<%BANNER%>

Landscape ecology of apiculture in the Maya area of La Montana, Campeche, Mexico


PAGE 1

LANDSCAPE ECOLOGY OF APICULTURE IN THE MAYA AREA OF LA MONTAA, CAMPECHE, MXICO By LUCIANA PORTER BOLLAND A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREME NTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2001

PAGE 2

Dedicado a los apicultores y apicultoras de La Montaa.

PAGE 3

iii ACKNOWLEDGMENTS This work was possible thanks to the collaboration and assistance of many people of the La Montaa area; particularly the Montoy Koh family, especially Jorge and Paula. Don Juan Montoy and family, Gregorio Montoy and family, Don Beto Cama ra, Guadalupe May Pacheco, Gabriela Martin Ek, the Gongora Cauich family at Xnoha, and many others, who taught me much of what is in these pages and welcomed me into their homes. I am also thankful to the people of Oxfam/Comadep (particularly Leticia Coel lo, Bernard Vanderweerd, and Lamberto Huh) who facilitated my stay at Xmaben and Mrida. I learned a lot about beekeeping from Victor Camara, who also provided me with insights and suggestions throughout the field work and the writing phase. Juan Jmemez Osornio was very helpful by assisting me at the Universidad Autnoma de Yucatn in Mrida. Juan Tun Garrido helped me with the botanical identification at the Alfredo Barrera Marin herbarium of the Universidad Autnoma de Yucatn, making this work possibl e. Eddie Ellis, while at work at the Universidad de Quintana Roo, was very helpful by assisting me in the satellite data analysis and image classification. Hugo Galletti Busy also provided me with insights during different points of my field work. I am f orever grateful to all of them. I am very happy to have had Henry Gholz and Mary Duryea as advisors, as they have always been patient and helpful in many aspects. I am forever indebted to them. I thank Hugh Popenoe, Marianne Schmink and Michael Binford f or serving on my

PAGE 4

iv committee. I am grateful to Jack Putz, Malcom Sanford, Martin Barker, and Michael Valki for reading drafts of some of the chapters and giving me suggestions. I am also grateful to the statistics department (particularly Ziyad Mahfoud, Be rnhard Klingenberg and David Hitchcock) for helping me with analysis. My friends in Gainesville have helped make it a home and are to a great extent responsible for my having achieved my study goals. I am particularly thankful to Mercedes Prieto, Jorge A ragon, Mariana Varese, Gabriella Scollo, Michael Valki, Paulina and Raul Aragon, Victoria Reyes, Vincent Vades, Monica Espinosa, Victoria Saenz, and many others. I am also thankful to my family for always being there for me. I am specially thankful to my father, Luis Porter, as he has always been a source of inspiration to me in many aspects of my life and particularly in the academic one. The research and writing phase of this work were made possible by funding from Fulbright/Garcia Robles/CONACYT schol arship, a TCD field grant, a Compton Foundation grant through the PSTC, an Inter American Foundation field grant, a TCD fellowship, and support from the School of Forest Resources and Conservation..

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ iii LIST OF TABLES ................................ ................................ ................................ ........... viii LIST OF FIGURES ................................ ................................ ................................ ............ ix ABSTRACT ................................ ................................ ................................ ........................ xi CHAPTERS 1 INTRODUCTION ................................ ................................ ................................ ........... 1 2 THE MAYA LANDSCAPE AT LA MONTAA: PAST AND FUTURE .................... 5 Introduction ................................ ................................ ................................ ..................... 5 Study Area ................................ ................................ ................................ .................. 7 Landscape History ................................ ................................ ................................ ..... 12 Maya Period ................................ ................................ ................................ .......... 13 Colonial and post Colonial period ................................ ................................ ........ 19 Modern period ................................ ................................ ................................ ....... 24 Current situation ................................ ................................ ................................ .... 26 Methods ................................ ................................ ................................ ......................... 27 Surface Analysis ................................ ................................ ................................ ... 29 Results ................................ ................................ ................................ ........................... 31 Discussion ................................ ................................ ................................ ..................... 40 Clearing Rate ................................ ................................ ................................ ............. 40 Land Use Change at La Montaa ................................ ................................ .............. 42 Conclusions ................................ ................................ ................................ ................... 45 3 BEEKEEPING IN THE MAYA AREA: HISTORY AND CURRENT STATUS ....... 47 Introduction ................................ ................................ ................................ ................... 47 Beek eeping Industry in Co ntext: Mexicos Place in World Honey Production and Relevance of the Yucatan Peninsula. ................................ ................................ ............ 48 History of Beekeeping in the Maya Area. ................................ ................................ ..... 51 Beekeeping with the Africanized Bee ................................ ................................ ........... 53 Africanized Bee ................................ ................................ ................................ ......... 53 New Challenges and Opportunities ................................ ................................ ........... 55 Conservation Concerns ................................ ................................ ............................. 59 Conclusions ................................ ................................ ................................ ................... 60

PAGE 6

vi 4 SOCIO ECONOMIC IMPORTANCE OF BEE KEEPING IN THE MAYA FORESTS: THE EXAMPLE OF THE LA MONTAA AREA ................................ ................ 62 Introduction ................................ ................................ ................................ ................... 62 Methods ................................ ................................ ................................ ......................... 63 Study Site ................................ ................................ ................................ .................. 63 Research Conducted ................................ ................................ ................................ .. 64 Results ................................ ................................ ................................ ........................... 65 Beekeeping at La Montaa ................................ ................................ ....................... 65 Socioeconomic Importance of B eekeeping ................................ ............................... 67 Honey Production and Marketing ................................ ................................ ............. 71 The Lol Kax Cooper ative Example ................................ ................................ ......... 75 Conclusions ................................ ................................ ................................ ................... 83 5 FLOWERING PHENOLOGY OF MELLIFEROUS SPECIES AND ITS RELATION TO THE BEEKEEPING CYCLE ................................ ................................ ............. 86 Introduction ................................ ................................ ................................ ................... 86 Study Area and Methods ................................ ................................ ............................... 89 Results and Discussion ................................ ................................ ................................ .. 93 Melliferous/ Polliniferous Species and Their Relation to the Beekeeping Cycle ...... 93 Community Level Flowe ring Phenology ................................ ................................ .. 98 Flowering patterns of the different species and their contribution to apiculture. ... 101 Honey Harvest Flowering ................................ ................................ ....................... 102 High Water Content Honey Flowering ................................ ................................ .. 110 Dearth Period Flowering. ................................ ................................ ........................ 113 Recovery Period Flowering ................................ ................................ .................... 115 Multiple Flowering Species ................................ ................................ .................... 118 Conclusions ................................ ................................ ................................ ................. 119 6 BEEKEEPING AND THE LANDSCAPE ................................ ................................ .. 123 Introduction ................................ ................................ ................................ ................. 123 Methods ................................ ................................ ................................ ....................... 125 Study site ................................ ................................ ................................ ................. 125 Distribution and abundance of melliferous species ................................ ................ 126 Beehive location ................................ ................................ ................................ ...... 128 Results ................................ ................................ ................................ ......................... 129 Distribution and abundance of melliferous and polliniferous species .................... 129 Beehive locati on ................................ ................................ ................................ ...... 137 Discussion ................................ ................................ ................................ ................... 143 Conclusions ................................ ................................ ................................ ................. 149 7 CONCLUSIONS ................................ ................................ ................................ .......... 153

PAGE 7

vii APPENDICES A LIST OF THE IMPORTAN T MELLIFEROUS AND PO LLINIFEROUS SPECIES ACCORDING TO THEIR F LOWERING PERIODS IN RELATION TO THE ANNUAL BEEKEEPING CY CLE AT LA MONTAA, C AMPECHE, MEXICO, 163 B FLOWERING EPISODES OF SELECTED SPECIES OBSERVED FROM JANUARY 1999 TRHOUGH MARCH 2000 AT LA MONTAA, CAMPECHE, MXICO ................................ ................................ ................................ ................. 169 LIST OF REFERENCES ................................ ................................ ................................ 172 BIOGRAPHICAL SKETCH ................................ ................................ ........................... 184

PAGE 8

viii LIST OF TABLES Table Page 1. Major soil/vegetation types identified by local people at La Montaa, Campeche. ... 14 2. Change detection matrix obtained by combining classifications made of the La Montaa portion of Landsat 5 TM, 02/05/96 and Landsat 7 ETM 03/27/00 scenes. ................................ ................................ ................................ ................. 35 3. Changes in pixel count, area, and percentage of total area of land use/ cover types obtained from a post classification analysis of Landsat TM, 02/05/96 and Landsat 7 ETM, 03/27/01. ................................ ................................ ................... 37 4. Some production practices taught at the Technical Training Program for Lol K ax members and used in the cooperatives apiaries at La Montaa. ......................... 81 5. Average of dominance (D ), dominance of mellifero us spe cies (MD ** ) and proportion of MD (MP *** ) for transects sampled for the different lowland young, intermediate and older, and upland young, intermediate and older forest types at La Montaa, Campeche ................................ ................................ .................. 130 6. Average proportion of species that flower during the high water content honey period (MP1), of specie s that flower during the honey harvest period (MP2), and of species that flower during the dearth and recovery periods (MP3) f or transects sampled for the different lowland young, intermediate and older, and upland young, intermediate and older forest type s at La Montaa, Campeche .............. 132 7. Dominance of pollen yie lding species (PD), and dominance of nectar yielding species (ND) f or transects sampled for the different lowland young, intermediate and older, and upland young, intermediate and older forest types at La Montaa, Hopelchen, Campeche ................................ ................................ ......................... 132 8. Melliferous importance value index and its ranking (MIVI and MIVI Rank), importance value index and its ranking (IVI and IVI Rank), and the sum of MIVI values for lowland species of young, intermediate and older vegetation ages. ................................ ................................ ................................ ..................... 135 9. Melliferous imp ortance value index and its ranking (MIVI and MIVI Rank), importance value index and its ranking (IVI and IVI Rank), and the sum of MIVI values for upland species of young, intermediate and older vegetation ages ................................ ................................ ................................ ..................... 136

PAGE 9

ix LIST OF FIGURES Figure Page 1. La Montaa area at Hopelchen, Campeche, Mxico ................................ ........................ 11 2. Soil/vegetation types identified by local people at La Montaa, Campeche and their relation to topography. See Table 1. ................................ ................................ .... 16 3. Conceptual diagram of p rocedures for conducting a post classification c omparison change detection ................................ ................................ ................................ .. 30 4. Image classification of La Montaa portion of a) Landsat 5 TM image02/05/96; and b) Landsat 7 ETM 03/27/00 ................................ ................................ ..................... 33 5. Change map created by combining classification data of the La Montaa portion of Landsat 5 TM, 02/05/96, and Landsat 7 ETM 03/27/00 scenes. ......................... 36 6. La Montaa area, Campeche, Mxico. Buffers made at: A. 1, 3 and 5 km from roads; B. At 3, 5 and 8 km from roads; and C. At 0.5 and 1.0 km from lowlands ......... 39 7. Number of beekeepers that gained, lost, or had no change in their colony numbers from 1988 to 1999, accordi ng to a survey of 64 beekee pers of the communities of Chunek, Pachuitz, Ukum and Xmaben conducted during 1999. ......................... 73 8. Honey produced by Lol Kax during 2000. Values are given for 18, 19 and 20 percent water content and for total production. Data obtai ned from Lol kax financial summary of 2000. ................................ ................................ ................................ 78 9. Metric tons of honey produced in 2000 by each of the 74 producers belonging to the Lol Kax cooperative. Data obtained from Lol kax financial summary of 2000. ................................ ................................ ................................ ..................... 82 10. Flowering phenology of melliferous and polliniferous species at La Montaa, Campeche, Mxico. ................................ ................................ .............................. 96 11. Number of flowering melliferous and polliniferous species and their relation with the beekee ping annual cycle at La Montaa, Campeche, Mxico. ............................ 97 12. Flowering phenology and average monthly precipitation at La Montaa, Campeche, Mxico. A. Flowering phenology of m elliferous and polliniferous species according to data obtained at La Montaa from November 1998 to June 2000.

PAGE 10

x B. Mean monthly precipitation at the meteorological station of Chunchintok (data from 1984 to 1999) ................................ ................................ ..................... 99 13. Perce ntage of melliferous dominance partitioned according to the important periods identified for the beekeeping annual production cycle: High water content honey period, Honey Harvest Period, and Dearth and Recovery periods, at La Montaa, Campeche. ................................ ................................ ............................ 131 14. Locations of a sample of apiaries of 4 communities at La Montaa, Campeche, Mexico. ................................ ................................ ................................ ................ 140 15. A sample of apiaries at Xonha in La Montaa, Campeche, Mexico . ............................. 142 16. A sample of apiaries at Xmaben in La Montaa, Cam peche, Mexico ............................. 143

PAGE 11

xi Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy LANDSCAPE ECOLOGY OF APICULTURE IN THE MAYA AREA OF LA MONTAA, CAMPECHE, MXICO By Lu ciana Porter Bolland December 2001 Chair: Henry L. Gholz Co Chair: Mary L. Duryea Major Department: School of Forest Resources and Conservation Apiculture is one of the most important economic activities for rural communities in the Yucatan Peninsula. In part, this results from the rich natural melliferous and polliniferous flora found in the area and from the beekeeping tradition of the Maya population. At La Montaa, Campeche, Mxico, beekeeping is currently the number one income generating activity. Honey production is currently carried out with Africanized bees ( Apis mellifera L.), for marketing to national and international markets. An analysis of some economic aspects showed that profits are benefiting only a small sector of the population that ha s better management skills and investment capacity. Better technology and technical training, as well as the diversification of the activity could provide more benefits to the region. Nevertheless, for this to happen external investment is needed, partic ularly to increase the organizational, administrative and technical skills of beekeepers.

PAGE 12

xii The landscape at La Montaa greatly benefits apiculture as the different vegetation types are rich in nectar and pollen yielding species. More than 100 species from at least 67 genera and 31 families of trees, shrubs, vines, woody vines and annuals were found to be important for honey production. These species are present in the different vegetation types found in the region, including lowland and upland forests, fa llow areas, recently disturbed sites, and savanna communities. The plant pollinator landscape where honey bees forage, is a function of the bees response to spatial and temporal variation in floral resources (Bronstein 1995). Floral resources are distr ibuted in the landscape both in space and time, according to the distribution and abundance of melliferous and polliniferous species, and to their flowering periods. The patchy nature of the landscape at La Montaa offers diverse resources for bees to for age in, favoring apiculture. An analysis of a 4 year period of Landsat Thematic Mapper remotely sensed imagery showed that forest conversion may be intensifying at La Montaa, as cattle ranching becomes a management goal for people in the area; communica tion at the regional, national and international levels increases; and the population grows. As a buffer zone for the largest reserve in Mexico, the Calakmul Biosphere Reserve, management practices for La Montaa that favor forest conservation are of the utmost importance. Honey production at La Montaa and similar areas represents a promising activity that can favor both natural resources conservation and socio economic development. Nevertheless, for this to occur, there needs to be a directed effort th rough the implementation of specific management practices. Alternative markets may also play a crucial incentive role.

PAGE 13

1 CHAPTER 1 INTRODUCTION Beekeeping in the Yucatan Peninsula has been a traditional activity since ancient times. Maya people are known to have practiced intensive management of stingless bee species, including Melipona beecheii Benneth (Chemas and Rico Gray 1991). Although meliponiculture (beekeeping with native stingless bees of the subfamily Meliponinae) is now seldom practiced, beekeeping with Apis bees has become an important activity in the region. Beekeeping is now carried out with modern techniques (movable frames) and with bees of the species Apis mellifeara L. Over 40 % of Mexicos honey production comes from this region, contributing to Mexicos importance as one of the worlds leading producers and exporters of honey (Angulo Carrera 1992, Crane 1990, Ec hazarreta et al. 1997, Guzman Novoa 1996). Although comprising only 8% of the Mexican territory, the Yucatan Peninsula contributes nearly 40% of the national honey production (Echazarreta et al. 1997). Given that nearly 80% of the beehives are owned by s mall farmers, beekeeping constitutes an important activity for rural communities (Guzman Novoa 1996). The landscape in the Yucatan Peninsula is known to be rich in nectar (melliferous) and pollen (polliniferous) yielding species (Rico Gray et al. 1991). T his characteristic favors beekeeping, as nectar and pollen represent the major food sources for bees. Floral resources are distributed in the landscape both in space and in time, according to the distribution and abundance of melliferous and polliniferous species, and to their flowering periods. Foraging bees are favored by a diverse landscape where

PAGE 14

2 patches of different types of vegetation offer flowering sources through time and space (Chemas and Rico Gray 1991). In areas where forests are still a major component of the landscape, beekeeping has been recognized as having great potential for socio economic development while preserving the natural forests (Acopa and Boege 1998, Galletti 2000). The objective of this research is to analyze the social and e cological implications of beekeeping at the landscape level of the La Montaa area, in Campeche, Mxico. This region is found north of the Calakmul Biosphere Reserve, and is a buffer zone for this protected region. At La Montaa, beekeeping represents on e of the most important economic activities (Galletti 2000, Porter 1995). Through analyzing the social and ecological implications of beekeeping, it will be possible to assess beekeepings status as a management alternative that could favor the conservati on of forested areas, and at the same time increase the well being of its population. In the tropical areas of Mexico, as in other parts of the world, the need to implement alternative activities that may favor forest conservation is of outmost importance given the fact that forest conversion is rapidly occurring in the face of modern human activity. According to an FAO report from 1993, Mexico ranks first worldwide in deforestation, with an estimated annual rate of 1.9% for the tropical deciduous forests (Cairns et al. 1995). While the causes of deforestation in Mexico are multiple, forest conversion to agriculture, particularly cattle ranching, represents the leading factor. Many tropical forests have been shaped to different extents by human societies creating a mosaic of forest patches that have recovered from different disturbances (Browder 1996). Actually, in most protected tropical forests, archeological, historical and ecological evidence exists showing a high density of human population in the past

