pleistocene faunal extinctions in southern Patagonia. Science 228: 1110- 1112. Marks, S. A. 1973. Prey selection and annual harvest of game in a rural Zambian community. East African Wildlife Journal 11:113-128. 323 Marshall, L. G. 1988. Land mammals and the Great American Interchange. American Scientist 76(4):380-388. Marshall, L. G., et al. 1984. Geochronology of the continental mammal-bearing Quaternary of South America. Palaeovertebrata, Mem. Extr., 1984:1-76. Martin, P. S. 1973. The discovery of America. Science 179:969-974. Matamoros H., Y. 1982. Notas sobre la biologia del tepezcuintle, Cuniculus paca, Brisson (Rodentia: Dasyproctidae) en cautiverio. Brenesia 19/20:71-82. McCoy, M., C. Vaughan, and V. Villalobos. 1983. An interesting feeding habit for the collared peccary (Tayassu tajacu Bangs) in Costa Rica. Brenesia 21:456-457. McCoy, M. B., C. S. Vaughan, M. A. Rodriguez, and D. Kitchen. 1990. Seasonal movement, home range, activity and diet of collared peccaries (Tayassu tajacu) in Costa Rican dry forest. Vida Silvestre Neotropical 2(2):6-20. McDonald, D. R. 1977. Food taboos: a primitive enviroimiental protection agency (South America). Anthropos 72:734-748. McMahan, L. R. 1986. The international cat trade. Pp. 461-488 in Cats of the world (S. D. Miller and D. D. Everett, eds.). National Wildlife Federation, Washington, DC, 501 pp. Mediz Bolio, A. 1987. La tierra del faisan y del venado. Producci6n Editorial Dante, S. A., Merida, Yucatan, Mexico, 156 pp. Meffe, G. K., A. H. Ehrlich, and D. Ehrenfeld. 1993. Human population control: the missing agenda. Conservation Biology 7(1): 1-3. Mellink, E., J. R. Aguirre R., and E. Garcia Moya. 1986. Utilizaci6n de la fauna silvestre en el altiplano potosino-zacatecano. Colegio de Postgrados (Instituci6n de Enseiianza e Investigaci6n en Ciencias Agricolas), Mexico, 105 pp. Mendez, E. 1979. Las aves de caza de Panama. [Edici6n privada], Panama, 290 pp. Mittermeier, R. A. 1991. Hunting and its effects on wild primate populations in Suriname. Pp. 93- 107 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Mohr, C. O. 1947. Table of equivalent populations of North American small mammals. American Midland Naturalist 37:223-249. Morley, S. G. 1927. New light on the discovery of Yucatan and the foundation of the new Maya empire. American Journal of Archeology, 2d. ser., 31:51-69. Morley, S. G. 1938. The Maya new empire. Carnegie Institution of Washington, Publication 501:533-565. Morley, S. G. 1956. The ancient Maya, 3d ed. (revised by G. W. Brainerd). Stanford University Press, Stanford, 507 pp. 324 Moseley, E. H., and E. D. Terry (eds.). 1988. Yucatan: a world apart. University of Alabama Press, University, 335 pp. Murie, J. O. 1977. Cues used for cache-finding by agoutis (Dasyprocta punctata). Journal of Mammalogy 58(l):95-96. Murphy, J. 1990. Indigenous forest use and development in the "Maya Zone" of Quintana Roo, Mexico. M.E.S. thesis, York University, Ontario, 181 pp. Myers, N. 1987. The extinction spasm impending: synergisms at work. Conservation Biology 1(1):14-21. Myers, N. 1989. Extinction rates past and present. BioScience 39(1): 39-41. Nations, J. D., and R. B. Nigh. 1980. The evolutionary potential of Lacandon Maya sustained-yield tropical forest agriculture. Journal of Anthropological Research 36(1): 1-30. Navarro L., D., T. Jimenez A., and J. Ju^ez G. 1990. Los mamlferos de Quintana Roo. Pp. 371- 450 in Diversidad biol6gica en la Reserva de la Bi6sfera de Sian Ka'an, Quintana Roo, Mexico (D. Navarro L. and J. G. Robinson, eds.). Centro de Investigaciones de Quintana Roo, Chetumal, and Program for Studies in Tropical Conservation, University of Florida, Gainesville, 471 pp. Navarro L., D., and J. G. Robinson (eds.). 1990. Diversidad biol6gica en la Reserva de la Bi6sfera de Sian Ka'an, Quintana Roo, Mexico. Centro de Investigaciones de Quintana Roo, Chetumal, and Program for Studies in Tropical Conservation, University of Florida, Gainesville, 471 pp. Negreros, P. 1991. Ecology and management of mahogany (Swietenia macrophylla King) regeneration in Quintana Roo, Mexico. Ph.D. dissert., Iowa State University, Ames, 132 pp. Nesbitt, P. H. 1980. The Maya of Yucatan. Pp. 41-59 in Yucatan: a world apart (E. H. Mosely and E. D. Terry, eds.). University of Alabama Press, University, 335 pp. Nietschmann, B. 1972. Hunting and fishing focus among the Miskito Indians, eastern Nicaragua. Human Ecology l(l):41-67. Nietschmann, B. 1973. Between land and water: the subsistence ecology of the Miskito Indians, eastern Nicaragua. Seminar Press, New York, 279 pp. Nietschmann, B. 1979. Ecological change, inflation, and migration in the far western Caribbean. The Geographical Review 69(1): 1-24. Noguez-Galvez, A. M. 1991. Changes in soil properties following shifting cultivation in Quintana Roo, Mexico. M.S. thesis. University of Florida, Gainesville, 64 pp. Nowak, R. M., and J. L. Paradiso. 