PAGE 15

3 and sites of continuous occupation over many centuries; and also intensively managed and constantly changing environments (Gmez Pompa and Kaus 1992). In that way, forest patches making up the tropical landscape are dynamic entities, where wildli fe and vegetation are controlled and influenced by groups of interacting biotic and abiotic factors, of which historic and present land use patterns are of great importance (Lyon and Horwich 1996). Understanding the role of current human activity in this landscape is needed to understand how forests are changing over time and across space (McIntosh 1981) and is critically important for the management and conservation of tropical forests. This research is divided into 3 parts. First, the status of the curr ent landscape is analyzed, considering its history and its current trends. This information is presented in Chapter 2, and it sets the background of what the main economic activities are today and what their implications may be for the future landscape st ructure. The second part of the research is presented in Chapters 3 and 4 and provides an understanding of the current status of apiculture in the region as well as some of its socio economic implications. In Chapter 3, a literature review is presented in order to address different issues related to the honey industry in Mexico and specifically in the Yucatan Peninsula. The historical context and the changes that have resulted from the Africanization of Apis mellifera bees are discussed. Findings by diff erent authors that have addressed the Africanization impacts are provided to show limitations and opportunities for beekeeping in different contexts. Chapter 4 includes an analysis of the socio economic implications of beekeeping at La Montaa derived fro m information from local beekeepers through interviews and close collaboration for a period of two honey production seasons. Issues such as the benefits that the local families and

PAGE 16

4 communities are obtaining from this activity, as well as the strengths and limitations of honey production and marketing are discussed. Chapters 5 and 6 present the third part of the research and include issues related to the landscape ecology of apiculture, specifically flowering resource availability for bees through space and time and its relation to beekeeping. Chapter 5 deals with the time dimension and presents the results obtained from a community level flowering phenology study to assess resource availability for bees and its relation to the beekeeping annual cycle. Thi s chapter lists the most important melliferous and polliniferous species for this area, according to beekeepers information. Issues related to the implications for honey production and marketing, and related to forest conservation are also addressed. In Chapter 6 the space dimension is analyzed and an assessment of the distribution and abundance of the melliferous and polliniferous species is presented. The objective is to analyze how the landscape structure favors apiculture (particularly the role of fo rested areas). The management given by beekeepers to vegetation for honey production is also discussed in order to increase the understanding of the role of this activity to the future structure and composition of forests. Finally, in Chapter 7 some concl uding remarks are made by integrating the information derived from the previous chapters. An understanding of the ecological implications of apiculture as well as the socio economic characteristics involved in beekeeping provides some insight into the rol e of beekeeping as a management alternative that favors the conservation of this protected region.

PAGE 17

5 CHAPTER 2 THE MAYA LANDSCAPE AT LA MONTAA: PAST AND FUTURE Introduction Forest systems in tropical areas are dynamic, and their structure and function are molded by different processes acting at different scales (Waide and Lugo 1992). According to Holling (1992) three different classes of processes operating in a hierarchical manner determine structure and function in ecosystems. Factors at the micro scale of centimeters to tens of meters in space and days to decades in time, drive the vegetative processes that d etermine plant growth, plant form, and soil structure. At the other extreme, macro scale factors drive the slow geomorphological processes that dominate the formation of topographic and edaphic structures at scales of hundreds to thousands of kilometers a nd centuries to millennia in time. In the middle, meso scale factors drive the formation of patterns over spatial scales of hundreds of meters to hundreds of kilometers and on time scales of years to decades. Meso scale factors include natural disturbanc es such as fire, insect outbreaks, plant disease, and human activities, that interact with topography and regional climate to form the ecosystem specific patterns on the landscape, shaping the structure and composition of plant and animal communities (Holl ing et al. 1996). Meso scale processes, therefore, lead to cycles of ecosystem development, disruption, and renewal, configuring structure and pattern (Holling 1986). Forman (1995) further differentiates between disturbance and stress. The latter refers to periodic

PAGE 18

6 changes that become part of the ongoing background pattern of variation, while disturbances are events that significantly alter the pattern of variation in the structure or function of a system. Resilience is the ability of the system to maint ain its structure and patterns of behavior in the face of disturbance, which at the same time is molded by the same pulses and variations of the system (Holling 1986). There is increasing understanding of the role that disturbances (and stressors) play in determining the structure and composition of tropical forests. Disturbances and stressors differ in scales that vary in severity, spatial extent, frequency and duration (Waide and Lugo 1992). While cycles of disturbance are integral to the maintenance an d generation of natural diversity (Asquith and Whittaker 1992, Goldammer 1992), tropical forest ecosystems are also drastically being changed in the face of contemporary human impacts and possibly climate change. Therefore, knowing the pattern, kind, freq uency, and magnitude of disturbances (or stressors) in the landscape is important for understanding how forests are changing over time and across space (McIntosh 1981) and is critically important for the management and conservation of tropical forests. Mos t tropical forests have been inhabited by human societies for millennia and have been submitted to numerous cycles of management and abandonment. Recent palynological evidence suggests that such disturbances may have occurred for much longer than previous ly believed (Binford et al. 1983, Flenley 1992). Many tropical forests can be regarded as human social spaces that have been shaped to different extents by human societies, creating a mosaic of forest patches that have recovered from different disturban ces (Browder 1996). In most protected tropical forests, archeological, historical and ecological evidence exists showing a high density of human

PAGE 19

7 population in the past and sites of continuous occupation over many centuries; and also intensively managed and constantly changing environments (Gmez Pompa and Kaus 1992). In that way, forest patches making up the tropical landscape are dynamic entities, where wildlife and vegetation are controlled and influenced by groups of interacting biotic and abiotic fa ctors, of which historic and present land use patterns are of great importance (Lyon and Horwich 1996). The objective of this chapter is to assess the current status and trends of the landscape at La Montaa, Campeche, Mxico, by interpreting remotely sen sed digital image data combined with household level information. A change detection analysis of two satellite images was used to assess forest clearing for agriculture from 1996 and 2000. The following questions were posed: What are the activities affec ting land use that are being carried out by the local population? What are the impacts in the landscape in terms of forest clearing? And, how do current activities differ from traditional ones? Through greater understanding of how the current landscape is being molded by human activity, considering its history of land use, it should be more possible to understand how current activities will affect future composition. Since La Montaa represents a buffer zone for one of the largest reserve areas in th e neotropics, the Calakmul Biosphere Reserve, this has important implications for the search for management alternatives compatible with larger regional conservation goals. Study Area The study area is north of the Calakmul Biosphere Reserve, in the center of the Yucatan Peninsula, Mexico. The area is known as La Montaa and is part of the region

PAGE 20

8 named Los Chenes in the state of Campeche (Figure 1). The study took place in the eastern portion of this region and included 8 of the 11 ejidos 1 that constitute La Montaa: Xmejia, Ukum, Xmaben, Chan chen, Chun ek, Pachuitz, Nuevo Chan yax che, and Xkanha; it also includes the ejido of Belha belonging to the Municipality of Calakmul. The northeastern portion of the area is located at 19 latitude and 89 longitude, and the southwestern portion at 18 and 89, respectively. The total land area is approximately 200,000 ha, with a total population of less than 2,800 inhabitants, of which only 620 are ejidatarios (the persons holding the land title, which in most cases include only the head of the family, most commonly adult males) (Galletti 2000). La Montaa is considered a physiographic, economic and ethnic unit (Galletti 2000) with the population mostly Yucatec Mayan. The land titles of th ese ejidos were legally established between 1920 and 1940, when the Mexican public policy for land was to grant forestlands to ejidal property (Galletti 2000). La Montaa is a forested area and has been relatively isolated until recent times. The primary economic activities are related to timber and nontimber forest products, together with subsistence agriculture. North of La Montaa is the area known as La Milpa, a larger region constituting different communities including the central Municipality, Hopel chen. This is the economic and population center of the Los Chenes region. Topographic relief in this northern part is less abrupt and there is greater population density, with more 1 An ejido is a tenure system in which the members forming the specific community have usufruct rights to the land. Generally land is worked individually, although in some cases (e.g., pastures or forest lands) rights are provided in common (Goldring, 1995). Ejidatarios (ejido members not including their descendents or other family members) constitute a corporate group, the ejido assembly, which is the highest authority. Decisions on individual or collective use of land are made through this assembly. The State grants permits and determines rules and regulations that must be followed. A 1993 amendment to Article 27 of the Constitution changed the ejido legislature and now ejido lands can be priv atized.

PAGE 21

9 agriculturally oriented communities. The population in this area is als o mostly Yucatec Mayan, although there is a growing population of Mennonites. South of La Montaa is the Calakmul Biosphere Reserve, created in 1989 and registered in the UNESCO Man and the Biosphere Program in 1993. The reserve, the largest tropical rese rve in Mexico, totals 723,185 ha. It extends south into Guatemala, abutting with the Maya Biosphere Reserve, making a 1.2 million hectare corridor 2 of permanent forest reserves, including the Sian Kaan Biosphere Reserve at the northeastern end in the sta te of Quintana Roo (Acopa and Boege 1998). Most of La Montaa falls outside the Calakmul Biosphere Reserves boundaries, except for part of the territory of Xkanha, and Belha, but is considered a buffer zone for the protected region. The climate of the area has been determined, according to the Kppen classification system as modified by Garcia (1973), to be a subhumid, warm tropical type, having marked rainy and dry seasons (Flores and Espejel 1994). The mean annual temperature is 26.2 C (1984 2000). The mean annual precipitation is 1223 mm, with high variation from year to year, with a low of 648 mm (1986), and a high of 1807 mm (1999) (CNA, unpublished data). The rainy season lasts from summer (May or June to August) into fall (September to Octobe r). Semi evergreen and semi deciduous upland forests cover most of the landscape in the area, although substantial portions are also covered with lowland forests. Lowland forests that are inundated for several months of the year, referred to as akalche are found in poorly drained sites and are sometimes large, particularly in the northeastern 2 This proposed corridor is now being formally established through the PNUMA (Programa de las Naciones Unidas para el Medio Ambiente); La Montaa has been proposed to be the Focal Area #2 with high priority for conservation (Galletti, personal comm unication).

PAGE 22

10 portion of La Montaa. Deciduous lowland forests also occur in the area. The mean tree height in the upland forests ranges between 25 and 35 m During the dry season, 25 to 50% (semi evergreen) or 50 to 75% (semi deciduous) of the tree species shed their leaves. Deciduous lowland forests have a mean tree height that ranges from 6 to 15 m, and most trees shed their leaves during the dry season. Akalche forests have a mean tree height of 7 m and about 50% of the trees shed their leaves during the dry season (Flores and Espejel 1994). Patches of secondary vegetation in different successional stages as well as patches of agricultural areas are found throughout the area. Hydrophilic and savanna type vegetations are also found (Ucan et al. 1999). These vegetation types comprise mostly Gramineae and Cyperaceae with differing associations of shrubby an d arboreal species. The distribution of these latter plant communities depends in part on the levels of inundation and periods of desiccation, which vary highly (Flores and Espejel 1994). Fire is an important disturbance, particularly in savannas (Pennin gton and Sarukhan 1998), as was reported to the author by people in the area who annually use fire to enhance sprout growth for cattle grazing, to attract deer for hunting, and to favor flowering of the melliferous tree Crecentia cujete Soils in the area were formed mostly by the dissolution of calcareous rocks and over time have resulted in a karst topography (Flores and Espejel 1994). The topography is typically undulating with few outstanding characteristics (Gates 1992). Drainage is subterraneous wit h no superficial flows of water, except in the rainy season during storms and periods of inundation, when rain can be very intense and water moves in temporary surface flow channels. Lowlands are water nourished by filtration of Karst during the humid sea son and become dry to varying degrees depending on recharge through

PAGE 23

11 Figure 1. La Montaa area at Hopelchen, Campeche, Mxico

PAGE 24

12 the dry season (Gates 1999). Some of the lowlands in the area have permanently humid soils and there are permanent or semi permanent water bodies at some sites. Water is the most limiting factor for the people in the area. In the dry season, people rely primarily on aguadas small water bodies that fill up during the rainy season. Aguadas are formed by depressions originating from the collapse of part of the terrestrial crust, as a conse quence of the existence of subterranean currents that dissolved underlying calcareous rocks. These depressions are then filled by sediments with fine textures that hold water and allow only slow drainage to occur (Morales 1992). Flores and Espejel (1994 ) point out that the Maya, based on great knowledge, have developed soil and vegetation classification systems that are still in use today. The Mayan soil classification system is based on soil properties such as color, organic content, drainage, ferrous oxides, and rock content (including the distinction of anthropogenic rock remains), as well as associated vegetation (Table 1, Figure 2). Landscape History The structure and composition of the landscape at La Montaa has been molded to a great extent by di fferent factors, both human induced and natural occurring, that have played important roles in this forest ecosystem over space and time. According to Konrad (1999), hurricanes in the area are recurrent and can be considered a constant factor affecting th e dynamics of succession as well as the stability of human settlement since pre Hispanic times. Between 1871 and 1990, nearly 14 major cyclonic events affected the area (Konrad 1999). According to Flores and Espejel (1994), the return time of hurricanes considered dangerous in the region is from 8 to 15 years. People in the area commonly report the devastation of Hurricane Janet in 1955, which left large portions of the forests resembling land recently felled for agriculture. In their most

PAGE 25

13 extreme c ases, hurricanes create large natural clearings (Snook 1996), but these forceful storms also result in the extensive inundation of lowlands and cultivated areas, as well as defoliation and consequent forest fires, which can sometimes be very extensive (Kon rad 1999). Human intervention and management has shaped the landscape as well. Human activity has a long history with periods of increasing and decreasing populations with differing management goals. The four periods identified and described below are th e Maya period, the colonial and post colonial period, the modern period, and the current situation. Maya Period Vestiges of occupation in the Maya lowlands date to at least 1,000 BC, flourishing during its Classic period at approximately A.D. 250 to 850 (D eevey et al. 1979). This was the time when population densities were highest, reaching numbers greater than 100 per km 2 (Gmez Pompa and Kaus 1992, Turner et al. 2001). Although much of the population was concentrated in urban centers such as Calakmul (l ocated 190 km South of La Montaa), known to have been inhabited since the pre Classic period and becoming the largest city of the Maya area during the late Classic times (Folan 1999), there is archaeological evidence of occupation throughout the entire re gion. Although not much archaeological exploration has been carried out at La Montaa, local people report the existence of archaeological remains throughout the area. Folan (1999) has described the northern part of the Rio Bec region and the Los Chenes area as one politically and economically organized in a decentralized manner, with many different governing groups distributed throughout the region.

PAGE 26

14 Table 1. Major soil/vegetation types identified by local people at La Montaa, Campeche. Their assoc iation to the FAO soil classification system is provided. Mayan name Characteristics Vegetation type Importance to agriculture Ak'alche' Yellowish soils, limy and deep, related to chromic vertisols and gleysols (1). Lowlands that inundate at certain per iods where debris accumulates (2). People in the area distinguish different types of ak'alche' forests according to characteristics such as the soil type and the time it remains inundated.) Low stature forests are present, including perennial and deciduous Aquatic vegetation and savanna type communities are also present (2 Some of the communities identified are: Suut forests. This name is given by local people to describe lowlands sites that inundate for long periods of time and that often are found borde ring savannas. Kat luk are ak'alche' forests where soil becomes very sticky when wet. Chichi'tok are ak'alche' forests that do not inundate very often (dry ak'alche'). People also distinguish a different type of savanna community, named Xoch chakan *, wi th a greater abundance of Haematoxylon campechianum. Chak'an vegetation communities. These are savannas established in ak'alche' terrain with a distinctive vegetation community mainly of grasses from the Graminaceae and Ciperacea, as well as dispersed low stature trees (no greater than 4 or 5 m in height) (2). These communities have been related to anthropogenic activities and depend on continuous disturbances such as fire (3). Local people report that these areas are intentionally burned annually to encou rage grass sprout in order to feed cattle, attract wild animals for hunting, and favor the flowering of Cresentia cujete an important melliferous tree. These soils are not good for agriculture although there is a recent trend to grow one year of milpa an d after harvest establish pasture for cattle*.

PAGE 27

15 T able 1 Conti nued Mayan name Characteristics Vegetation type Importance to agriculture Ya'axhom Dark brown and dark reddish brown soils, limy and deep, related to chromic luvisols (2). Flat lands with greenish dark soil, with mud that absorbs water with great quantities of vegetative organic matter. They have limestone that water currents bring and that mixed with salt obtain a greenish color (1). High, medium and low stature forests (subperennial and perennial) are found (1). This type of soil is very fertile and often cleared for milpa establishment*. Tsek'el Stony dark or reddish soils, less than 30 cm deep. Moderate to rapid drainage and easily eroded. Lithosols and lithic rendzinas are associated (1). Tsek'el s oils contain big stones or sheets of limestone with little organic soil content. Medium and low stature forests are present (2). Not good for agriculture*. K'ankab Flat lands made up of red soil that when dry brake up, becoming hard and compacted. These red, or brownish yellowish soils (from redzinas) have a silt texture, and are usually deep (> 80 cm.), having a rapid internal drainage, and a superficial drainage from moderately slow to slow (1) High and medium stature forests are present (1). This is th e favored soils type for mechanized agriculture*. Kakab or Ramonal These are places characterized by having anthropogenic remains of past times. They contain a deep layer of vegetative soil found over a limestone sheet usually fertilized by part of the li me from archeological sites, becoming very fertile soils (2) Medium and high stature forests. Also called ramonal by local people because of the abundance of the ramon tree ( Brosimum alicastrum ). Other trees common to these areas are: guava ( Talisia olivae formis ), beek ( Ehretia tinifo lia), cocoyol ( Acromia mexicana ) and ceiba ( Ceiba pentandra ). These trees are some of the most important trees for the Maya culture, past and present*. This is the favorite soil type for milpa establishment although it is not often cleared for this purpose since these are areas containing many useful trees that serve as fodder, fruit, hunting areas, to name a few*. (1)Arellano, Rodriguez, and Uuh 1992. (2)Flores and Espejel 1994. (3)Penningrong and Sarukhan 1998. *Personal co mmunication from local informants at La Montaa, Campech

PAGE 28

16 Figure 2. Soil/vegetation types identified by local people at La Montaa, Campeche and their relation to topography. See Table 1. Studies of ancient Maya production sy stems indicate that the Maya managed and used a rich and diverse mosaic of ecosystems. A list of the different practices hypothesized to have been used by ancient Maya, though not exhaustive, includes terraces, raised fields and other hydraulic systems, s acred groves, door yard gardens, other silvicultural systems such as the tolche and pet kot (see below), and the shifting agricultural system called milpa (Barrrera et al. 1977, Caballero 1992, Gmez Pompa 1991, Gmez Pompa and Bainbridge 1995, Marcus 198 2, Puleston 1982, Remmers and Koeijer 1992, Turner and Harrison 1981). The use by ancient Maya of terraces and intensive hydraulic systems such as channels and raised fields has been observed in a large number of depressions in the Maya lowlands (Adams e t al. 1981, Matheny 1978). The high population densities during Classic times are believed to have been sustained as a result of the implementation of these types of intensive systems (Fedick and Ford 1990, Turner and Harrison 1981). Although the abrupt collapse of the Maya civilization (at approximately A.D. 900) is not Kankab Tsekel Kakab Seasonal water line

PAGE 29

17 well understood, it is believed to be due to a combination of different factors, including environmental strain, economic and political crisis, and climate change (Deevey et al. 1979, Hod ell et al. 1995). This period of high occupation and intensive land use left a dramatic imprint on the landscape, as evidenced in pollen data that shows that forests were greatly disturbed (Deevey et al. 1979). However, there is also evidence that Maya p roduction systems relied greatly on forest management systems, with a high degree of biological diversity. This helps to explain both why the flora and fauna of the area are still rich in species, and why the conservation of some natural ecosystems over t heir long span of use was possible (Gmez Pompa and Bainbridge 1995). After the Maya collapse, lower population densities and the absence of intensive agricultural systems gave way to forest recovery (Binford et al. 1987, Turner et al. 2001). Turner (1990 ) estimated population changes during the two centuries between A.D. 800 to 1000, to have decreased from 2.6 or 3.4 million to around 1 million people, representing a 53 to 65 % decrease. Maya agrosilvicultural techniques persisted after the Maya collapse (Marcus 1982). Barrera et al. (1977) have argued that the present floristic composition of forests in the Maya area is a result of ancient silvicultural practices, where human selection processes have interacted with the regeneration of the natural syste m. Evidence for this is based on current traditional management practices and by the dominance in the mature vegetation of tree species that were and still are the same species protected, spared, or planted in the land cleared for crops in the practice of shifting agriculture (Gmez Pompa and Kaus 1992).

PAGE 30

18 The milpa is the traditional shifting agricultural method of a mixed agroforestry system where corn, beans, and squash are the main products. In this practice, uncontrollable fires are rarely caused but r ather individual trees are saved from being burned for the purpose of attracting wild animals, for fruit or wood, or simply for shade (Barrera et al. 1977). The traditional milpa lasts for a period of 3 to 4 years, after which the land is left fallow. Th e Maya had extensive knowledge about the regeneration processes of vegetation. Different stages were identified according to the age of the fallow and were used and managed in different ways (Gmez Pompa 1987). Dooryard gardens or homegardens were managed in many ways to provide different goods and services: food, medicine, fodder, wood for construction, fuel, areas for beekeeping, recreational spaces, etc. They are regarded as centers for plant and animal experimentation. Dooryard gardens are managed th rough intensive silvicultural practices and are structurally very diverse (Remmers and Koeijer 1992). The term pet kot means enclosed vegetation in a wall of stones and it is a reserved area in the forest, enclosed by stonewalls where numerous fruit tree s and other edible or useful plants are concentrated (Remmers and Koeijer 1992). These places are chosen for the protection and cultivation of useful trees. They resemble forests in composition and structure but have been enriched and manipulated, either by favoring wild species coming from local forests, or by introducing different species from distant and more humid forests of the Maya region (Gmez Pompa 1987). The tolche is a 10 20 m wide strip of forest, usually created in the practice of making th e milpa These areas serve as boundary markers, providing protected status to corridors between forest patches. The management of the vegetation as a whole, not individually, characterizes

PAGE 31

19 them, and their purpose is to provide multiple uses such as shelt ered routes for moving between fields, firebreaks, sources of different plant and animal species, and areas of bee forage (Remmers and Koeijer 1992). Another practice that suggests that the Maya played an important role in structuring the vegetation is th at of protecting natural areas through sacred groves (Gmez Pompa 1987). Colonial and post Colonial period La Montaa was still influenced by traditional land use systems. Nevertheless, external factors and land use changes occurred at different points in time as a result of several factors. European contact and its resulting epidemics during the colonial period brought about yet another decrease of the population throughout the Maya area. This depopulation is estimated to have been greater than the de population after the Maya collapse. In a period of 100 years (during the 16 th century) population decrease is estimated to have been as high as 75 to 95 % (Crosby 1972 as cited by Faust 1998). The area of La Montaa, as well as much of the area that toda y makes up the Calakmul Biosphere Reserve, remained somewhat in isolation from European contact until recently. At the same time, though, it has still been an area where external interests have managed to constantly exploit natural resources (Konrad 1999) During the Colonial period, the area of La Montaa was characterized by an indigenous, low density, nomadic population that appeared and disappeared, depending on the milpas cycle. These so called independent Mayas, known as huites had little contact with the white population and maintained a traditional subsistence system based on the cultivation of corn and beans and on trade. Trade occurred to the north with population centers that had been established by Spanish friars -e.g., Hopelchen, which

PAGE 32

20 w as established in 1621 and by 1630 had grown to over 20,000 native people -and to the south with English settlers. Some of the products which were traded included honey, wax, copal, cotton, tobacco, medicinal and aromatic plants, fibers, wood, live birds and bird products, meat and skin from deer and other animals, and spices such as pepper and annatto. In exchange, the Mayas obtained salt, rope, tools, and gunpowder (Ramayo Lanz 1996). After Mexican independence (1821), the Maya population throughout th e Peninsula suffered the consequences of the new republic and its modernizing project. With the objective of integrating the milpa areas into modern agriculture, post colonial agrarian policy formalized the dispossession of lands from native communities ( Ramayo Lanz 1996). The forests of the Yucatan Peninsula suffered from the increase of farming and ranching, where products such as rice and sugar cane were cultivated. The tribute system, mainly in the form of payments of cotton and beeswax, as well as f orced labor, was maintained after independence from Spain. The imparting of justice, as well as clerical services, reflected a new government that remained profoundly racist (Ramayo Lanz 1996). Ancient hierarchical forms of power, which the Mayas had bee n able to sustain throughout the colonial period, also fell to the new governments social and political reforms. According to Ramayo Lanz (1996), these factors, together with false promises made to the Mayas by different political groups of the then form ing government (centralistas and federalistas) in order to recruit them into their militias, led to the breakout of the so called War of the Castes. This long and violent period lasted over 50 years (from 1847 to the beginning of the 20 th century) and was decisive in the history of Mexico in many aspects; it also strongly affected the future of La Montaa. This area

PAGE 33

21 became the settlement lands of the pacificos, a group of Maya willing to negotiate with the government, as opposed to the more radical group, the rebeldes. Manipulated by the government, this division among the Mayas ultimately reduced their strength and led to their eventual defeat (Ramayo Lanz 1996). The Maya population known as pacificos del sur (coming from other parts of the Pen insula) congregated into settlements at La Montaa. The population in this area fluctuated throughout the second half of the 19 th century, according to different events and circumstances. At that time, the valleys of the northern most part of Los Chenes had already become prosperous agricultural areas that produced export goods, such as sugar cane, corn, rice, cotton and tobacco. This area was at the highest point of its economy, and was generating pressure to the south. Many Mayas fleeing from slavery increased population numbers in the southern hills of La Montaa, an area that maintained a certain autonomy (Ramayo Lanz 1996). By 1860, the area of La Montaa was considered a municipality and had a population estimated at 14,400 (Dumond 1997). By the second half of the 19 th century, Campeche had become an international economic enclave. With its proyecto agroexportador, sugar cane became an important industry that competed with that of Cuba (Ramayo Lanz 1996). Palo de tinte or logwood ( Haematoxylon campechianum L. ) had already been an important export product since the end of the previous century, particularly for the English market. An important ingredient used for fixing dyes, logwood from Campeche was exported in vast quantities to supply Englan ds flourishing textile industry (Ferr Damar 1997). La Montaa formed part of the commercial corridor that connected the southwest to the

PAGE 34

22 southeastern English settlements through a network of paths where logwood, mahogany and other forest goods found t heir way to the old world (Ramayo Lanz 1996) 1 Nevertheless, the Caste War had its effect on a growing economy, which was based to a large extent on slavery. Los Chenes was severely affected by the war, and its proyecto agroexportador failed because of a reduction in the labor force (Ramayo Lanz 1996). At La Montaa, different confrontations both with the government and with the Maya rebeldes resulted in a decrease of the population as well. Whole towns were abandoned. By the end of the 1860s there were only two population centers of pacificos left: Xkanha and Ikaiche, and estimates are that the population had been reduced to 12,000 people. By 1900 estimates are that the population had been further reduced to around 8,000 individuals and the only remaining population center was Xkanha (Dumond 1997). The agricultural project of Los Chenes was further affected by a devastating locust outbreak that occurred at the end of the 19 th century. At La Montaa, hunger and disease greatly diminished the rema ining population, and by the beginning of the 20 th century, the government declared the forests of La Montaa to be despoblados, or uninhabited areas 2 (Ramayo Lanz 1996). A new period began for the forests of Campeche, characterized by the exploitation of resources through foreign hands and a few local businessmen (since many had already gone north to invest in the henequen project of Yucatn). At the end of the 19 th century, the government of Porfirio Diaz had already set the stage for forest 1 The importance of palo de tinte in the textile industry ended after synthetic dyes were invented by William Henry Perking at the end of the 1850s (Ferr Damar 1997). 2 The population remaining at that time is not estimated, but it stoppe d being considered economically important.

PAGE 35

23 exploita tion. Communication networks were built and forest concessions were granted to foreign companies, particularly from North America 3 (Ramayo Llanz 1998). La Montaa was left in isolation for most of the 20 th century. During the first half of the 1900s the area was characterized by the exploitation of the chicle tree ( Manilkara zapota L. Van Royen). The forests of Campeche and Quintana Roo produced approximately 80% of the worlds supply of the chicle resin used for making chewing gum, an important export product, particularly during the period of the two World Wars 4 (Acopa and Boege 1998). The product was brought from the forests along a network of airplane landing strips, and chicleros (those who tapped the chicle trees) were imported seasonally to work in the area. According to Konrad (1999) between 1901 and 1910, 10 million km of chicle were exported from Campeche, representing 50% of all national exports. The benefits accrued during the early chicle period remained in the hands of the foreign compani es and contractors, with little profit left for local people in the area. Nonetheless, the presence of chicle camps, together with water availability, guided the re establishment of communities at La Montaa (Ramayo Lanz 1996). During the 1920s, after t he Mexican revolution (1910 1917), the population at La Montaa was granted ejidal property rights. During the 1930s and 1940s, large forest tracts were provided to the communities (Galletti 2000). These forestry ejidos responded to chicle production nee ds and their large tracks were intended to guarantee good production rates per person. At times population density was as low as 400 ha per person 3 In 1886 the Mexican government signed a contract in which several North American companies, including the Pennsylvania Campeche Land and Lumber Company, the Laguna Corporation, the Mexican Gulf Land and Lumber C ompany, the Campeche Timber and Fruit Company, and the Campeche Development Company were granted concession to approximately 10,000 km 2 of Campeches territory (Ramayo Lanz 1996). 4 The U.S. Army provided chewing gum to its soldiers as part of their ration s (Acopa and Boege 1998).

PAGE 36

24 (Acopa and Boege 1998). After the 1930s, chicle production in the area was commercialized through the Coope rativa de Los Chenes (and later and until 1995, through the Banco Mexicano de Comercio Exterior IMPEXNAL), regulated by the government. After the chicle boom ended (shortly after World War II, when synthetic gums replaced latex for making chewing gum), c hicle trees continued to be tapped with up and down market demand, but to a much lesser degree. According to Galletti (2000), timber exploitation at La Montaa began during the 1960s when the Zoh Laguna Mill extended its activities to the north. Over th e following decades, the forests were mined, and the most valuable species, particularly mahogany ( Swietenia macrophyla King) and spanish cedar ( Cedrela odorata L.) were substantially reduced. This period of timber extraction also left little or no benefi t to the area (instead reducing its economic potential), as lumbering was carried out with neither management plans nor the participation of the local population more than as seasonal workers. The result of this period of forest exploitation was the impov erishment of forests, and the creation of a network of roads with its future consequences for resource exploitation (Galletti 2000). Modern period During the 1970s and 1980s government initiatives were promoted with the objective of integrating the area in to the national economy. At that time, the rural sector in Campeche was subject to a rash of agricultural development projects that promoted colonization, mechanized agriculture and cattle ranching that resulted in the deforestation of at least half a mi llion hectares of forests in Campeche (Gates 1999a, 1999b). Although the area of La Montaa remained somewhat inaccessible until the construction of the Dzibalchen Xpujil road in 1995, this period set the stage for the

PAGE 37

25 current situation in the area. Seve ral agricultural projects were promoted, including beekeeping with Apis bees, which was adopted by traditional beekeepers in the early 1970s. In the early 1980s, a large scale mechanized rice production project was established, which resulted in the defor estation of 1000 ha of lowland forests (information obtained from local people). Although this project failed and was abandoned shortly after it was established as a result of technical and organizational problems (as occurred with many other such project s in different areas of Campeche), it opened land for the introduction of cattle ranching, and brought expectations for mechanized agriculture to people in the area (Gates 1999b). These changes towards more agriculturally oriented production systems occur red slowly in the area, but greatly accelerated after the construction of the Dzibalchen Xpujil road. Pressure from areas surrounding La Montaa has occurred as a result of changes brought about by the modern period of the rural sector in Mexico. The Es carcega Chetumal road that was built in 1967 brought thousands of settlers from all over the country to the area that today is the Calakmul Biosphere Reserve (Gates 1999b). Population in the region of Xpujil is estimated to have quadrupled from 1980 to 19 90 (Ericson 1997). Human settlements within the Calakmul Biosphere Reserve and surrounding areas (an area that remained until the 1950s in greater isolation than La Montaa) have been estimated to total approximately 24,000 people settled in 72 establishm ents (Sanchez 1999). This population is very heterogeneous, coming from 23 different Mexican states, and including some Mayan ethnic groups such as Yucatec, Chol and Tzeltal (Acopa & Boege 1998). These farmers mainly farm for subsistence, practice forest extraction, grow commercial plantations of chile serrano ( Capsicum sp L.), and

PAGE 38

26 work as migrant laborers in nearby cities such as Campeche and Mrida (ECOMAT 1995). Cattle raising is a strong factor in some of these communities. Beekeeping and chicle extr action are also carried out, mainly by the Maya populations. Other experiments in conservation are taking place related to the promotion of alternative land use systems after the establishment of the Calakmul Biosphere Reserve in 1993, including the pro duction of other non timber forests products, such as allspice ( Pimenta dioica (L.) Merrill), the establishment of agroforestry systems based on fruit tree plantations, the use of native legume species as green fertilizers, and the encouragement of ecotour ism as an alternative industry (Acopa & Boege 1998). North of La Montaa, a group of Mennonite settlements has been increasing in population and area. Mennonites first occupied lands among the territory of the old established Maya population. During rece nt decades this population has increased, arriving from Canada and northern parts of Mexico. By 1995 Mennonites occupied a total land area of 45,000 ha and their population increased to 1,100 from 490 in 1990 (Ericson 1997). There are indications that th is population is still increasing. In 2000, a group of Mennonites bought 5,000 ha from one of the communities at La Montaa, despite some disagreement within the community. This represents a major concern because the Mennonites pursue intensive agricultu re, felling a large fraction of the forests, utilizing heavy machinery, and applying large loads of agrochemicals (Ericson 1997). Current situation The current population of La Montaa is about 2,800 (Galletti 1999), with some slight growth. In 1996, the ejido of Chan yax che was created by a relatively small and heterogeneous group of migrants (about 12 families of people from Tabasco and

PAGE 39

27 Veracruz). Sporadic immigration into old ejidos also has been occurring, in some cases one or two migrant families a sking the ejidos assembly for permission to occupy land, but in other cases, by having many outside families as much as double the ejidos population (as was the case of Chan chen) promoted by government intervention. Nevertheless, recent migration to the area as a whole has resulted in only limited total population growth, with densities still only 116 to 783 ha per ejidatario (Galletti 1999). The economy of the area depends on the use of a wide array of ecosystems from which people obtain different goods and services. The traditional milpa is the basis for subsistence. Other types of agriculture, like various agroforestry combinations including homegardens, are also common and important for meeting family needs. People in the area obtain meat from anim al husbandry in their homegardens and from hunting in the forest. Use of timber and non timber products is important as well, as people obtain wood for fuel and construction, fruits, medicines and other goods. People conduct beekeeping and chicle extract ion for cash income and occasionally sell timber to outside companies. Cattle raising also has been gaining importance in recent years (Porter 1995, Galletti 2000). Methods Remotely sensed digital image data were used for describing the current landscape at La Montaa by means of a change detection analysis of two scenes taken at different dates. The objective was to monitor human activity as evidenced from forest clearing. The scenes used for this study were a Landsat 5 TM scene from February 5, 1996, a nd a Landsat 7 ETM scene from March 27, 2000 (WRS 2, Path 20, Row 47). According to Sader et al. (1994), who conducted an analysis of forest change estimates in