1983. Walker's mammals of the world. Fourth edition. The Johns Hopkins University Press, Baltimore 2:569-1362. Nye, P. H., and D. J. Greenland. 1961. The soil under shifting cultivation. Technical Communication No. 51. Commonwealth Agricultural Bureau, Bucks, England, 156 pp. 325 Ojasti, J. 1991. Human exploitation of capybara. Pp. 236-252 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Ojeda, R. A., and M. A. Mares. 1984. La degradaci6n de los recursos naturales y la fauna silvestre en Argentina. Interciencia 9:21-26. Olmsted, 1., and R. Duran. 1991. Vegetaci6n de Sian Ka'an. Pp. 1-12 in Diversidad biol6gica en la Reserva de la Bi6sfera de Sian Ka'an, Quintana Roo, Mexico (D. Navarro L. and J. G. Robinson, eds.). Centro de Investigaciones de Quintana Roo, Chetumal, and Program for Studies in Tropical Conservation, University of Florida, Gainesville, 471 pp. Otanez Toxqui, G., and B. Equihua Enriquez. 1981. Comercializaci6n del chicle en Mexico. Boletin Tecnico del Instituto Nacional de Investigaciones Forestales (Mexico), Number 70, 74 pp. Pacheco, T. 1987. Impacto de un inventario forestal sobre la fauna silvestre en la zona del rio Lobo (Requena-Peru). Matero 1(1):9-17. Paolisso, M., and R. Sackett. 1985. Traditional meat procurement strategies among the Ir^a-Yukpa of the Venezuela-Colombia border area. Research in Economic Anthropology 7:177-199. Parker, E., D. Posey, J. Frechione, and L. F. da Silva. 1983. Resource exploitation in Amazonia: ethnoecological examples from four populations. Annals of Carnegie Museum 52(8): 163-203. Parra Lara, A. del Campo. 1986. Uso y manejo tradicional de la fauna silvestre y su relaci6n con otras actividades productivas en San Pedro Jicayan, Oaxaca. Cuademos de Divulgaci6n INIREB (Instituto Nacional de Investigaciones sobre Recursos Bi6ticos, Xalapa, Veracruz), No. 27, 51 pp. Patton, J. L., B. Berlin, and E. A. Berlin. 1982. Aboriginal perspectives of a mammal community in Amazonian Peru: knowledge and utilization patterns among Aguaruna Jivaro. Pp. 111-128 in Manunalian biology in South America (M. A. Mares and H. H. Genoways, eds.). University of Pittsburgh, Pymatuning Laboratory of Ecology, Special Publication No. 6, Linesville, PA, 539 pp. Paynter, R. A., Jr. 1955a. The omithogeogr^hy of the Yucatin Peninsula. Peabody Museum of Natural History, Bulletin, 9:1-347. Paynter, R. A., Jr. 1955b. Additions to the omithogeography of the Yucatan Peninsula. Postilla No. 22, 4 pp. Pereira, N. 1944. A utiliza^ao d came do jacare na Amazonia. Boletim Geografico 2:150-152. Peres, C. A. 1990. Effects of hunting on western Amazonian primate communities. Biological Conservation 54:47-59. Perry, E., A. Reeve, R. Sanborn, and L. Marin. 1990. Comment and reply on "Geologic and environmental aspects of surface cementation, north coast, Yucatan, Mexico." Geology 18(8):802-804. 326 Perry, E., J. Swift, J. Gamboa, A. Reeve, R. Sanborn, L. Marin, and M. Villasuso. 1989. Geologic and environmental aspects of surface cementation, north coast, Yucatan, Mexico. Geology 17(9);818-821. Peterson, J. T. 1981. Game, farming, and interethnic relations in northeastern Luzon, Philippines. Human Ecology 9(1): 1-22. Peterson, J. T. 1982. The effect of farming expansion on hunting. Philippine Sociological Review 30:33-50. Peterson, R. T., and E. L. Chalif. 1973. A field guide to Mexican birds: Mexico, Guatemala, Belize (British Honduras), El Salvador. Houghton Mifflin Company, Boston, 298 pp. Pipemo, D. R., M. B. Bush, and P. A. Colin vaux. 1990. Paleoenviroimients and human occupation in late-glacial Panama. Quaternary Research 33:108-116. Pohl, M. E. D, 1976. Ethnozoology of the Maya: an analysis of fauna from five sites in Peten, Guatemala. Ph.D. dissert.. Harvard University, Cambridge, 317 pp. Pohl, M. (ed.). 