PAGE 40

28 a similar area (northern Petn region of Guatemala), a 4 year time period between satellite da ta acquisition is a good one for monitoring forest clearing patterns, because a fallow milpa site of two years can be clearly distinguished from active agricultural areas. The scenes were analyzed by performing separate unsupervised multispectral classific ation for each, and then conducting a post classification detection of changes. The softwares used for processing the data and conducting analyses were Envi and Arc View. Multispectral classification is the procedure of grouping image pixels into spectra l clusters. This is carried out by assigning pixels with similar spectral levels into specific classes of known reference signatures (Sohn et al. 1999). The post classification comparison change detection is used to compare the two classified images of t he different dates on a pixel by pixel basis using a change detection matrix (Jensen 1996). This procedure creates from to change class information that is used to create a change image map. A conceptual diagram depicts the procedures for analysis tha t were followed (Figure 3). After resizing the two scenes to an area that included the extent of La Montaa, the 1996 scene was rectified to the already georeferenced Landsat 7, 2000 image, obtaining an RMS error of less than 0.5 pixels. An ISODATA class ifier, one for each scene, was run using all but the thermal band. The classifier generated narrowly defined spectral clusters (the procedure was set to obtain a maximum of 100 clusters), 85 for the 1996 scene and 86 for the 2000 scene. Spectral pattern m atching using information from a classification conducted for the same 2000 scene but at a smaller scale at the University of Quintana Roo, and information derived from field work in the area during January 1999 to May 2000, were used to

PAGE 41

29 merge classes and create a pre classification of the study area. An image of this scene was taken to the field during June 2001 for verification. Participatory image analysis was conducted with 5 local people, and land/cover use types were assigned to the pre classified i mage according to visual interpretation and to the identification of specific sites. This step was followed by transect walks conducted with the same people for observation, and at different sites spatial data was collected using a hand held Garmin 12 Glo bal Positioning System (GPS). A re classification followed, using a total of 78 GPS points as reference points. Once each cluster was assigned to one of the reference classes, spectral clusters that belonged to the same land cover/use type were merged together to create a classified 2000 image. This image was re evaluated by the group of participants in the field. The 1996 image was classifi ed accordingly, using the information generated through the 2000 image classification. The two classified images were combined to conduct a post classification comparison change detection. This generated a change detection matrix that was used to create a change image map. Surface Analysis A surface analysis was conducted to address different factors that may influence forest conversion. The factors analyzed were distance to roads, distance to settlement areas and soil type. For each of these factors se veral buffers were created and subsequently related to the change image map. For assessing disturbance as a factor of distance from roads buffers were created at 1, 3 and 5 km intervals. For assessing disturbance as a factor of distance from settlements buffers were created at 3, 5 and 8 km. For assessing the influence of soils, disturbed areas were quantified within a 0.5and 1.0 km buffer from lowlands.

PAGE 42

30 Figure 3. Conceptual diagram of procedures for conducting a post classification c omparison change detection

PAGE 43

31 Results The classified land cover/use maps of 1996 and 2000 are shown in Figure 4. The 2000 image wa s classified into 7 types, including: 1) upland forest, 2) lowland forest, 3) advanced secondary succession/burned areas (quemadal/acahual), 4) secondary succession/agriculture (acahual/agriculture), 5) agriculture/urban, 6) aquatic vegetation/ savanna 7) and water. The classification of the 1996 scene included the same land cover/use types, with an additional cloud mask class added. The latter was created after performing a procedure to mask out clouds and their shadows; this was not done in 2000 si nce the scene was cloud free. Since the objective of the classification was to document forest clearing, no attempt was made to more finely classify different types of upland (e.g., sub perennifolius vs sub deciduous), or lowland forests (e.g., akalche vs deciduous lowlands). Regarding the local peoples classification system, upland forest includes tzequel kakab kankab and yash hom ; and lowland forest includes akalche and suut (Figure 2). Advanced secondary succession includes areas locally known as quemadales which are areas that have undergone fire (either natural or intentionally set) and areas locally known as acahuales which is a Spanish term used to refer to secondary vegetation. Secondary vegetation/agriculture refers to early stages of rec overy after an intensive disturbance (e.g., agriculture). Since the spectral signatures of active agricultural areas and urban settings were not differentiated, they were placed together as a single class. Aquatic vegetation/ savannas, a class that inclu des areas with a distinctive vegetation cover, can sometimes also be spectrally confused with quemadales or acahuales

PAGE 44

32 The post classification comparison yielded a change detection matrix (Table 2). Since the accuracy of the change detection in this type of procedure is dependent on the accuracy of the two separate classifications (Jensen 1996), several things should be taken into consideration. For change detection using remotely sensed digital image data, the date of data acquisition within the year is important, due to seasonal variations in the spectral signatures (Brondizio et al. 1994). Both images were obt ained during the dry season. But the 1996 image was taken at the beginning of February, when soil moisture was still high and trees had only begun to shed their leaves, while the 2000 image was taken at the end of March, when the dry season was advanced, s oils dryer and the trees that shed their leaves had already done so. Additionally, the quality of the 1996 image was not very good, presenting radiometric error shown as striping. Although several filtering procedures were used to correct this error, the condition did not markedly improve. Nevertheless, a principal components analyses for the 1996 image showed that only 15% of the variation may have been the result of radiometric error.

PAGE 45

33 Figure 4. Image classification of La Montaa portion of a) Landsat 5 TM image02/05/96; and b) Landsat 7 ETM 03/27/00

PAGE 46

34 The 56 clusters generated in the post classification comparison wer e merged to produce a change image map (Figure 5). This map was merged into 11 classes using the following criteria (the different colors in Table 1): 1) upland forest, 2) lowland forest, 3) recovered vegetation (including vegetation that has recovered to either upland or lowland forests) 4), new quemadal (including areas that have undergone a transition from lowland or upland forests to quemadal 5) same quemadal/acahual (including areas that have remained as quemadal and areas that are recovering from d isturbance), 6) new agriculture/urban (areas that have been cleared either from upland, lowland, or from quemadal including new urban areas), 7) Same agriculture/urban (areas that have remained under agriculture, acahual or urban), 8) cloud mask; 9) aquat ic vegetation/savanna (this includes the areas previously identified as aquatic/savanna in the two images and areas that at either of the two dates were covered with water, suggesting that they may be aquatic vegetation or savannas and whose spectral signa ture when dry, is confused with acahual quemadal or agriculture (agriculture is very unlikely to occur in areas inundated for long periods); 10) water; and 11) suut (these areas were tentatively classified as suut representing lowland forests that are i nundated for long periods of time, since they were classified at either date as forest and in the other one as either water or aquatic vegetation. The three last classifications (aquatic vegetation/savanna, suut and water) are indicative of the complex h ydrology of the area. Approximately 35.5% of the area classi fied either as upland or lowland forest was misclassified, as shown in Table 2 in the matrix boxes belonging to upland/lowland and lowland/upland. This probably resulted from environmental variations given by differences in the scene dates and from the l ow quality of the 1996 scene. Also, a vegetation sampling carried out in the area (Chapter 6) reviled that 90 species were common for both forest types, out of a total of 115 for lowlands and 112 for uplands, which could result in

PAGE 47

35 Table 2. Change detection matrix obtained by combining classifications made of the La Montaa portion of Landsat 5 TM, 02/05/96 and Landsat 7 ETM 03/27/00 scenes. Numbers are in hectares difficulties for differe ntiating spectral signatures. Other areas such as those classified as quemadal or acahual, may als o contain some errors. However, these errors do not affect our interpretation of land use change since the areas classified as agriculture and urban are unlikely to be in error, as their shape (e.g., squares, rectangles) and position in the landscape (e.g ., near roads, settlement) are other good indications of their land use/cover types. For Figure 6 upland and lowland forests were classified according to the 2000 image.

PAGE 48

36 Figure 5. Change map created by combining classification data of the La Montaa portion of Landsat 5 TM, 02/05/96, and Landsat 7 ETM 03/27/00 scenes.

PAGE 49

37 Four percent of the area was cleared for agriculture or developed as urban areas duri ng the four year period (Table 3). This includes areas that were in the early stages of recovery ( acahual ). Additionally, 2.6% of the area remained as agriculture, urban, or acahual Together, areas in 2000 that were represented as active agriculture, u rban development, or early stages of recovery ( acahual ), added up to 6.6% of the total area. Areas in advanced regeneration stages or that had undergone a fire ( quemadal ) represented 4.5% of the area, of which 3.5% were recently disturbed. Areas that had recovered to either upland or lowland forests represented 2.5% of the area. Not considered in the classification was 1.5% of the area masked out as clouds or shadows. Table 3. Changes in pixel count, area, and percentage of total area of land use/ cover types obtained from a post classification analysis of Landsat TM, 02/05/96 and Landsat 7 ETM, 03/27/01. Results from the surface analyses show that although proximity to roads and settlements may be a factor in forest clearing, most of it is occur ring as a function of soil type. This corresponds with information obtained from local people who stated to choose their agricultural sites mostly on the basis of environmental conditions (soil type and vegetation cover), although accessibility to town wa s also considered important.

PAGE 50

38 Buffers made at 1, 3 and 5 km from roads show that disturbance is not very related to proximity to roads (Figure 6 A). Approximately 7.7%of total area falls inside a 1 km buffer from major roads and only 1.7% of this area is d isturbed, which corresponds to 16.7% of the total disturbed area in the image. A 3 km buffer accounts for about 26% of total area of which 3.9% is disturbed (38% of total disturbed area). A 5 km buffer from roads accounts for about 40% of the total area of which only 5.4% corresponds to disturbed areas, accounting for only 52% of total disturbed area in the image. It should be taken into consideration that the roads are relatively new and that people have been established in the area for at least a centu ry. Buffers from settlements were made at 3, 5 and 8 km (Figure 6 B). A 3 km buffer from settlements accounts for almost 13% of total area of which only 2..5% is disturbed, representing 24.3% of total disturbed area. A 5 km buffer from roads accounts f or 28.5% of total area of which 4% is disturbed, corresponding to almost 40% of total disturbed area. An 8 km buffer accounts for almost 50% of total area of which little over 6% is disturbed, representing 60% of total disturbed area in the image. While proximity to settlements may be a factor in establishing agricultural areas, people do not limit themselves to areas close to their towns and almost 40% of total disturbed areas falls outside the 8 km limit from settlements. People in the area report that very often they establish their agricultural sites in areas where they have to camp, sometimes for over two weeks.

PAGE 51

39 Figure 6. La Montaa area, Campeche, Mxico. Buffers made at: A. 1, 3 and 5 km from roads; B. At 3, 5 and 8 km from roads; and C. At 0.5 and 1.0 km from lowlands

PAGE 52

40 Buffers from lowlands indicate that people largely choose their agricultural sites based on soil type since up to 53% of the total disturbed area falls within a 0.5 km buffer from lowlands, and up to 67% within a 1 km buffer (Fi gure 6 C). This corresponds with information provided by local people that indicated the yash hom and kakab soil/vegetation types were preferred for milpa establishment (Figure 2, Table 1). Discussion Clearing Rate If forest clearing at La Montaa is ass umed to have occurred equally over the four year period, an estimated annual clearing rate of 1% and an annual regeneration rate (also divided between the four year period) of 0.635% provides a net deforestation figure of 0.365% per year. This figure cons iders only the recently disturbed areas. Additionally, 2.6% of the area previously disturbed has not been allowed to recover, probably remaining under continuous use. If the latter figure is partitioned into the four year period, an additional 0.65% prov ides a clearing rate of a little over 1% per year. Another 4.5% of the area has been disturbed as quemadal of which most, 3.5%, has occurred over the four year period. Areas under quemadal either resulted from wildfires, or were burned or felled by loca l people, indicating that they may be in a transition to a more intensive use (when preparing agricultural areas from mature forest people in the area usually fell trees at the beginning of the year, during February or so). Quemadales may also be areas in an advanced stage of regeneration. They may also be misclassified and could be areas of very dry forest that shed most of their leaves (considering that the 2000 image was taken during the peak of the dry season).

PAGE 53

41 Turner et al. (2001) reported annual d eforestation rates caused by anthropogenic disturbance in the region of the Calakmul Biosphere Reserve to be between 0.32 and 0.39 % for the years 1969 to 1997. Sader et al. (1994) reported annual rates of forest clearing at the northern Peten region of G uatemala to have been 0.4 % at the end of the 1980s, with a regeneration rate of only 0.03%. They consider the clearing rate as low, which they attribute to the inaccessibility of the interior of the Maya Biosphere Reserve. Nevertheless, in a second stud y, Sader et al. (1997) reported annual clearing rates at the buffer zone of the Maya Biosphere Reserve (in the western and southern borders) to have been as high as 3 % at some sites during the early 1990s. They attribute the higher forest clearing rates to agricultural expansion and accessibility (e.g., roads, pipe lines, and rivers). According to an FAO report from 1993, Mexico ranks first worldwide in deforestation, with an estimated annual rate of 1.3 % (Cairns et al. 1995). Considering the different forest types, tropical deciduous forests are second in their rate of forest clearing, after tropical evergreen forests, with an estimated figure of 1.9 % per year. The causes of deforestation in Mexico are multiple, with forest conversion to agricultural uses, particularly cattle ranching, the leading factor. Forest fires, hurricanes and commercial logging are also considered important factors (Cairns et al. 1995). Although the estimated 1% annual forest clearing rate at La Montaa is lower than the aver age estimated figure of 1.9 % for deciduous tropical forest in other areas of Mexico and Guatemala, it is still higher than the rate estimated for the interior of the Calakmul and Maya Biosphere Reserves ( .4%) (Sader et al. 1994, Turner et al. 2001). Th is suggests that forest clearing may be intensifying, although further research would be

PAGE 54

42 needed to understand more about the regeneration process of these forest ecosystems, and consequently more about their dynamics (e.g., through conducting a similar ana lysis utilizing images taken at a previous date particularly before the construction of the Dzibalchen Xpujil road). Land Use Change at La Montaa There are factors that indicate that land use may be intensifying and that management objectives for the l ocal populations are changing. The milpa is traditionally established in upland forest preferably of the yash hom kakab and kankab types (Figure 2) (note that uch of the agricultural area in the images are shown bordering lowlands). None of it is estab lished in akalche forests, since the probability of losing the crop by inundation is high. Nevertheless, local people report that the recent trend is to establish the milpa in akalche areas and, after the first years harvest, to plant grass. This new trend changes the traditional slash and burn system, where forests are left to regenerate, to a system where fallowed areas are replaced with grass for rearing cattle. It can be observed in the images that a large portion of the newly cleared areas is fou nd within the lowland forest areas. Also the 2.6% of the area that showed continued use over the four year period suggests that forests are not being left to recover (this could be further tested through analyzing images of more dates). Cattle raising a t La Montaa is an activity that has been gaining in importance in recent years. Government programs have aided in its promotion, and even though most people have only a few head of cattle, some individuals are investing much of their capital and labor in to increasing their herds and pasturelands. Cattle raising is currently perceived as an activity important for accumulating capital and gaining social status, as well as a good source for quick cash. In recent times, the government has provided

PAGE 55

43 considera ble credit to some individuals for establishing cattle ranches (Porter 1995). There are also people who invest in pastures even though in some cases they still do not own cattle (Galletti 2000). Cattle raising at La Montaa has strong implications for for ests in the region and may be the reason for the forest clearing rate observed here. According to Masera et al. (1997 citing FAO 1990) the deforestation process in Mexico is weakly correlated with population growth and, although the key driving forces of deforestation are varied, direct or indirect subsidies to cattle ranching have been one of the outstanding factors. Changes in the traditional agricultural systems may be posing more rapid threats to the current forest ecosystems at La Montaa than popula tion growth. However, people at La Montaa carry out other forestry related activities, which are economically important and which favor (or could favor) the conservation of a forested landscape. These activities include chicle production, timber extracti on, beekeeping, traditional agroforestry systems, and the traditional milpa Chicle production continues to be an important economic activity in the area. This activity is favored and at the same time favors the conservation of forests, where large chic le trees are found. Although restrained by market fluctuations and by government permits, it is potentially important for the economy of the region. Timber harvesting is limited by impoverished forests, where the most commercially valuable species (mahog any and Spanish cedar) have been extensively extracted. Nevertheless, timber harvesting still represents an important potential economic activity in the region, which given the necessary conditions could favor forest conservation. Species with commercial potential, such as granadillo ( Platymiscum

PAGE 56

44 yucatanum Standley ), siricote ( Cordia dodecandra A.DC ), guayacan ( Tabebuia guayacan ), and chaka ( Bursera simaruba (L.) Sarg. ), are currently being sold from forests at La Montaa. However, timber extraction is an activity limited by government permits and markets. Currently, there is no organization or structure in which the local population (or the forests) can benefit, and timber extraction is being carried out with no formal management plans. Usually, wood is sold as standing volume to companies that bring their trucks and equipment to extract the timber, many times illegally. Local people mainly receive money for the labor of felling the tree, but with not much other gain, and most of the time they are una ware of the volumes that are being extracted (Porter 1995). Timber extraction carried out in this way is clearly detrimental to forest conservation. Cairns et al. (1995) point out that the deforestation process in much of the Mexican forests typically be gins with timber extraction in this unorganized way. The consequences are that the economic potential of forests is reduced after valuable timber species are depleted and the associated road infrastructure encourages colonization and agricultural expansio n. However, timber extraction could become an incentive for forest conservation, provided the necessary conditions are created. Honey production is currently an important income generating activity in the region. Although beekeeping is a traditional acti vity, beekeeping with Apis mellifera L. is relatively new and has been developing into an important economic activity in recent years, establishing market ties with national and international industries. In the region, beekeeping with Apis bees has been f avored by a mosaic like landscape formed by patches of different vegetation ages and types, rich in melliferous flora. The extent to which this activity favors (or could favor) forest conservation is an important question

PAGE 57

45 that should be addressed in the s earch for sustainable management alternatives for the development of the region (see following chapters). The traditional milpa as well as different agroforestry systems including home gardens and the parcela agroforestal 1 are important subsistence activi ties for people at La Montaa. These activities are inherited traditional practices based on solid environmental knowledge that result in a high degree of biological diversity. The slash and burn milpa system as practiced traditionally can be regarded as an agroforestry system where fallow and regeneration of the vegetation are part of the management practices (Gomez Pompa 1987, Sohn et al. 1999) and where multiple products and services are obtained. While the milpa system itself does not guarantee fores t conservation or a sustainable use of natural resources, as populations grow, forests are converted to other land uses and new technologies are adopted (e.g., use of chainsaws, use of agrochemicals, seed sources, etc.). This must be balanced by the wealt h of knowledge that the Maya people have regarding vegetation, soil, hydrology, and climate that should not be disregarded when management alternatives are sought for forest conservation at La Montaa and similar areas in the tropics. Conclusions The fores t landscape of La Montaa has been shaped by a range of factors, that include natural forces such as hurricanes and fires and human induced disturbances, through the long span of history of occupation in the area. Human activity in the area of 1 A traditional agroforestry system similar to that of homegardens but which may or may not be established next to the house, and that currently has been including the establishment of different native and non native fruit and tim ber trees encouraged by NGO.

PAGE 58

46 La Montaa has been constant, with episodes of higher and lower occupation. Since recovery of the forests from the high populations and intensive use during the Classic Maya period, forest cover has been steady. Human activities in the area until recently have been related to the use of timber and non timber products, guided by internal and external interests that have molded the current status of the forested landscape. The traditional milpa agriculture, as well as other traditional agroforestry systems of the May a, developed into management practices where regeneration of forests was part of the systems, becoming endogenous stresses to the forest ecosystems and very likely affecting the structure and composition of the vegetation and animal communities through t ime. Chicle extraction provided an incentive for the legal establishment of forestry ejidos in the area. Timber extraction, on the other hand, left an impoverished forest, depleted of the most valuable timber species. Today, land use activities are c hanging towards more agricultural production, with potentially large implications for the forested landscape of La Montaa. This analysis of the four year period of remotely sensed imagery indicates that forest conversion may be intensifying, as cattle ra ising becomes more common; communication at the regional, national and international levels increases; and the population grows. Although the economy in the region is still greatly dependent on the management of forest resources, changes are occurring whic h are visible at the landscape level. The question now is how will current decision making regarding land use affect the future structure and composition of forests at La Montaa. The following chapters will analyze beekeeping as an activity that could f avor the conservation of a forested landscape in this area.

PAGE 59

47 CHAPTER 3 BEEKEEPING IN THE MAYA AREA: HISTORY AND CURRENT STATUS Introduction Beekeeping in the Yucatan Peninsula has been a traditional activity since ancient times. Maya people are known to have practiced intensive management of stingless bee species, including Melipona beecheii Benneth. In Mayan philosophy, humans and bees were very related; the Zayawinnicoob or first humans, were considered to be very industrious people and were represented as having bee like eyes (Chemas and Rico Gray 1991). Although melip oniculture (beekeeping with native stingless bees of the subfamily Meliponinae) is now seldom practiced in the Maya area, beekeeping with Apis bees has become an important activity in the region. Beekeeping is now carried out with modern techniques (movab le frames) and with bees of the species Apis mellifeara L. Over 40 percent of Mexicos honey production comes from this region, contributing to Mexicos importance as one of the worlds leading producers and exporters of honey (Angulo Carrera 1992, Crane 1990, Echazarreta et al. 1997, Guzman Novoa 1996). A literature review is presented to address issues related to the honey industry in Mexico and specifically in the Yucatan Peninsula, considering its historical context and the changes that have resulted f rom the Africanization of Apis mellifera bees. Findings by different authors that have addressed the Africanization phenomena are provided in order to illustrate limitations and opportunities for beekeeping in different contexts.

PAGE 60

48 Beekeeping Industry in Co ntext: Mexicos Place in World Honey Production and Relevance of the Yucatan Peninsula. The beekeeping industry provides society with many important products and services. The traditional products, honey and beeswax, have been used since ancient times in all parts of the world (Crane 1990). Honey, the final product of nectar collected and modified by bees, is used as a sweetener, either served at the table, or in the food and drink industries. It also has many uses in medicine and cosmetic manufacturing. Beeswax, which is secreted by worker honey bees from wax glands, was more widely used in the past, before synthetic products were substituted (e.g., paraffin). Nevertheless, beeswax is still used as a component of cosmetics and pharmaceuticals, as well as for other minor uses. According to Crane (1990), most of the beeswax that is produced today is wasted, some of it is used in traditional ways, and a small part is used in the industry. A big portion of beeswax is re used in the beekeeping industry for producing comb foundations. Other products derived from beekeeping began to be produced after World War II, when low prices for honey, replacement of beeswax by other substances for many industrial purposes, new technologies, and a widespread new interest in health foods and dietary supplements led beekeepers to diversification. These products include pollen, propoli, royal jelly, bee venom, and bee brood (Crane 1990). The beekeeping industry is also important for providing pollination services for agric ulture, and has become an integral part of production technology worth millions of dollars in industrialized countries (Basualdo 1996). Currently, the pollination of fruit and seed crops by bees (including bees other than honey bees) provides a greater ec onomic benefit to agriculture

PAGE 61

49 world wide than the beekeepers income from honey and other hive products (Crane 1990). The basis of todays beekeeping industry is the honey bee Apis mellifera L. (order Hymenoptera; family Apidae; subfamily Apinae), even in areas of the world where this species is not native (such as the American continent). Honey is also produced in different parts of the world by other Apis species and by different species of the subfamily Meliponinae (the stingless bees), but mostly for l ocal consumption or regional trade (Crane 1990). Honey production from Apis mellifera is the most widespread and has an established world trade market. Worldwide honey production from this species reaches numbers that approximate one million metric tons per year (Angulo Carrera 1992). Mexico is one of the principal countries in the world wide beekeeping industry. In 1991, Mexico was the fourth honey producer, after the former USSR, China and the United States (Snchez Vazquez and Colli Ucan 1992). Data obtained from the Centro de Estadstica Agropecuaria 1 report an average annual honey production in Mexico of 60,000 metric tons of honey from 1980 to 1999, and an average annual production of 4,217 metric tons of beeswax for that same period. According to Angulo Carrera (1992), Mexico was the leading exporter for the period of 1974 to 1984. In 1995, Mexico was the ninth honey producer, after China, USA, Argentina, the former USSR, and four other Asian countries, which together accounted for 46% of that ye ars world production, and the fourth largest exporter, after China, Canada and Argentina (Troncoso 2000). The use of bees for pollination has also gained importance in some states of Mexico, mainly in the

PAGE 62

50 northern states, and has been estimated to have a greater value than honey production and beeswax for that area (Guzman Novoa 1996). The southeastern part of Mexico is one of the most important honey producing regions. This honey producing area is comprised of the Yucatan Peninsula (Campeche, Quintana R oo, and Yucatan). Representing only 8% of the national territory, this region accounts for 30 to 40% of annual national honey production (Echazarreta et al. 1997). The relevance of this area as a honey producing region has been explained by Guzman Novoa (1996) to be the result of its exuberant native melliferous/polliniferous flora, and because of the beekeeping tradition of the Maya people, who have been keeping bees since ancient times. In Mexico, honey production is very important for the rural popul ation as an alternative income activity and 80% of the beehives are owned by small farmers (Guzman Novoa 1996). Therefore, the honey industry represents an important source of income and jobs for low income families in the rural sector. Beekeeping has al so been considered an alternative land use in areas where forests are to be conserved and as a way to provide added value to secondary vegetation and mature forests (Acopa and Boege 1998). Being a profitable non timber forest product with an established m arket, beekeeping is considered to have great potential in protected regions such as the Calakmul Biosphere Reserve and its surrounding areas (Acopa and Boege 1998, Galletti 2000). 1 The Centro de Estadstica Agropecuaria (Agricultural Statistical Center) is part of the Secretara de Agricultura, Ganadera y Desarrollo Rural (SAGAR), the Mexican Secretariat o f Agriculture and Rural Development.

PAGE 63

51 History of Beekeeping in the Maya Area. Although there are no native honey bees ( Apis ) in the American continent, Mesoamerican cultures managed stingless bees of the genera Melipona and Trigona (order Hymenoptera; family Apidae; subfamily Meliponinae) for over 3,000 years (Guzmn Novoa 1996). Of particular importance is the mana gement given to the species Melipona beecheii Bennett by the Mayas. Management of the xunaan kab ( xunaan = queen or princess, and kab =bee or honey), the Mayan name given to Melipona beecheii developed into an extensive bee culture, where honey and wax w ere used for trade, religious ceremonies, and medicinal purposes (Sanford 1973). Bishop Diego de Landa (who arrived in Yucatan in 1549) described the life and customs of Maya Indians, including beekeeping and the different festivities and offerings dedica ted to bee gods (Weaver and Weaver 1981). The Troano Codex also refers to different aspects related to beekeeping in the Yucatan Peninsula (Labougle and Zozaya 1986). Weaver and Weaver (1981) describe some of the bee related festivities dedicated to gods that were still practiced until recently by current Mayas. Honey produced by the xunaan kab is still (in few cases) used for making balche an alcoholic drink used in religious ceremonies. Balche is prepared using the bark of the balche tree ( Loncho carpus longistylus Pittier ) and honey made by Melipona beecheii Honey was paid as tax since pre Hispanic times and was the subject of an intense commerce that took place from Tabasco to different areas of the Peninsula, by sea to Honduras and Nicaragua, and to the west with the Mexica Empire. Mayas traded beeswax and honey probably for cacao seeds and precious stones (Labougle and Zozaya 1986). After the Spanish conquest, honey and wax produced from xunaan kab were paid as taxes to the Spaniards. Durin g this time, beeswax was particularly important in

PAGE 64

52 religious activities carried out by the Spaniards. Honey as a sweetener was gradually replaced by sugarcane. Nevertheless, the beeswax produced from stingless bees supplied a vast area of the Spanish Emp ire practically for the entire duration of the Colonal period (Labougle and Zozaya 1986). Labougle and Zozaya (1986) reported taxes paid to Spaniards by the Maya population in the year1547; of 173 settlements in the state of Yucatan, only 5.8% did not pay with honey or beeswax. Total production in this area during that year approximated 29,300 kg of wax and 3,300 kg of honey; every 20 persons contributed 12 kg of wax and every 295 persons, 12 kg of honey. European honey bees ( Apis mellifera ) were not int roduced into the Yucatan Peninsula until the end of the 19 th and beginning of the 20 th centuries. Although there are reports of their presence in the center of Mexico at the end of the 18th century, this did not prove to be the introduction to the Yucatan Peninsula. Modern apiculture (carried out with Apis mellifera bees and using movable frames) began to be practiced after the 1920s (Labougle and Zozaya 1986). In the 1930s and 1940s, large amounts of Apis honey started to be produced in the Yucatan Peni nsula (Chemas and Rico Gray 1991). During the 1960s and 1970s beekeeping with Apis mellifera by the Maya population gained importance, becoming one of the principal economic activities for the rural sector (Echazarreta et al. 1997). Beekeeping in the area is now carried out primarily with Apis bees using modern techniques (Sanford 1973), although management of Meliponinae bees is still practiced by a decreasing number of Maya people. The arrival of the Africanized bees (see below) into Mexico in the last half of the 1980s has brought different challenges and opportunities for beekeepers in the area.

PAGE 65

53 Beekeeping with the Africanized Bee The Africanization of the European races of Apis mellifera in the American continent in the last fifty years has resulted i n profound changes in the beekeeping activity in a very vast area of tropical and subtropical America. These changes bring different challenges and opportunities for beekeepers that will be briefly discussed in the next section. These changes also raise important questions regarding the role of exotic honey bees as pollinators in natural areas and the possible effects that may occur on the native flora and fauna populations. Africanized Bee Africanized bee is the name given to races of European Apis melli fera bees that have hybridized with an African race of Apis mellifera ( Apis mellifera scutellata Ruttner) in the American continent (Crane 1990, Daly 1991). The Africanization process began in 1956, when 26 absconding swarms of Apis mellifera scutellata e scaped from an apiary in Camaquan, So Paulo, Brazil. These bees had been brought from Africa in order to conduct breeding experiments. By 1991 Africanized bees were found in many different habitat types in an area about 20 million km 2 that extended from northern Argentina to Mexico (Roubik 1991). In the last decade these bees also reached parts of the southern United States and in all likelihood will continue to expand in North America. The Africanized bee has been able to quickly occupy most of tropica l and subtropical America. This astonishing success is in part due to a highly evolved capacity to exploit unpredictable and varied resource conditions and in part due to its unique reproductive biology (Ratnieks 1991). It has been calculated that the c olonizing front of Africanized bees has moved at an average rate of 300 to 500 km/year (Crane 1990, Otis 1991).

PAGE 66

54 The African bee that was brought to Brazil was adapted to arid and semi arid environments ranging from tropical East Africa to subtropical Sout h Africa. The environmental conditions in this area are varied, but the major selection pressure for the bees was likely the unpredictability of rainfall which influences flowering. On the other hand, the European bee evolved in a much more predictable e nvironment where annual snow melt provided a secure source of moisture for flowering to occur (Rinderer and Hellmich 1991). This adaptation to unpredictable conditions has provided a great advantage to the Africanized bee when colonizing new areas. A summ ary of the characteristics of the reproductive biology that have helped the colonizing success of the Africanized bee (and that affects traditional beekeeping schemes) is as follows (Otis 1991, Ratnieks 1991, Rinderer and Hellmich 1991): Swarming This is the process by which bees reproduce (through colony fission). European colonies usually swarm once a year, while Africanized bees can swarm up to four times annually. Absconding. Africanized bees tend to abandon their nests or abscond. Absconding occu rs either as a result of disturbance to the colony, or as a result of shortages in resources. The latter occurs in periods of extended dearth (when there are no flowering resources or water available), or when other resource rich areas have been identifie d. European bees rarely abscond; it is more common that the colony perishes when faced with resource shortages or external threats. Differential reproductive rates Africanized bees develop large populations in short periods in response to the existence of harvestable resources (as a result of their adaptation to unpredictable resource availability). They have higher fecundity and shorter development times than European bees. Social reproductive parasitism This is a term that includes different behav ioral characteristics of African races of Apis mellifera A form of social reproductive parasitism is the capacity of reproductive females of this subspecies to take the place of the queen of a different colony, accelerating Africanization. Another form of social reproductive parasitism is that drone populations (drones are male bees whose basic function in the colony is to mate queens) are larger in African colonies. Therefore, during mating flights there is a greater probability that the queen will mat e with drones of the Africanized population.

PAGE 67

55 As a result of these collective characteristics, Africanized bees have now spread to most of the American continent, replacing European honey bees. Africanized honey bees are now found not only in areas where traditional beekeeping existed, but also in natural areas forming numerous feral populations (wild colonies), where before they were relatively scarce (Michener 1973, 1975). One of the characteristics that makes the Africanization process even more striki ng is the fact that African traits of the Apis mellifera subspecies have persisted almost entirely. Genetic studies indicate that Africanized bees resist changes in gene frequency toward the European type through partial reproductive isolation between Eur opean and Africanized bees. This partial reproductive isolation operates in ways that perpetuate the African ancestry and give the Africanized bee further advantage over European bees (Daly 1991, Fletcher 1991, Ratnieks 1991, Spivak 1991). New Challenges and Opportunities Changes in honey bee populations have brought about many challenges to beekeepers. Behavioral characteristics of the Africanized bee, such as their defensive nature and their tendency to swarm and abscond, have destabilized beekeeping an d honey production. At the same time, it has become common that after an initial period of adjustment, the Africanization actually favors beekeeping, at least in some tropical and semitropical settings. The behavioral trait of Africanized bees that has made them widely known among the common public, and one of the biggest challenges for beekeepers, is their defensive behavior (Breed 1991, Rinderer and Collins 1991). Africanized bees show greater sensitivity to colony disturbance, have a greater ability to communicate alarm within and

PAGE 68

56 between colonies, and have a greater capacity to respond quickly by mounting massive attacks on intruders (Michener 1973). This defensive nature of the Africanized bee makes their management even more difficult. Beekeepers have had to understand different aspects of the Africanized bee biology in order to modify management practices and adapt to these new conditions. To deal with the defensive nature of the Africanized bee, beekeepers have learned to relocate apiaries away from homes, barns, animals and roads (Kent 1991). They also now need to use more appropriate clothing. Beekeepers have learned when to work in the apiary, since the temperament of a colony varies depending on weather and foraging conditions, food stores, prior treatment, colony size, etc. (Michener 1973). The tendencies of the Africanized bee to frequently swarm and abscond have made beekeeping a more intensive activity, requiring more visits to apiaries than before. Beekeepers have learned to capture s warms in bait hives as a procedure to increase colony numbers and reduce wild populations (Helmich and Rinderer 1991). Beekeepers have also learned to select toward mildness by destroying and requeening the most defensive colonies (Michener 1975). Africa nized populations do not represent a well defined phenotype, but rather display a wide range of behavioral characteristics, ranging from relatively gentle and manageable to extremely defensive and difficult to manage (Spivak 1991). Spivak (1991) provides examples from Costa Rica where manageable and productive Africanized colonies with intermediate characteristics have been successfully maintained by beekeepers who have modified their management practices and requeen or kill undesirable colonies.