1985. Prehistoric lowland Maya environment and subsistence economy. Papers of the Peabody Museum of Archaeology and Ethnology (Harvard University) 77:1-209. Pollock, H. E. D. 1940. The architectural survey of Yucatan. Carnegie Institution of Washington, Year Book 39:265-267. Posey, D. A. 1982. The keepers of the forest. Garden 6:18-24. Posey, D. A. 1983. Indigenous ecological knowledge and development of the Amazon. Pp. 225-256 in The dilemma of Amazonian development (E. Moran, ed.). Westview Press, Boulder, 347 pp. Posey, D. A. 1984. Keepers of the campo. Garden 8(6):8-12, 32. Posey, D. A. 1985. Indigenous management of tropical forest ecosystems: the case of the Kayap6 Indians in the Brazilian Amazon. Agroforestry Systems 3:139-158. Posey, D. A. 1987. Ethnoentomological survey of Brazilian Indians. Entomologia Generalis 12(2/3): 191-202. Posey, D. A., and J. F. Camargo. 1985. Additional information regarding the keeping of stingless bees (Meliponinae) by the Kayap6 Indians of Brazil. Annals of Carnegie Museum 54(8):247- 274. Posey, D. A., J. Frechione, J. Eddins, L. Francelino da Silva, with D. Myers, D. Case, and P. MacBeath. 1984. Ethnoecology as applied anthropology in Amazonian development. Human Organization 43(2):95-107. Puleston, D. E. 1982. The role of ram6n in Maya subsistence. Pp. 353-366 in Maya subsistence: studies in memory of Dennis Puleston (K. V. Flannery, ed.). Academic Press, New York, 368 pp. Pulliam, H. R. 1988. Sources, sinks, and population regulation. American Naturalist 132:652-661. 327 PuUiam, H. R., and B. J. Danielson. 1991. American Naturalist 137(Supplement):S50-S66. Ramamoorthy, T. P., R. A. Bye, A. Lot, and J. E. Fa. 1992. Biological diversity of Mexico. Oxford University Press, New York, 512 pp. Ramirez Pulido, J., R. L6pez Wilchis, C. Mudespacher, and 1. Lira. 1982. Catalogo de los mamiferos terrestres natives de Mexico. Editorial Trillas, Mexico, 126 pp. Ramirez Pulido, J., R. L6pez Wilchis, C. Mudespacher Ziehl, and I. E. Lira. 1983. Lista y bibliografia reciente de los mamiferos de Mexico. Universidad Aut6noma Metropolitana, Unidad Iztapalapa, Mexico, 363 pp. Re Cruz, A. 1992. The two milpas of Chan Kom: a study of socioeconomic and political transformations in a Maya community. Ph.D. dissert.. State University of New York, Albany, 284 pp. Redfield, R. 1960. The little community and peasant society and culture. University of Chicago Press, Chicago, 182 + 92 pp. Redfield, R., and A. Villa Rojas. 1962. Chan Kom, a Maya village. Carnegie Institution of Washington, Publication No. 448, 387 pp. (First published in 1934 by the University of Chicago Press, Chicago, 387 pp.) Redford, K. H. 1992. The empty forest. BioScience 42(6) :4 12-422. Redford, K. H., and J. G. Robinson. 1985. Hunting by indigenous peoples and the conservation of their game species. Cultural Survival Quarterly 9:41-44. Redford, K. H., and J. G. Robinson. 1987. The game of choice: patterns of Indian and colonist hunting in the Neotropics. American Anthropologist 89(3):650-667. Redford, K. H., and J. G. Robinson. 1990. A research agenda for studies of subsistence hunUng in the Neotropics. Florida Journal of Anthropology, Special Publication, No. 6, pp. 117-120. Redford, K. H., and J. G. Robinson. 1991. Subsistence and commercial uses of wildlife in Latin America. Pp. 6-23 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Redford, K. H., and A. M. Stearman. 1989. The Chimane Indians and the Beni Biosphere Reserve, Bolivia. Vida Silvestre Neotropical 2(l):49-56. Reed, N. 1964. The Caste War of Yucatan. Stanford University Press, Stanford, 308 pp. Reina, R. E. 1967. Milpas and milperos: implications for prehistoric times. American Anthropologist 69:1-20. Rewald, M. 1989. Stump sprouts in the milpa cycle and their role in forest regeneration. M.E.S. thesis, York University, Ontario, 167 pp. Reyes Castillo, P. 1981. La fauna silvestre en el plan Balancan-Tenosique. Cuademos de Divulgacidn INIREB (Instituto Nacional de Investigaciones sobre Recursos Bi6ticos, Xalapa, Veracruz, Mexico), No. 4, 45 pp. 328 Rice, D. S. 1991. Roots. Natural History 2/91:10-14. Robinson, J. G. 1993. The limits of caring: sustainable living and the loss of biodiversity. Conservation Biology 7(l):20-28. Robinson, J. G., and J. F. Eisenberg. 