PAGE 69

57 According to Michener (1973), most of the beekeepers in Brazil have learned how to handle the new honey bee and actually prefer it in spite of its drawbacks, because of its high productivity. He found that Africanized bees in Brazil are reported to produce more ho ney per colony than European bees. According to Gonalves et al. (1991), beekeepers in Brazil, with the aid of government technicians and university scientists, have been able to adapt. Brazil began producing large quantities of honey and became one of t he largest honey producers in the world only after the introduction of the Africanized bee. Most beekeepers in Brazil are happy working with Africanized bees because it is easy to obtain bees and because of their higher productivity (Gonalves et al. 1991 ). Kent (1991) reviewed the impact of Africanized bees in Peru, comparing areas with and without Africanized bees. He concluded that average colony production was consistently greater in Africanized regions. He found that it is common after an initial pe riod of adjustment when honey yields may decline, that yields then tend to increase notably. Small beekeepers tended to abandon beekeeping, while commercial beekeepers with greater financial resources and technical knowledge were more likely to continue. To adapt, beekeepers became better equipped and more progressive in their management practices as well as more involved in information networks to share experience and knowledge. Spivak (1991) concluded that in different regions of Costa Rica, the African ized bee caused beekeepers who were not willing to modify their practices and adapt, to abandon beekeeping. Nevertheless, other beekeepers were able to reduce the impact of the Africanized bee by employing good management and selection practices. She sta tes that if beekeepers are willing to follow advice from other beekeepers and researchers who

PAGE 70

58 have had positive experiences with Africanized bees, the Africanization process can have a positive effect overall. Hellmich and Rinderer (1991) show that the int roduction of the Africanized bee to Venezuela greatly decreased beekeeping initially. However, after a decade, beekeepers began to learn to manage the Africanized bee. These authors conclude that a new generation of beekeepers has emerged in Venezuela, b etter informed and prepared. However, they report that honey yields were still greater before the Africanized bee, attributing lower present productivity to the increased competition for nectar sources by the numerous feral colonies. It has been the gener al case that beekeeping in many areas of South and Central America has been favored by Africanization. However, it is important to note that beekeepers that have been able to adjust and take advantage of the new situation have been those with the capacity of acquiring equipment and becoming better informed, either by technical training or through different media channels (by personal means or through government programs). In Mexico, the effects of Africanization on total national honey production have been minimal, basically because beekeepers have had time to adapt to the new management conditions (Echazarreta et al. 1997). Governmental programs such as the Programa Nacional para el control de la Abeja Africana (PNPCAA) formed as an agreement between the Agricultural Secretary (at that time SARH 2 ) and the United States Department of Agriculture (USDA), has aided in the mitigation of the impact of Africanization (Snchez Vazquez and Colli Ucan 1992). Data from the Centro Estadstico Agropecuario show that national annual honey production decreased slightly

PAGE 71

59 in the 1990s, averaging approximately 58,000 metric tons from 1990 to 1999, as opposed to an average of 60,000 for the previous decade. Part of this decrease may be the result of Africanization, but othe r factors such as deforestation, intensive use of agrochemicals, national and international low prices, market competition, and the spread of Varroa jacobsoni (an exotic mite that affects honey bees and that has been spreading in recent years negatively af fecting honey production), have also been used to explain such decreasing tendency (Angulo Carrera 1992, Snchez Vazquez and Colli Ucan 1992). Nevertheless, certain areas, such as La Montaa (see chapter 4), have actually increased honey production in rec ent years, favored by Africanization and other factors. Conservation Concerns There is increasing concern about possible effects that introduced bee species may have on native fauna and flora. Exotic honey bees may compete with other species of flower fee ding animals, such as nectar feeding birds and other bees (Sugden 1996, Sugden et al. 1996). One concern is the effect that the increasing number of Africanized bees, particularly as feral populations, may have on the natural assemblage of pollinators (Bu chmann 1996, Corbet 1996). Not only is there evidence that the Africanized bee displaces native pollinators (Roubik 1978, 1996a, 1996b; Roubik et al. 1986), but also that they may be ineffective at pollinating native flowers and may interact in complex wa ys with native pollinators, reducing the amount and efficiency of pollen transfer (Kearns and Inouye 1997). Although the direct effect that this may have on the faunal and floral compositions in the mid and longer terms is hard to measure (Buchmann 1996, Roubik 1996a, Sugden 1996), it is unlikely that it will have no effect. It is 2 Secretara de Agricultura y Recursos Hidrulicos

PAGE 72

60 important to keep this considerations in mind when promoting beekeeping as an activity compatible with forest conservation. Research regarding the biology, reproduction and eco logical relation with other organisms of stingless bees in the Yucatan Peninsula and in Mexico is only beginning (Quezada Euan 2001). Not only does beekeeping with stingless bees represent part of a cultural wealth of indigenous populations in Mexico, inc luding the Maya area, but it is also considered to have great economic potential for honey production and for pollinization of commercial products. Appropriate technology could make this activity a viable and profitable one (Quzada Euan and Gonzlez Acere to 1994) and should be promoted in the context of current honey production. Conclusions Apiculture is an important economic activity in Mexico and the Yucatan Peninsula is one of the most important areas within the country (Echazarreta et al. 1997, Snchez Vazquez and Colli Ucan 1992). In part, this results from the rich natural melliferous and polliniferous flora found in the area (Chapter 5) and from the beekeeping tradition of the Maya population. Beekeeping with native bees of the subfamily Melinponin ae began with the Mayas over 3000 years (Guzman Novoa 1996). Intensive management of certain species, particularly Mellipona beechei resulted in a bee culture important in religion, medicine, food, and trade (Labougle and Zozaya 1986, Sanford 1973). Dur ing the 20 th century, beekeeping with Apis mellifera started in the Yucatan Peninsula, increasing in importance throughout the century. During the 1960s and 1970s beekeeping with Apis mellifera by the Maya population also gained importance, becoming one o f the principal economic activities for the rural sector (Echazarreta et al.

PAGE 73

61 1997). During the last half of the 1980s, the arrival of the Africanized honey bee resulted in new challenges and opportunities for beekeepers, as it did throughout most of the A merican continent (Spivak et al. 1991). In the Yucatan Peninsula and the rest of Mexico, beekeepers have had time to adapt to the new management conditions and the effects of Africanization have been minimal. While beekeeping in the Yucatan Peninsula with Apis mellifera results in a viable and profitable economic activity, more efforts should be made to promote beekeeping with stingless bees. Future studies should address the impacts that beekeeping with Africanized Apis mellifera may have on native polli nators, and ultimately to the function and structure of flora and fauna.

PAGE 74

62 CHAPTER 4 SOCIO ECONOMIC IMPORTANCE OF BEEKEEPING IN THE MAYA FORESTS: THE EXAMPLE OF THE LA MONTAA AREA Introduction Beekeeping in the Yucatan Peninsula has been a traditional activity since ancient times. Maya people are known to have practiced intensive manage ment of stingless bee species such as Melipona beecheii Benneth. Although meliponiculture (beekeeping with native stingless bees of the subfamily Meliponinae) is now seldom practiced, beekeeping with Apis bees is an important activity in the region. Beek eeping is now carried out with modern techniques (movable frames) and with bees of the species Apis mellifer L. Over 40% of the national honey production comes from this region, contributing to a large honey production industry with world wide importance, since Mexico is one of the leading producers and exporters of honey (Angulo Carrera 1992, Crane 1990, Echazarreta et al. 1997, Guzman Novoa 1996). In the Yucatan Peninsula, about 80% of the beehives are owned by small farmers, making beekeeping an impor tant source of income and jobs for the rural sector (Guzman Novoa 1996). In protected regions such as the Calakmul Bioshpere Reserve and surrounding areas, honey production is considered to have great potential for development (Acopa and Boege 1998, Galle tti 2000). The Maya landscape hosts a very rich melliferous and polliniferous flora found in the different forests types, yielding abundant nectar and pollen for honey production, and representing an important economic resource (Rico Gray et al. 1991).

PAGE 75

63 The objective of this research it to analyze the socio economic implications of beekeeping at La Montaa, an area found north of the Calakmul Biosphere Reserve. Beekeeping in this area is the number one income generating activity (Galletti 2000, Porter 19 95). Through conducting research with beekeepers from the area for a period of two honey productive season, information was sought to understand the socioeconomic importance of beekeeping, the benefits that the local families and communities are obtaining from this activity, as well as the strengths and limitations of honey production and marketing. Methods Study Site La Montaa is an area north of the Calakmul Biosphere Reserve in the state of Campeche, Mexico. It is formed by a group of eight ejidos oc cupying an area of approximately 185,000 ha, and having a population of about 2700 inhabitants mostly of Yucatec Mayan origin (Galletti 1999). The type of climate for the area has been determined, according to the Kppen climate classification system as m odified by Garcia (1973), to be subhumid, warm tropical (Flores and Espejel 1994). The mean annual temperature is above 22C and the mean annual precipitation is 1,223 mm 1 The most dominant vegetation type is medium stature subperennifoleous forests, wh ere the mean height for trees is 25m and where from 25 to 50% of the tree species shed their leaves during the dry season. This type of vegetation is mixed with high stature subperennifoleous forests, low stature forests that inundate periodically, aquati c vegetation, and secondary vegetation in different successional stages (Ucan et al. 1999).

PAGE 76

64 Soils in the area have been formed mostly by the dissolution of calcareous rocks that over time has resulted in a karst topography (Flores and Espejel 1994). Drain age is subterraneous with no superficial flows of water, except in the rainy season during storms and inundations when rain can be very intense and water is transported in temporary surface flow channels (Gates 1999). Research Conducted The field study la sted from November 1998 to June 2000 and included two honey production seasons. The research was facilitated by Oxfam/Comadep 2 a non governmental organization working in the area, which was assisting a group of beekeepers to develop a legally established cooperative to commercialize and increase honey production. Formal and informal interviews with 94 beekeepers, many of them members of the cooperative formed with the assistance of Oxfam/Comadep (Sociedad de Solidaridad Social, Productores Unidos Lol ka x from now on I will refer to them as Lol kax), provided information about honey production and marketing, as well as about other relevant issues related to beekeeping in the area. Information was also derived by assisting the cooperative in their finan cial and organizational tasks. The members of Lol kax are representative of beekeepers from seven different localities (Xmaben, Ukum, Chunek, Pachuitz, Xmejia, Belha, and the ranch of Xnoha) that have been organized since 1998. Other beekeepers in the a rea are members of different cooperatives or market their honey through the informal honey market. Informal interviews were also conducted with external key people, such as the veterinarian working in the area, Victor 1 Data for temperature and precipitation were obtain ed from the Comision Nacional del Agua for the period from 1984 to 2000.

PAGE 77

65 Camara, who has worked in the Yucatan for more than 30 years, and with members of Miel Maya (the company that has bought honey from Lol Kax). Regular assistance to the training sessions carried out with subgroups of beekeepers in the different localities, especially a group of women beekeep ers from the Belha community, provided information about different issues of beekeeping management practices in relation to honey marketing. Results Beekeeping at La Montaa The inhabitants of La Montaa, of Mayan descent, have had a tradition of beekeepin g for many generations. People in the area remember their parents, or themselves when young, keeping large numbers of colonies of xunaan kab These bees are traditionally kept in specially made hollow logs (called in Mayan, hobones ) in thatched structur es within the solar (the Maya homegarden), or near their agricultural plots (the milpa ). Wax and honey from these bees are greatly valued for different religious and medicinal attributes. Meliponiculture (the termed used for beekeeping with bees from th e subfamily Meliponinae) is now seldom practiced. Even though occasionally some still keep a few hobones of these bees, beekeeping activity has changed to be a more commercial one, directed towards regional, national, and international markets, and carrie d out with Africanized strains of Apis mellifera Currently, beekeeping is the number one income generating activity at La Montaa, bringing considerable amounts of capital into the area (see below). 2 Oxfam/Comision Mesoamericana de Asistencia Para el Desarrollo Popular, Asociacin Civil, at that time Oxfam Belgium who began a project in the area in 1993 funded by the European Co mmunity and by the Oxfam office in Belgium, that has extended until recently.

PAGE 78

66 Modern beekeeping was promoted in the area during the e arly 1970s by government initiatives (specifically by the PIDER the Programa de Infraestructura para el Desarrollo Rural) through an effort to integrate the area into the national economy (Camara, personal communication). During that time beekeeping was carried out with the Italian bee ( Apis mellifera ligustica ), one of the European subspecies. The arrival of Africanized strains of Apis mellifera ( Apis mellifera scutellata) into this area occurred during the last half of the 1980s (Crane 1990; Camara, personal communication). With this, many people discontinued beekeeping. Numerous stinging incidents, fatal in some cases, are reported to have occurred. Nevertheless, beekeeping continued, especially among the younger population. During the 1990s, b eekeeping actually gained importance. Two main factors made this possible. First is the construction of the Hopelchen Xpujil road, which was finished in the early 1990s. This road opened access to the area and led to many changes particularly in communi cations and market accessibility, and with these, the involvement of the population in the regional economy. Changes have been astonishingly visible from year to year in many different social, economical and political aspects. For beekeeping in particula r, alternatives to the traditional commercialization scheme have emerged. For example, instead of one or a few intermediaries (the coyote ) with absolute control over markets (buying the honey from individual producers and re selling it to a buyer), many p eople are now organized into beekeeping cooperatives, although the coyote scheme also remains in operation. The road has also meant access to information and awareness of the different limitations and possibilities for commercializing bee products (aided by non governmental organizations that have established in the area). Second, the arrival of the

PAGE 79

67 Africanized bee, with high numbers of feral colonies, has meant greater bee availability. A prospective beekeeper needs only to acquire the basic equipment ( mainly wooden boxes which are produced in the area) to establish a productive unit. Many of the local population (particularly younger people) have taken advantage of the opportunities for producing and commercializing honey. Socioeconomic Importance of B eekeeping Honey production has traditionally been part of a larger natural resources management system in which activities complement each other in space and time. Segregation in space is acomplished by the use of different ecosystems for different produc tive activities, and in time by the distribution of different activities throughout the year according to season (Porter 1995). In this way, people follow the strategy indicated by Toledo (1991) to be typical of indigenous management systems, maximizing t he utilization of all available ecosystems and getting a number of essential products throughout the year. The strategy buffers households and communities against market fluctuations and environmental hazards. People at La Montaa practice a type of slas h and burn agriculture, called the milpa in which corn, beans and squash (among other perennial and annual crops) are planted. This is an important traditional activity carried out by all men in the area. It is a subsistence activity and most food consu med by families, or used for feed for domestic animals, is produced in this way. Other types of agriculture, like agroforestry plots and homegardens, are also common and important for meeting family needs. Agricultural activities are mainly for subsisten ce and play only a small role in bringing direct cash to the families, since only a few of the products are sold.

PAGE 80

68 Cattle raising is an activity that has recently gained in importance. Government programs have aided in its promotion and even though most pe ople have only a few head of cattle, some individuals are investing much of their capital and labor into increasing their herds and pasture lands. A socioeconomic analysis conducted in the community of Xkanha showed that about 68% of the families owned ca ttle, although, of these, 75% had <10 head, 22% had >10 but <16, and only 4% had >40 animals (Porter 1995). Nevertheless, cattle raising is an increasing trend and an activity perceived as important for accumulating capital and gaining social status. Als o, cattle are a source for quick cash and used in emergencies. In recent times, the government has given considerable credit to some individuals for establishing cattle ranches. Currently, cattle is at an investment stage with many people investing their surplus income in preparing pasture lands; there are people who invest in pastures even though in some cases they still do not own cattle. According to Galletti (2000), the last 5 years at La Montaa have shown potrerizacin (pasture growth) and cattle r aising continues to gain importance. Use of timber and non timber resources is important for people in the area to different extents. Timber extraction is not a very important economic activity for these communities in general. Most of the commercial spe cies, mainly cedro ( Cedrela odorata L. ) and mahogany ( Swietenia macrophylla King ), have already been eliminated from the forests by logging. Other native species with commercial potential such as granadillo ( Platymiscum yucatanum Standley ), siricote ( Cord ia dodecandra A.DC ) and chaka ( Bursera simaruba (L.) Sarg. ), among others, are being currently extracted. Although timber extraction is limited by government permits and markets, there is no official organization or structure in which the local populatio ns can benefit. Usually, wood is

PAGE 81

69 sold as standing volume to companies that bring in their own trucks and equipment to extract the timber. Local people often receive money for the labor of felling trees, but not for much else, and most of the time they ar e unaware of the volumes that are being extracted. Usually people obtain very little revenue from this activity, which favors companies or only a few of the local people involved in woodshops for making furniture and beekeeping equipment. The only commun ity that is currently benefiting in a significant economic sense from timber harvesting is Xkanha, where most of the timber extraction is being done in an illegal manner with no management plan. Even in this case, extraction has again favored only a few s pecific families (Porter 1995). Chicle extraction has historically been an important economic activity in the area. Chicle is the gum tapped from the Manilkara zapota tree, which is used for making chewing gum. Historically, this activity has been very i mportant in the area and until recently it was the activity that generated the most income. Chicle has always been an export product. At different points during the last century (specifically during the periods of the World Wars), chicle camps guided the establishment of communities. Nevertheless, most of the income benefited outside companies and contractors and little benefit was felt by the local people (Ramayo Lanz 1998). After the chicle boom ended (shortly after World War II), basically when synth etic gums replaced latex for making chewing gum, chicle trees continued to be tapped with an up and down market demand, but still constituting an important source of income. Now, chicle production is the second income generating activity after honey produ ction. In an interview conducted by Galletti (2000), 43% of total ejidatarios at La Montaa were reported to be practicing chicleros (many chicleros although skilled, stopped conducting the activity because of

PAGE 82

70 age or other reasons, since it is a highly d angerous job), and in some of the communities practically all ejidatarios are chicleros (e.g., Xkanha, and Pachuitz). Annual chicle production in the region ranges from 17 to 26 metric tons of dried gum (Galletti 2000, Porter 1995). As a result of the As ian economic crisis, starting in 1997, chicle has not been bought from the area because the Japanese company that bought most of the product went bankrupt. Chicleros are now waiting for a new contract to continue with this activity. Other economic activit ies contribute in different degrees to the economies of La Montaa families and communities. Some individuals have woodshops or other commercial establishments, or own a car or a chainsaw. People may also work for wages either within or outside the commu nities. Women make embroidery and other crafts, which although limited by market availability and low prices, sometimes represent an important income to the family (mainly because it goes directly to family use). Other activities such as hunting may also be important for increasing the well being of families, mainly because meat is an important protein source. Government subsidies can also contribute greatly to total family income, but this depends on the specific government programs at the moment (Galle tti 2000, Porter 1995). Beekeeping is currently the number one income producing activity in the area. As explained before, it is a relatively recent activity as practiced today with Africanized Apis mellifera and given the current marketing possibilities through better accessibility to the area. In a survey conducted in the region, 76% of the men interviewed reported practicing beekeeping (Galletti 2000). In a study conducted in Xkanha in 1995, over 60% of ejidatarios were involved in beekeeping (Porter 1995). An average of about 200

PAGE 83

71 metric tons of honey are produced annually in the region, depending on weather conditions (Galletti 2000). Prices vary greatly (from 1994 to 2000 it varied from 1 to 12 pesos/km), but considering a common current price in t he region of $0.7 USD/km, yields an annual income of $140,000USD/year. Only about 40% is re invested into the activity, but considering that the minimum wage can be as low as $3.00 USD/day, it is a considerable amount of money. Most of the population has only small productive units, with few colonies and involving a minimum of investment. In interviews carried out with 63 beekeepers of four different communities conducted as part of this research in 1999, 32% of the producers had <10 colonies, 46% had 11 to 20, 17% had 21 to 30 and only 5% had >30 colonies. In a study conducted in Xkanha, 60% of total income made from beekeeping in 1995 was generated by only three producers (Porter 1995). Thus, beekeeping revenues still largely benefit a relatively smal l sector of the population. As practiced today, beekeeping is limited by technical knowledge and the investment capacity (time and money) to increase colony numbers. Nevertheless, current programs established in the area (e.g., Oxfam/Comadep, and EDUCE,A .C. 3 ) aim at increasing the potential in the region, by assisting in the organization of beekeepers cooperatives, establishing markets and providing technical advice. Honey Production and Marketing Honey harvest season at La Montaa runs from mid to late February through late May or early June. This timing varies from year to year depending on the winter rains and their timing. There are several factors that determine production success, besides the

PAGE 84

72 necessity of having the skills to work with Africanize d bees. The length and regularity of precipitation controls the flowering period (Chapter 3). Also, beekeepers need to know how to manage their colonies to prevent absconding and swarming. People report that it was as easy to expand the number of coloni es by catching moving swarms, as it was to lose them all because of lack of assistance or an external factor that led colonies to abandon their nests. Of the 63 beekeepers that I surveyed during 1999, only 14% reported no change in colony numbers from 199 8 to 1999 (Figure 7). Actually one producer reported to have gained 100% of his colonies during 1999 and another one reported to have lost 100%, so that colony numbers can clearly change greatly from year to year. The main cause of colony loss was reported to be absconding during the dearth period. Other factors that affect colony numbers included pests and predators that either lead colonies to abscond or to perish. The most threatening predators were reported to be ants, especially army ants of the Eci toninae subfamily, locally known in Mayan as xulab, and the oso colmenero or hive eater ( Eira barbara senex ) a mammal of the Mustelidae, known by its Mayan name as sanjol These predators are also reported in the literature to disturb beekeeping in other parts of the Peninsula (Echazarreta et al. 1997, Snchez Vazquez and Colli Ucan 1992). The varroa mite ( Varroa jackobsoni ) is a recent pest that can greatly affect colonies. According to Echezarrieta et al. (1997) the varroa mite has become an increasin g problem in the Yucatan Peninsula since 1995. At La Montaa, all beekeepers reported to have problems with varroa to different degrees and needed to treat 3 Educacion, Cultura y Ecologa, A. C., a non governmental organization based in Hopelchen assisting the organization of the beekeepers cooperative named Kabi tah, of which some of the beekeepers of the La Montaa region are members.

PAGE 85

73 colonies either with chemical products (e.g., fluvalinate or flumethrin) or organic ones (e.g., for mic acid). Other diseases are reported to affect the well being of apiaries to lesser degrees. The absconding of colonies can be avoided to a certain extent by applying appropriate management practices (e.g., killing ants before they attack the colony; fe eding; removing or providing frames; Camara, personal communication). The effects of diseases can also be diminished through management. Factors other than colony number, such as the age of the queen and colony size and strength, are also important in de termining productivity. These factors can also be controlled with management (Camara, personal communication). 0 2 4 6 8 10 12 14 Number of producers 10% 11-30% 31-50% >50 no change 10% 11-30% 31-50% >50 % of colonies gained % of colonies lost Figure 7. Number of beekeepers that gained, lost, or had no change in their colony numbers from 1988 to 1999, accordi ng to a survey of 64 beekeepers of the communities of Chunek, Pachuitz, Ukum and Xmaben conducted during 1999. In summary, production is determined by both natural factors and management practices which depend on the beekeepers skills and the resources in vested in the activity (both in time and supplies). The average kg of honey produced per colony was

PAGE 86

74 reported by a group of 64 beekeepers of four different communities to be 60 kg/colony, although some reported to have produced 10kg/colony and some up to 1 50 kg/colony. As mentioned above, honey is commonly sold through a coyote (the intermediary), who buys honey directly from a producer and re sells it to a buyer. The intermediary, thus determines price. The coyote also has the role of providing credit t o producers for maintaining and managing apiaries throughout the year. Credit provided during the off season benefits the coyote, who then sets the prices, terms and conditions. Within this commercializing scheme, there is little room for capitalization or re investment. Most recently, some producers have organized themselves into cooperatives, with assistance of non governmental organizations (NGOs). The purpose is to create a system where members are provided with social and administrative mechanisms i n order to (1) increase their technical knowledge and productivity (through technical assistance facilitated by the non governmental organizations), (2) obtain government credit, (3) negotiate directly with markets, and (4) create mechanisms of capitalizat ion that increase the cooperatives capacity (technical, organizational and administrative) and ultimately create structures that benefit members in the mid and long terms. Cooperatives have also been formed for chicle production, with assistance from th e Plan Piloto Chiclero. The objective of the Plan Piloto Chiclero is to organize chicleros to create a commercialization front, in which producers are empowered to negotiate directly with the buying companies, establishing terms and conditions in their favor. The cooperative also has the role of creating services to benefit members in the mid and long terms, such as employing new technology, and offering loans, life insurance, and retirement funds (Aldrete, personal communication).

PAGE 87

75 The Lol Kax Cooper ative Example The cooperative of Lol kaax was formed at the beginning of 1999 with the assistance of Oxfam/Comadep, which started a program of technical assistance for beekeepers in 1998. The initiative for forming the cooperative came both from Oxfam/Co madep and from a group of young beekeepers who had already participated in the Kabi tah cooperative, but had renounced their membership. Kabi tah is a cooperative whose headquarters are in Hopelchen (formed through the assistance of EDUCE, A.C.) which has been in operation for some years. The Kabi tah cooperative incorporates a much larger region, particularly in the north. Some of the people at La Montaa found that they were in at a disadvantage because their harvest season begins later than in the nor th. At the end of the harvest season, people at La Montaa are affected the most if resource shortages occur at the cooperative (cooperatives function in a way that they need capital to buy honey from individual producers before receiving payment from the buying company). Also, after a Kabi tah experience in which a shipment of honey exported to Germany was found to be contaminated with chemical residues and rejected (something that at that time resulted in a general crisis), beekeepers realized the impor tance of having adequate technical training and a more regulated supervision of quality control. Oxfam/Comadep assisted Lol Kax to legally establish the cooperative, to obtain credit, and to locate markets. It has also carried out a formal technical ass istance program. Nevertheless, Lol kax has been in charge of much of its own management. Currently, Lol Kax has 72 members from 6 communities (Xmaben, Ukum, Pachuitz, Chunek, Xmejia, and Belha) and one settlement (Xnoha a group of people that have set tled in National Territory for over 50 years). During the last two years, they have sold

PAGE 88

76 part of their product to Miel Maya Honing, part of the Fair Trade Market 4 with headquarters in Belgium. The experience of Lol Kax in their export enterprise has no t been an easy one. For example, Lol kax has had to establish mechanisms of self financing for product commercialization. They also have had to cope with strict quality control norms established by the international market. Miel Maya buys honey of the best quality: no more than 18% water content and free of chemical residues. This has resulted in a very strict program of quality control and has necessitated increased technical training for cooperative members. Contamination of honey usually results fr om use of chemicals and antibiotics for pest control, so that producers have had to use alternative pest control measures. Water content in honey can vary greatly, based on natural causes such as nectar type and weather conditions, as well as management p ractices. High water content results in fermentation. Although the international market usually only buys honey with a water content %, % water content is considered to be good quality honey within the national market. In 1999, Lol kaax established a contract for 20 tons of honey with Miel Maya, although the cooperative was able to only gather <12 tons, mainly because Lol kax di d not find mechanisms to finance the commercialization until the end of the season, when the honey had a naturally high water content. The honey that arrived in Belgium had a water content >18 % and failed to meet the terms of the agreement. Nevertheless Miel 4 Fair Trade markets belong to the International Federation of Alternative Trade (IFAT) whose code of practice establishes: To trade with concern for the social, economic and environment well being of marginalized producers in developing countries. This means equitable commercial terms, fair wages and fair prices. Unfair trade structures, mechanisms, practices, and attitudes will be identified and avoided. To cooperate and not compete. To promote fai r trade and social justice in the interest of the producer, and not to maximize profit at the producer's expense (taken from internet web page http://www.goedwerk.nl/engels/en_page2.html ).

PAGE 89

77 Maya gave Lol kaax a second opportunity and established a new contract for another 10 tons of honey for the year 2000. After one of Miel Mayas representatives came to the area to supervise and test honey quality, Lol kaax shipped to Belgium almo st 12 metric tons of honey of excellent quality. Lol kax produced approximately 44.5 metric tons of honey in the year 2000 (Figure 8), of which 26% was honey with % water content (that sold to Miel Maya). The rest was honey of <19% water content (47% of total honey produced), or <20% water content (27% of total honey produced). Honey of 19 and 20% water content was marketed nationally. Quality control in this case was much lower and Lol Kax was able to sell honey of < 20% water content at the same p rice and were also exempted from conducting lab tests for chemical residues. The price of honey sold to Miel Maya (the honey of best quality) was set at $10.50 pesos per kilogram (approximately $1.05 USD), with Miel Maya paying for shipment and containers The price of honey sold to the national company was also $10.50, but Lol Kax had to pay for shipping and containers, which resulted in a net price of $9.00 pesos per kilogram of honey sold (approximately $0.9 USD). The amount of honey produced during 2000 is not representative of the production capacity of Lol Kax. That year was droughty, with only 22.3 mm of precipitation for the months of January to May compared to a 17 year mean of 56 mm (SD 22.1). People complained about the lack of flowering es pecially at some locations.

PAGE 90

78 0 5 10 15 20 25 30 35 40 45 50 <18% <19% <20% Total Maximum water content (%) Metric tons of honey Figure 8. Honey produced by Lol Kax during 2000. Values are given for 18, 19 and 20 percent water content and for total production. Data obtained from Lol kax financial summary of 2000. Total production for Lol Kax in 19 99 was not recorded, since most of the people sold their honey in that year to the coyote. In my survey conducted after the harvest season of 1999, 65 beekeepers reported that they had produced 69 tons in 1999, and 58 tons of honey in 1998. These figures are only estimates and more years of production records are needed before production capacity for Lol Kax can be reliably predicted. An assessment of the advantages of selling honey internationally through alternative markets such as Miel Maya is critica l. Although prices have been better when honey is sold to Miel Maya, this has been achieved only through a large effort by the cooperative and through subsidies from Oxfam/Comadep. Although the Fair Market provides better prices, the cooperative needs to find mechanisms for self financing. Producing the high quality honey demanded by the international market means increased investment, not only by producers who need to use alternative products for pest control, more intensive management, etc., but also from the cooperative, which lacks

PAGE 91

79 mechanisms for self financing and does not have adequate infrastructure. Much of the Lol Kax failure to produce 18% honey in 1999 was the result of poor handling practices due to a lack of infrastructure (containers were not well sealed and were stored in the open). Competing in the highest quality honey market with no infrastructure places the cooperative at a disadvantage and in a very fragile position. The Miel Maya experience resulted in relationships of trust and d istrust among the different parties. The company has not been sure of the cooperatives capacity to produce good quality honey, but has nevertheless given it the opportunity to do so. The cooperative is wary of the companys fairness, because it has not shown sensitivity to the cooperatives current situation and has not paid in advance, even while knowing that there is a lack of self financing capacity and the infrastructure needed to meet the strict quality control terms. There is also some mistrust fr om Lol kax members towards the cooperative itself, given that the Lol Kax enterprise of selling honey to alternative markets has implied eliminating the coyote producer relationship without sufficient alternative financing. Lol kax membership has chang ed from year to year; about 36% of the members resigned in 1999, while 50% of the members were new in 2000. It is commonly agreed that the cooperative Miel Maya business relationship has potential to bring great benefits in the future. However, for this t o happen Lol Kax needs to strengthen its self financing mechanisms, develop a trustworthy relationship with all of its members, and understand and predict production capacity. For Miel Maya, Lol Kax as a supplier is still an experiment, and it will no t consider the cooperative as part of the enterprise until further demonstrations of production capacity are completed. Lol Kax 2000s commercialization was largely a failure, because although it was able to

PAGE 92

80 purchase honey produced from all of its member s (through state credit 5 ) and fulfilled its agreement with Miel Maya successfully, this represented only 25 % of the total production. The rest of the honey was sent to a national market under a previous word agreement. This honey was not paid for as of 2000, and Lol kax has also delayed paying the state government. Thus, Lol Kax is in a very fragile position. However, the cooperatives experience and knowledge about honey commercialization is increasing and perhaps moving it closer to achieving pr oduction and commercialization goals. Prospects for Lol Kaax Regardless of the problems that have come about with establishing markets, Lol kax has begun to create mechanisms for establishing their enterprise by investing with the only capital they have: bees and equipment. With the contribution of one brood chamber of bees per member, the cooperative established several apiaries at the beginning of 2000. The idea was to create self sustaining production systems to support the cooperative. Apiary produ ction for 2000 was very low (4.2 % of total production) because the apiaries were still in the establishment phase. Nevertheless, they were established with the supervision of an apiculture specialist, using advanced techniques for obtaining high producti on. Some of the production practices carried out in the cooperatives apiaries and presented in technical training programs to Lol K'ax members are provided in Table 4. 5 The credit was provided by the Fondo Nacional de Apoyo a la Empresa Social (FONAES), which translates into National Fund for Aiding Social Enterprise.

PAGE 93

81 Table 4. Some production practices taught at the Technical Training Program for Lol K ax members and used in the cooperatives apiaries at La Montaa. H ARVEST MANAGEMENT ME THODS SPECIFICALLY T O CONTROL WATER CONT ENT C ONTROL OF CONTAMINAT ION OF HONEY FROM TH E USE OF CHEMICALS A ND ANTIBIOTICS USED IN CONVENTIONAL PEST AN D PARASITE CONTROL U SE OF ORGANIC ALTERN ATIVES FOR CONTROLLI NG PESTS AND PARASIT ES Q UEEN REARING Q UEEN SYSTEMATIC REPL ACEMENT U SE OF QUEEN EXCLUDER S A DEQUATE SUPERVISION OF BROOD AND HONEY S UPERS U SE OF ORGANIC WAX FO UNDATIONS ( MADE FROM BEESWAX RA THER THAN SYNTHETIC ALTE RNATIVES ). U SE OF ADEQUATE CLOTH ING AND EQUIPMENT The cooperatives apiaries are intended to provide a source of cash for the cooperative itself, as well as to serve as models and learning centers. As explained before, beekeeping skills in the area a re low and technical support for increasing productive capacity in the region is very important. Only a small group of beekeepers are substantially contributing to production. 2000s production is very illustrative of the latter. Only 15% of the membe rs of Lol Kax produced over 44.4% of the total production for the year. The rest, about 85% of total producers, accounted for only 55.6% of total production (Figure 9). Another objective of the cooperatives apiaries is to establish productive units wh ere organic honey can be produced. This will potentially bring added value or market share to production. At the same time, it is intended that in the mid to long term, the apiaries can serve as models for producing other bee products for sale in altern ative markets.