1985. Group size and foraging habits of the collared peccary Tayassu tajacu. Journal of Mammalogy 66(1): 153-155. Robinson, J. G., and K. H. Redford. 1986. Body size, diet, and population density of Neotropical forest mammals. American Naturalist 128(5):665-680. Robinson, J. G., and K. H. Redford. 1989. Body size, diet, and population variation in Neotropical forest mammal species: predictors of local extinctions? Pp. 567-594 in Advances in Neotropical manmialogy (K. H. Redford and J. F. Eisenberg, eds.). Sandhill Crane Press, Gainesville, 614 pp. Robinson, J. G., and K. H. Redford (eds.). 1991a. Neotropical wildlife use and conservation. University of Chicago Press, Chicago, 520 pp. Robinson, J. G., and K. H. Redford. 1991b. The use and conservation of wildlife. Pp. 3-5 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Robinson, J. G., and K. H. Redford. 1991c. Sustainable harvest of Neotropical forest animals. Pp. 415-429 in Neotropical wildlife use and conservation (J. G. Robinson and K. H, Redford, eds.). University of Chicago Press, Chicago, 520 pp. Romanoff, S. 1976. Informe sobre el uso de la tierra por los matses en la selva baja peruana. Amazonia Peruana 1(1):97-130. Romanoff, S.A. 1984. Matses adaptations in the Peruvian Amazon. Ph.D. dissert., Columbia University, New York, 306 pp. Ross, E. B. 1978. Food taboos, diet, and hunting strategy: the adaptation to animals in Amazon cultural ecology. Current Anthropology 19(1): 1-16. Roubik, D., R. Villanueva G., E. F. Cabrera Cano, and W. Colli Ucan. 1990. Abejas nativas de la Reserva de la Bi6sfera de Sian Ka'an, Quintana Roo, Mexico. Pp. 317-320 in Diversidad biol6gica en la Reserva de la Bi6sfera de Sian Ka'an, Quintana Roo, Mexico (D. Navarro L., and J. G. Robinson, eds.). Centro de Investigaciones de Quintana Roo, Chetumal, and Program for Studies in Tropical Conservation, University of Florida, Gainesville, 471 pp. Roys, R. L. 1972. The Indian background of colonial Yucatan. University of Oklahoma Press, Norman, 244 pp. First published in 1943 by Carnegie Institution of Washington, Washington, DC. Ruddle, K. 1970. The hunting technology of the Maraca Indians. Antropologica 25:21-63. Ruddle, K. 1974. The Yukpa cultivation system: a study of shifting cultivation in Colombia and Venezuela. Ibero-Americana No. 52. University of California Press, Berkeley, 197 pp. 329 Russell, J. K. 1981. Exclusion of adult male coatis from social groups: protection from predation. Journal of Mammalogy 62:206-208. Rzedowski, J. 1986. Vegetaci6n de Mexico. Editorial Limusa, Mexico, 432 pp. Rzedowski, J., and G. Calder6n de Rzedowski. 1989. Transisthmic Mexico (Campeche, Chiapas, Quintana Roo, Tabasco and Yucatan). Pp. 270-280 in Floristic inventory of tropical countries: the status of plant systematics, collections, and vegetation, plus recommendations for the future (D. G. Campbell and H. D. Hammond, eds.). The New York Botanical Garden, New York, 545 pp. Sabloff, J. A. 1991. Turning points. Natural History 1/91:6-11. Saffirio, G., and R. Scaglion. 1982. Hunting efficiency in acculturated and unacculturated Yanomama villages. Journal of Anthropological Research 38:315-327. Saffirio, J., and R. Hames. 1983. The forest and the highway. Working Papers on South American Indians 6:1-52. Bennington College, Bennington, VT. Sanchez-Herrera, G. T^llez-Giron, R. A. Medellin, and G. Urbano-Vidales. 1986. New records of mammals from Quintana Roo, Mexico. Mammalia 50(2):275-278. Santana, E., L. 1. Iniguez Davalos, and S. Navarro. 1990. Utilizaci6n de la fauna silvestre por las comunidades rurales de la Reserva de la Bi6sfera Sierra de Manantlan. Tiempos de Ciencia/Revista de Difusi6n CienuTica 18:36-43. SAS. 1988. SAS introductory guide for personal computers, release 6.03 edition. SAS Institute, Gary, NC, 111 pp. Scheie, L. 1991. The owl, shield, and flint blade. Natural History 11/91:6-11. Schemnitz, S. D. (ed.). 1980. Wildlife management techniques manual. Fourdi edition: revised. The Wildlife Society, Washington, DC, 686 pp. Schweinsburg, R. E. 1971. Home range, movements, and herd integrity of the collared peccary. Journal of Wildlife Management 35:455-460. Shaw, J. H. 1991. The outlook for sustainable harvests of wildlife in Latin America. Pp. 24-34 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Siskind, J. 1973. Tropical forest hunters and the economy of sex. Pp. 226-241 in Peoples and cultures of native South America (D. Gross, ed.). Doubleday & Natural History Press, Garden City, 566 pp. Slud, P. 1964. The birds of Costa Rica: distribution and ecology. Bulletin of the American Museum of Natural History 128:1-430. Smart, P. L., and F. F. Whitaker. 1990. Comment and reply on "Geologic and environmental aspects of surface cementation, north coast, Yucatan, Mexico." Geology 18(8):802-804. 330 Smith, C. E. Jr., and M. L. Cameron. 1977. Ethnobotany in the Puuc, Yucatan. Economic Botany 31:93-110. Smith, N. J. H. 1974. Destructive exploitation of the South American river turtle. Association of Pacific Coast Geographers 36:85-102. Smith, N. J. H. 1976. Utihzation of game along Brazil's transamazon highway. Acta Amazonica 6(4):455-466. Smith, N. J. H. 1981a. Colonization lessons from a tropical forest. Science 214:755-761. Smith, N. J. H. 1981b. Caimans, capybaras, otters, manatees, and man in Amazonas. Biological Conservation 19(3): 177-187. Smole, W. 1976. The Yanoama Indians: a cultural geography. University of Texas Press, Austin, 272 pp. Smythe, N. 1970. The adaptive value of the social organization of the coati (Nasua narica). Journal of Mammalogy 51:818-820. Smythe, N. 1978. The natural history of the Central American agouti (Dasyprocta punctata). Smithsonian Contribution to Zoology, Number 257, 52 pp. Smythe, N. 1983. Dasyprocta punctata and Agouti paca (Guatusa, Cherenga, Agouti, Tepezcuintl^, Paca). Pp. 463-465 in Costa Rican natural history (D. H. Janzen, ed.). University of Chicago Press, Chicago, 816 pp. Smythe, N., W. G. Glanz, and E. G. Leigh, Jr. 1982. Population regulation in some terrestrial frugivores. Pp. 227-238 in The ecology of a tropical forest: seasonal rhythms and long-term changes (E. G. Leigh, Jr., A. S. Rand, and D. M. Windsor, eds.). Smithsonian Institution Press, Washington, 468 pp. Snook, L. 1993. Stand dynamics of mahogany (Swietenia macrophylla King) and associated species after fire and hurricane in the tropical forests of the Yucatan Peninsula. Ph.D. dissert., Yale University, New Haven, 254 pp. Sowls, L. K. 1974. Social behavior of the collared peccary Dicotyles tajacu L. Pp. 144-165 in The behaviour of ungulates and its relation to management, 2 vols., (V. Geist and F. Walther, eds.). lUCN New Series, Publication 24, Morges, 940 pp. Sowls, L. K. 1978. Collared peccary. Pp. 191-205 in Big game of North America: ecology and management (J. L. Schmidt and D. L. Gilbert, eds.). Wildlife Management Institute and Stackpole Books, Harrisburg, 494 pp. Sowls, L. K. 1983. Tayassu tajacu (Saino, Collared Peccary). Pp. 497-498 in Costa Rican natural history (D. H. Janzen, ed.). University of Chicago Press, Chicago, 816 pp. Sowl, L. K. 1984. The peccaries. University of Arizona Press, Tucson, 251 pp. Sponsel, L. E. 1986. La cacen'a de los Yekuana bajo una perspectiva ecol6gica. Universidad Catolica Andres Bello, Caracas, Venezuela, 27 pp. 331 Standley, P. C. 1930. Flora of Yucatan. Publications of the Field Museum of Natural History, Botanical Series, 3:157-492. Steannan, A. M. 1984. The Yuqui connection: another look at Sirion6 deculturation. American Anthropologist 86(3):630-650. Steannan, A. M. 1989. Yuqui foragers in the Bolivian Amazon: subsistence strategies, prestige, and leadership in an acculturating society. Journal of Anthropological Research 45:219-244. Steannan, A. M. 1990. The effects of settler incursion on fish and game resources of the Yuqui, a native Amazonian society of eastern Bolivia. Human Organization 49(4):373-385. Stearman, A. M. 1992. Neotropical indigenous hunters and their neighbors: Sirion6, Chimana, and Yuqui hunting on the Bohvian frontier. Pp. 108-128 in Conservation of Neotropical forests: working from traditional resource use (K. H. Redford and C. Padoch, eds.). Columbia University Press, New York, 475 pp. Stearman, A. M., and K. H. Redford. 1992. Commercial hunting by subsistence hunters: Sirion6 Indians and Paraguayan caiman in lowland Bolivia. Human Organization 51(3):235-244. Steggerda, M. 1941. Maya Indians of Yucatan. Carnegie Institution of Washington, Pubhcation 531:1-280. Stelhi, F. G., and S. D. Webb (eds.). 1985. The Great American Biotic Interchange. Plenum Press, New York, 532 pp. Stephens, J. L. 1963. Incidents of travel in Yucatin (2 vols.). Dover Publications, Inc., New York, 303 + 336 pp. Stephens, J. L. 1969. Incidents of travel in Central America, Chiapas and Yucatan (2 vols.). Dover Publications, Inc., New York, 424 + 474 pp. (Reprint of the 1841 edition.) Stocks, A. 1983. Cocamilla fishing: patch modification and environmental buffering in the Amazon v^zea. Pp. 239-267 in Adaptive responses of native Amazonians (R. B. Hames and W. T. Vickers, eds.). Academic Press, New York, 516 pp. Stromgaard, P. 1986. Early secondary succession on abandoned shifting cultivation in the miombo of south central Africa. Biotropica 18:97-106. Stromgaard, P. 1991. Soil nutrient accumulation under traditional African agriculture in the miombo woodland of Zambia. Tropical Agriculture 68:74-80. Sullivan, P. 1987. Changing Maya livelihood in the forest of Quintana Roo: flexibility and freedom. Tulane Studies in Zoology and Botany 26(l):39-57. Sullivan, P. 1989. Unfinished conversations: Maya and foreigners between two wars. Alfred A. Knopf, New York, 269 pp. Terborgh, J. 1983. Five species of primates. Princeton University Press, Princeton, 260 pp. 332 Terborgh, J., J. W. Fitzpatrick, and L. Emmons. 1984. Annotated checklist of bird and mammal species of Cocha Cashu Biological Station, Manu National Park, Peru. Fieldiana (Zoology, New Series) 21:1-29. Terborgh, J., S. K. Robinson, T. A. Parker III, C. A. Munn, and N. Pierpont. 1990. Structure and organization of an Amazonian forest bird community. Ecological Monographs 60(2):213-238. Thompson, J. E. S. 1930. Ethnology of the Mayas of southern and central British Honduras. Field Museum of Natural History, Anthropological Series, 17(2):23-214. Thompson, J. E. S. 1966. The rise and fall of Maya civibzation (2d ed. enlarged). University of Oklahoma Press, Norman, 328 pp. Thompson, J. E. S., H. E. D. Pollock, and J. Chariot. 1932. A preliminary study of the ruins of Coba, Quintana Roo, Mexico. Carnegie Institution of Washington, Publication No. 424, 213 pp. Thompson, J. E. S., H. E. D. Pollock, and J. Chariot. 1940. Archaeological problems of the lowland Maya. Pp. 126-138 in The Maya and their neighbors (limited edition). D. Appleton-Century Company, New York, 606 pp. Thompson, R. A. 1974. Occupational sectors. Pp. 52-79 in The winds of tomorrow: social change in a Maya town (R. A. Thompson, ed.). University of Chicago Press, Chicago, 182 pp. Thomsen, J. B., and A. Braiitigam. 1991. Sustainable uses of Neotropical parrots. Pp. 359-379 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Tolstoy, P. 1991. Paper route. Natural History 6/91: 6-14. Tozzer, A. M. 1907. A comparative study of the Mayas and the Lacandones. Archaeological Institute of America, Report of Fellow in American Archaeology, 1902-1905, New York, 195 pp. (AMS Press, New York, 195 pp., reprint of the 1907 edition.) Tozzer, A. M. 1921. A Maya grammar. Papers of the Peabody Museum of American Archaeology and Ethnology (Harvard University), Vol. 9, 301 pp. Turner, B. L., II. 1974. Prehistoric intensive agriculture in the Mayan lowlands .^cience 185:118- 124. Turner, B. L., II. 1990. The rise and fall of population and agriculture in the central Maya lowlands: 300 BC to present. Pp. 178-211 in Hunger in history (L. F. Newman, ed.). Basil Blackwell, Cambridge, 429 pp. Turner, B. L., II. 1991. Environmental history. Environment 33(4):2-3, 45. Turner, B. L., II, and P. D. Harrison (eds.). 1983. PuUtrouser Swamp: ancient Maya habitat, agriculture, and settlements in northern Belize. University of Texas Press, Austin, 294 pp. United States Department of Agriculture, Soil Conservation Service. 1975. Soil taxonomy: a basic system of soil classification for making and interpreting soil surveys. U.S. Govenmient Printing Office, Washington, DC, 754 pp. 333 United States Department of Agriculture, Soil Survey Staff. 1960. Soil classification: a comprehensive system, 7th approximation. United States Govenunent Printing Office, Washington, DC, 265 -PP- Vaughan, C, and M. Rodriguez. 1991. White-tailed deer management in Costa Rica. Pp. 288-299 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Chicago Press, Chicago, 520 pp. Vargas Rivero, C. A. 1983. El ka'anche': una practica horti'cola maya. Bi6tica 8(2): 15 1-173. Vickers, W. T. 1980. An analysis of Amazonian hunting yields as a function of settlement age. Pp. 7-29 in Working Papers on South American Indians (W. T. Vickers and K. M. Kensinger, eds.). Bennington College, Bennington, VT. Vickers, W. T. 1983. Tropical forest mimicry in swiddens: a reassessment of Geertz's model with Amazonian data. Human Ecology ll(l):35-45. Vickers, W. T. 1984. The faunal components of lowland South American hunting kills. Interciencia 9(6):366-376. Vickers, W. T. 1988. Game depletion hypodiesis of Amazonian adaptations: data from a native community. Science 239:1521-1522. Vickers, W. T. 1991. Hunting yields and game composition over ten years in an Amazon Indian territory. Pp. 53-81 in Neotropical wildlife use and conservation (J. G. Robinson and K. H. Redford, eds.). University of Cnicago Press, Chicago, 520 pp. Villa Rojas, A. 1969. The Maya of Yucatan. Pp. 244-275 in Handbook of Middle American Indians, vol. 7 (E. Z. Vogt, ed.). University of Texas Press, Austin, 576 pp. Villa Rojas, A. 1987. Los elegidos de dios: emografia de los mayas de Quintana Roo. Institute Nacional Indigenista, Mexico, 574 pp. Walker, L. R., D. J. Lodge, N. V. L. Brokaw, and R. B. Waide. 1991. An introduction to hurricanes in the Caribbean. Biotropica 23(4a):313-316. Walsh, J., and R. Gannon. 1967. Time is short and the water rises. Thomas Nelson and Sons, Camden, NJ. Watts, E. S., and V. Rico-Gray. 1987. Los primates de la Peninsula de Yucatan, M6xico: estudio preliminar sobre su distribuci6n actual y estado de conservaci6n. Bi6tica 12(l):57-66. Watters, R. F. 1971. Shifting cultivation in Latin America. FAO Forestry Department Paper Number 17. Food and Agriculture Organization of the United Nations, Rome, 305 pp. Webb, S. D., and L. G. Marshall. 1982. Historical biogeography of Recent South American land mammals. Pp. 39-52 in Mammalian biology in South America (M. A. Mares and H. H. Genoways, eds.). University of Pittsburgh, Pymatuning Laboratory of Ecology, Special Publication No. 6, Linesville, PA, 539 pp. 334 Webber, I. L. 1980. Social organization and change in modern Yucatan. Pp. 172-201 in Yucatan: a world apart (E. H. Mosely and E. D. Terry, eds.). University of Alabama Press, University, 335 pp. Werner, D. 1983. Why do the Mekranoti trek? Pp. 225-238 in Adaptive responses of native Amazonians (R. B. Hames and W. T. Vickers, eds.). Academic Press, New York, 516 pp. West, R. C. 1964. Surface configuration and associated geology of Middle America. Pp. 33-83 in Handbook of Middle American Indians, vol. 1 (R. C. West, ed.). University of Texas Press, Austin, 570 pp. Whigham. D. F., I. Olmsted, E. Cabrera Cano, and M. E. Harmon. 1991. The impact of Hurricane Gilbert on trees, litterfall, and woody debris in a dry tropical forest in the northeastern Yucatan Peninsula. Biotropica 23(4a);434-441. Wilder, R. 1988. Damage report from a capricious hurricane: unequal treatment at Gilbert's hands. U.S. News & World Report 105(13):68. Wilkie, D. S. 1989. Impact of roadside agriculture on subsistence hunting in the Ituri Forest of northeastern Zaire. American Journal of Physical Antliropology 78:485-494. Wilkie, D. S., and J. T. Finn. 1990. Slash-bum cultivation and mammal abundance in the Ituri Forest, Zaire. Biotropica 22(1): 90-99. Willey, G., and D. Shimkin. 1973. The Maya collapse: a summary view. Pp. 457-501 in The classic Maya collapse (T. P. Culbert, ed.). University of New Mexico Press, Albuquerque, 549 pp. Wilson, D. E. 1983. Checklist of mammals. Pp. 443-447 in Costa Rican natural history (D. H. Janzen, ed.). University of Chicago Press, Chicago, 816 pp. Wilson, E. M. 1980. Physical geography of the Yucatan Peninsula. Pp. 5-40 in Yucatan: a world apart (E. H. Mosely and E. D. Terry, eds.). University of Alabama Press, University, 335 pp. Wing, E. S. 1974. Vertebrate faunal remains. Pp. 186-188 in Excavation of an early shell midden on Isla Cancun, Quintana Roo, Mexico (E. W. Andrews IV, M. P. Simmons, E. S. Wing, and E. W. Andrews V). Middle American Research Institute Publication, No. 21, Tulane University, New Orleans. Wing, E. S., and D. Steadman. 1980. Vertebrate faunal remains from Dzibilchaltun. Pp. 326-331 in Excavations at Dzibilchaltun, Yucatan, Mexico (E. W. Andrews IV and E. W. Andrews V). Middle American Research Institute Publication, No. 48, Tulane University, New Orleans. Wolf, E. R. 1957. Closed corporate peasant communities in Mesoamerica and central Java. Southwestern Journal of Anthropology 13(1)1-18. World Resources Institute. 1990. World resources 1990-91. Oxford University Press, New York, 383 pp. Yost, J. A., and P. M. Kelley. 1983. Shotguns, blowguns, and spears: the analysis of technological efficiency. Pp. 189-224 in Adaptive responses of native Amazonians (R. B. Hames and W. T. Vickers, eds.). Academic Press, New York, 516 pp. BIOGRAPHICAL SKETCH Jeffrey P. Jorgenson was born on 15 July 1951 in a small town in west central Minnesota. His earliest memories include enjoyable holiday gatherings with relatives, several unexpected moves with his family to new towns, and frequent summer vacations at farms belonging to his grandparents or other relatives. During these vacations he learned how to drive a tractor and stalk wild animals. These were pleasant times. After graduating from Benson Senior High School in 1969, he attended Augustana College (1969-1973), Sioux Falls, South Dakota, where he became interested in plants and bird watching. Subsequently he attended the University of North Dakota (1973-1977), Grand Forks, North Dakota, and earned the Master of Science degree. He conducted research on habitat use and movement patterns of the greater prairie chicken (Tympanuchus cupido pinnatus). In 1976 Jorgenson became a Peace Corps Volunteer and was assigned to Colombia (1976-1979). During the first part of his Peace Corps service he studied the ecology and behavior of the capybara {Hydrochaeris hydrochaeris), the world's largest rodent, while living on a cattle ranch in the northeastern part of the country. Later, he moved to a site in the Andes Mountains, near Santafe de Bogota, and conducted small mammal surveys. At that time Jorgenson also became interested in the ecology and conservation status of the spectacled bear (Tremarctos omatus). Upon completion of his Peace Corps service in 1979, Jorgenson was married to Amanda Barrera Rodriguez, whom he had met in Bogota. Upon his return to the United States in 1979, Jorgenson worked briefly with the Bureau of Reclamation in North Dakota, on the Garrison Diversion Project, before accepting a position in Washington, DC, with the Fish and Wildlife Service (1980-1986). In Washington, he worked as a biologist in the office that served as the U.S. Management Authority for the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). Jorgenson' s knowledge 335 336 of Spanish and his interest in international conservation problems were extremely useful at that time, especially when he served as special assistant to the U.S. CITES delegation at the 1985 meeting in Buenos Aires. In 1986 Jorgenson was accepted into the doctoral program at the University of Florida, where his interest in subsistence hunting by indigenous peoples developed. During his field research, Jorgenson lived for two years in a Maya Indian community in southeastern Mexico. In addition to the research, Jorgenson became the village typist, pawn broker, photographer, and ambulance service. Prior to leaving Mexico, he and his wife became the godparents of a young Maya girl, Nora Rubf Foot Uc. Upon his graduation from the University of Florida, Jorgenson would like to continue working in Latin America, conducting research that will be useful in solving critically important conservation problems in tlie Neotropics. I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Kent H. Redfferd, Chair Associate Professor of Forest Resources and Conservation I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope^d quality, as a dissertation for the degree of Doctor of Philosophy. Bums Professor of Anthropology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. A=,^ Eisenberg Jne Ordway Professoj Ecosystem Conservation I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Allyn M. Sj^afman Professor of Anthropology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. lelvin E. Sunquist Associate Scientist of Forest Resources and Conservation This dissertation was submitted to the Graduate Faculty of the School of Forest Resources and Conservation in the College of Agriculture and to the Graduate School and was accepted as partial fulfilhnent of the requirements for the degree of Doctor of Philosophy. December 1993 _^^^ Director, Forest Resources and Conservation Dean, Graduate School UNIVERSITY OF FLORIDA 3 1262 08556 8359