PAGE 94

82 0 5 10 15 20 25 0-0.25 0.251-0.5 0.501-0.75 0.751-1.0 1.01-1.25 1.251-1.5 1.501-1.75 1.751-2.0 >2.0 Metric tons of honey produced per beekeeper in 2000 Number of producers Figure 9. Metric tons of honey produced in 2000 by each of the 74 producers belonging to the Lol Kax cooperative. Data obtained from Lol kax financial summary of 2000. Benefits for Lol kax so far, in addition to the new technology and c ommercialization, have been the acknowledgment from members that beekeeping can provide a viable alternative income for beekeepers and their families, provided there is an investment. Another issue is the inclusion of women in honey production, especially if they want to be part of the Fair Market 6 Generally, Maya culture does not encourage the participation of women in these types of businesses. Currently, the beekeepers of the community of Belha are women, who were only given the opportunity of belong ing to the cooperative after Lol Kax learned about the advantages of having women as members. They now intend to include more women in beekeeping in the region. Lol Kax goals are to be able to strengthen the cooperative and to reinforce its ties with Mi el Maya, as well as with other markets. Ties with domestic markets are also important, since a large part of their production is not export quality honey. Other 6 The IFATs code of conduct also includes a section under Equal Employment Opportunities, which states t o oppose discrimination and ensure equality of employment opportunities for both men and women who suffer from the exploitation of their labor and effects of poverty and racial, cultural or gender bias (taken from internet web page http://www.goedwerk.nl/engels/en_page2.html ).

PAGE 95

83 alternative markets, such as the organic market and the market for other bee products, could expand possibilities for Lol Kax and for the region in general. Conclusions At La Montaa, honey production is currently the number one income generating activity. Although there are traces of traditional beekeeping, as some individuals still keep colo nies of native bees, this is becoming less and less common. A large proportion of the economically active population manage Apis mellifera bees, with the objective of commercializing their products through the regional, national and international markets. The arrival of the Africanized bee (with its high numbers of feral populations), together with better communication and transportation systems, has brought new prospects for beekeeping in the region. At La Montaa, the use of timber and non timber forest products has been the major land use type. Today, timber resources do not bring much profit to the communities, as most of the valuable species (e.g., Cedrela odorata and Swietenia macrophylla ) have already been logged out and the use of other species ha s not developed. Non timber forest products, such as chicle extraction and honey production, represent the current best alternatives for making a profit from forest resources. As agricultural activities gain importance (e.g., ranching and mechanization), it is important to consider their impacts on apiculture, as well as their viability of non timber forest products as economic options for development in the region, if the forest cover is to be conserved. Chicle has been a major export comodity in the reg ion throughout the 20 th century, but current market trends have not been favorable. On the other hand,

PAGE 96

84 beekeeping has recently developed, and much of the honey produced is being sold nationally and internationally. Unfortunately, profits are benefiting o nly a small sector of the population that has better management skills and investment capacity. Beekeeping could provide a better economic benefit to a larger sector of the population. An example of this is provided by Lol kax, a cooperative formed recen tly by a small number of beekeepers from different communities. Lol kax was formed with the assistance of Oxfam/Comadep with the objective of providing technical training to increase productivity and also to facilitate commercialization. Lol kax is beg inning to establish market ties with the international Fair market which provides better prices. Nevertheless, the experience has not been an easy one as Lol kax needs to find mechanisms of self financing and comply with high quality standards at a tim e when skills and resources are still extremely limited. Better technology and skills, as well as the diversification of the activity (e.g., to include the production of organic honey, pollen, royal jelly, propoli, etc.) could provide more benefits to the region. Nevertheless, for this to happen there needs to be external investment such as from NGOs or other agencies, particularly to increase the organizational, administrative and technical skills of beekeepers. In spite of the limitations, there are str ong prospects for beekeeping in the area. The next chapters will deal with the implications of beekeeping at the landscape level and its status as a non timber forest product that favors the conservation of the forest cover. It is important to keep in mi nd that beekeeping as practiced today favors the increase of an exotic bee species that has proven to displace natural pollinators with unknown consequences for floral and faunal composition (Buchmann 1996, Roubik 1996a, Sugden

PAGE 97

85 1996). Nevertheless, beekee ping at La Montaa and other similar areas in the Yucatan Peninsula presents an economic alternative to other land uses that will surely produce a great impact on forest composition, such as ranching.

PAGE 98

86 CHAPTER 5 FLOWERING PHENOLOGY OF MELLIFEROUS SPECIES AND ITS RELATION TO THE BEEKEEPING CYCLE Introduction Throughout the Maya area, honey production has been part of management systems since ancient times. Maya people are known to have carried out intensive manag ement of native bees of the subfamily Meliponinae, especially the stingless bee species Melipona beecheii Bennett (Chemas and Rico Gray 1991). The current landscape is now populated by introduced honey bees ( Apis mellifera L.) that share many of the same resources with native bee species, particularly of the genus Melipona (Roubik 1991, Roubik and Buchmann 1984, Roubik et al. 1986). Today, beekeeping with Apis mellifera represents an important economic activity for the Yucatan Peninsula. Although compris ing only 8% of the Mexican territory, it contributes nearly 40% of national honey production. This figure is more significant in light of Mexico as one of the most important global honey producers and exporters (Echazarreta et al. 1997). The landscape in the Yucatan Peninsula is known to be rich in nectar (melliferous) and pollen (polliniferous) yielding species (Rico Gray et al. 1991). This characteristic favors beekeeping as nectar and pollen represent the major food sources for bees. Floral sources a re distributed in the landscape both in space and time. In space, according to the distribution and abundance of melliferous and polliniferous species, and in time, according to the flowering periods of these species.

PAGE 99

87 Flowering phenology among species v aries in timing, duration and intensity, as well as in between year variability in response to shifts in exogenous cues (Bronstein 1995). From a plant pollinator landscape perspective, flowering phenology can be viewed as the regional flowering sequence of the entire set of acceptable plant species for a particular pollinator. According to Bronstein (1995), the plant pollinator landscape is determined by an interacting set of plant and pollinator attributes, including flowering phenology and the pollina tors search behavior and dietary specificities. For generalist pollinators, such as the honey bee, inter annual variation in flowering patterns of a single species makes little difference, but rather it is the entire set of acceptable plant species that is important. Honey bee colonies adjust their foraging patterns in time and space, tracking the changing mosaic of nectar sources (Sneyd and Camazine 1991). Therefore, foraging decisions are made at a spatial scale other than the one at which phenology v aries within a single species and foragers are likely to switch to a different species instead of searching out phenologically retarded patches or individuals of the same species (Bronstein 1995). Flowers provide bees with the food sources they require in the form of nectar and pollen, although extrafloral nectar, honeydew produced by insects from plant sap, and exposed sap may contribute to some extent in certain settings (Crane 1990). Nectar is produced by plant species as a reward to visitors and theref ore not all flowering species produce nectar and not all nectar produced by flowers may be accessible to honey bees (Bastiaan 1984). According to Crane (1990), only 16% of the worlds flowering plant species contribute to honey bees as a food source and o nly 1.6% are the source of most of the worlds honey. Nevertheless, the latter figure represents as many as 4000 different

PAGE 100

88 flowering species that produce nectar. Nectarless flowers that produce large loads of pollen and that are known to be visited by be es may also be important bee resources (Roubik 1991; Villanueva 1994). Lists of important melliferous/polliniferous species in the Yucatan Peninsula and other tropical areas have been compiled by a number of researchers (Crane 1990, Espina and Ordetx 1983, Lopez Palacios 1986, Roubik 1991, SAGAR 1998, Sarah Workman, unpublished, Souza Novelo 1981, Villanueva 1994). These species may be identified by conducting melissopalynological studies (pollen analysis in honey samples in order to determine the nectar s ources of honey bees) (Roubik 1991, Villanueva 1994), or by using scale hive records, measuring changes in the weight of a colony of bees and relating it to flowering sources (Crane 1990). However, the opinion of experienced beekeepers can serve just as well, if not better, in forming an idea of which nectar sources are of importance to honey bees (Chemas and Rico Gray 1991, Roubik 1991). The contribution of a species to honey production is not only a product of its flowering phenology, but also varies a ccording to nectar quality and load. The energy value of nectar and the nutritional value of pollen vary markedly among plant taxa. Nectar sugar concentrations in flowers range from 10 to 80 percent, with concentrations varying not only according to plant species, but also at the flower level throughout the day (Roubik 1991). The composition of pollen also varies greatly according to the plant source; for example, protein concentrations range from 7 to 30 % (Crane 1990). Relative abundance may also be im portant in determining a species contributions to honey production. Nevertheless, many flower visitors such as honey bees have ranges that include many patches in which they forage very selectively. For example, Roubik (1991)

PAGE 101

89 estimated that honey bees c an exploit an area up to 300 km 2 So total resource availability is not directly correlated to resource availability per unit area (Corlett 1993). At La Montaa region, north of the Calakmul Biosphere Reserve in Campeche, Mexico, beekeeping represent one of the most important economic activities (Porter 1995, Galletti 2000). The objective of this research was to establish what are the most important melliferous and polliniferous species in this area according to beekeepers information and to conduct a co mmunity level flowering phenology study in order to assess resource availability for bees through time. The questions leading this research were: What is the relation between flowering patterns and the beekeeping annual cycle in this forested area of the Maya landscape? How does this relation differ from other more agriculturally oriented areas? And, what are the implications for honey production and marketing and for forest conservation? Knowing the importance of the different species for honey pr oduction will help us understand the role of biodiversity for beekeeping, and the implications of this role for forest conservation. Study Area and Methods La Montaa is an area north of the Calakmul Biosphere Reserve in the state of Campeche, Mexico. It is formed by a group of eight ejidos occupying an area of approximately 184,340 ha and having a population of about 2700 inhabitants, mostly of Yucatec Mayan origin (Galletti 1999). The climate of the area has been determined, according to the Kppen clim ate classification system as modified by Garcia (1973), to be of the sub humid, warm tropical type (Flores and Espejel 1994). The mean annual

PAGE 102

90 temperature is above 22C and the mean annual precipitation is 1,223 mm (1984 to 2000, data obtained from the Com isin Nacional del Agua). The most dominant vegetation type is medium stature subperennifoleus forests, where the mean height for trees is 25 m and where from 25 to 50% of the tree species shed their leaves during the dry season. This type of vegetation is mixed with high stature subperennifoleus forests, low stature forests that are inundated periodically, aquatic vegetation, and secondary vegetation in different successional stages (Ucan et al. 1999). Soils in the area have been formed mostly by the dis solution of calcareous rocks, which over time has resulted in a karst topography (Flores and Espejel 1994). Drainage is subterraneous with no superficial flows of water, except in the rainy season during storms and periods of inundation, when rain can be very intense and water moves in temporary surface flow channels. Generally, lowlands are nourished by filtration of karst during the humid season and become dry to varying degrees depending on recharge through the dry season (Gates 1999). A list of the m ost important melliferous/polliniferous species for the La Montaa region was developed on the basis of information from beekeepers. The field research period lasted from November 1998 through June 2000. The list was constructed over the entire field per iod, during which time I was studying the beekeepers management process through formal and informal interviews and interacting with local beekeepers for two honey production seasons, as well as learning about the vegetation. For most of the species, I co llected a sample and took it for identification to the Alfredo Barrera Marin Herbarium at the Universidad Autnoma de Yucatn. Data on flowering phenology were derived utilizing three techniques: 1) regular monitoring of marked individuals of 32 of the mos t important species for honey bees, 2)

PAGE 103

91 daily observations of flowering during my 18 months of living in the forest, and 3) through informal and formal interviews with beekeepers. For the mark and monitor study, five individuals (the minimum number suggest ed by Frankie et al. (1974) for conducting phenological studies on plant populations) of each of 32 different species were tagged and monitored every 15 to 30 days for 14 months. These species were chosen on the basis of preliminary interviews with 10 bee keepers, although as research progressed, the list of important melliferous and polliniferous species grew. The marked individuals were located along two transects. One transect was 7 km long and ran along a west east line, starting from the major road, north of the Xmaben intersection. This transect was characterized as running through upland forests of different age classes, including recently abandoned milpas or fallow areas, then an older lowland forest, and ending with an open area characterized by savanna vegetation. The second transect was 2 km long and ran through an upland forest of intermediate age, northeast of the community of Xmaben. Every other week, walks along the two transects were made and flowering of selected species was recorded as: 0 = no flowering, 1= presence of flower buds, and 2= flowering. Similar studies were conducted by Alencar (1998), Bullock and Solis Magallanes (1990), Carabias and Guevara (1985) and Hilty (1980) (among others) for describing population or community repr oductive phenological patterns of groups of selected species in different seasonal tropical settings. The 32 monitored species represent only a sub sample of the important melliferous and polliniferous species, and my knowledge increased over time as I int eracted with beekeepers in the different seasons. Therefore, the mark and monitor

PAGE 104

92 data were complemented by recording flowering observations. These were made during regular visits to the different ejidos and supplemented by information provided by beekee pers at the different sites. I conducted monthly visits to 13 different apiaries in 5 different ejidos, made regular drives along the major road (going 40 km north to south) from Xmaben to Belha, and through the new Chanchen Chunek road, a 16 km road that had just been made the previous year by cutting through older forest. Examples of phenological studies conducted in such semi qualitative manner have also been reported by Corlett (1990 and 1993), who stated that making repeated observations on marked tr ees, even though it provides more detailed individual and species level phenologies, limits the number of species that can be monitored. Observations of flowering periods were annotated with information derived from informal talks with local beekeepers. At each visit to an apiary, which were made every two to three weeks over 18 months, I asked beekeepers what plants had flowered, which were flowering at the moment, and which were about to flower. This also provided information that related the flowering phenology to the beekeeping cycle. This ancillary information was judged to be reliable, since local beekeepers require this knowledge and their ultimate failure or success depends on it to a large degree. Beekeepers need to know, for example, when it i s time to introduce new combs, raise supers, get ready for harvest, or provide artificial feed to their bees to prevent their absconding. Thus, knowledge about flowering events, as well as other related ecological and climatological processes, is derive d through personal observation and trial and error, life long education, myths and tradition, and the everyday exchange of information, and becomes part of their overall management system.

PAGE 105

93 With this information, as well as with data collected through my ma rk and monitor study, I made a final flowering chart that was reviewed and corrected in formal interviews made with five key informants at the end of the field study. The flowering chart that I present here was constructed in collaboration with these 5 local beekeepers (although it includes information provided by many others). Their knowledge and input enabled me to extend this assessment to a period greater than the 18 months of my observations and data recording. Results and Discussion Melliferous/ Polliniferous Species and Their Relation to the Beekeeping Cycle The list of the most important melliferous/polliniferous species in La Montaa includes more than 100 species from more than 67 genera and 31 families of trees, shrubs, vines, woody vines, an d annuals found in the different vegetation types of the area (Annex 1). The importance of the different species as nectar or pollen sources depends on their abundance over the landscape, their nectar or pollen quality and load, and on their flowering phe nology in relation to the beekeeping cycle. The flowering chart depicts the flowering periods of the different species, and its relation to the beekeeping cycle (Figures 10 and 11). These include data from both my qualitative and quantitative observatio ns (Annex 2) as well as the input from local beekeepers. Four distinct seasons in the beekeeping management cycle are identified in relation to flowering patterns: 1. The honey harvest season occurs during the dry season, from mid February to early May, when there is a burst of flowering and conditions are good for bee foraging. This is the time when high quality honey is harvested.

PAGE 106

94 2. The high water content honey season occurs from the end of the dry season through the beginning of the rainy season, in May an d June. Flowering peaks during this time, but honey produced may not always be marketable due to high water content. At the same time, weather conditions may constrain bee foraging since, according to local beekeepers, bees do not forage on rainy days an d heavy rains may also wash out the nectar produced in the flowers before bees are able to access it. Beekeepers at this time prepare for the dearth period by making honey reserves either by not harvesting it or by saving it in containers for feeding it b ack to bees later. 3. The dearth season is the period when there is little flowering. This period occurs from August to October when only less than 10 % of the species may be in flower. The little flowering that occurs at this time becomes critical for co lony maintenance, although beekeepers may also need to provide feed to their bees. 4. The recovery period lasts from November to January or the beginning of February, the end of the rainy season and is characterized by having an increase in flowering (mostly by annual shrubs and vines) During this time, bee colonies recover strength and prepare for flower abundance. The extent in which bee colonies recover during this time varies from year to year, not only depending on flower availability, but also on weat her conditions given that bees reduce their activity on cold days (Mendes and Camargo 1991), and on management by beekeepers, who need to supply colonies with resources such as comb foundations. Another factor that affects recovery is the flowering of a tree species, Casimiroa tetrameria (Rutaceae), the nectar from which is poisonous to the exotic bees. A study conducted by Chemas and Rico Gray (1991) in the area of Tixcacaltuyub, Yucatan, reported that the annual honey production cycle in that area star ts in October (after the rainy season), with a good quality honey harvest period from November to March, and a period of high water content honey harvest from mid May to mid September. They also report that during the rainy season only a few species flowe r and bees may be threatened by a shortage of resources, forcing beekeepers to feed them sugar and honey with more frequency than at La Montaa. In the north, the beekeeping cycle is advanced (i.e., the honey harvest season starts and ends earlier). Site differences, such as more dry conditions and particularly the absence of mature forests, explain part of the differences observed in the beekeeping cycle of the northern Yucatan Peninsula.

PAGE 107

95 Even in areas closer to La Montaa, such as the northern Hopelche n area that is more agriculturally oriented, beekeepers report that the honey harvest begins and ends earlier. Flores (1990) reported the flowering periods of Leguminosae (the family with the most number of melliferous species at the Peninsula level) in relation to honey production in the Yucatan Peninsula. He found that, in general terms, shrubs flower at the end of the rainy period, annuals and climbers bloom in the rainy to dry period, and trees, even though different species flower during the differe nt seasons of the year, mainly flower in the dry to rainy transition period. This pattern also seems evident in other families. The higher abundance of shrubs and climbers in the more agricultural northern areas may also help expl ain the earlier honey harvest pattern there, since honey production depends greatly on these species that flower earlier. The reduced abundance of flowering trees during low flowering periods may explain the higher reliance of artificial bee feeding in th ese areas. Honey in the northern, more agriculturally oriented areas, may produce larger quantities of single species honey (particularly from Viguiera dentata, Gymnopodium floribundum and Turbina corymbosa ; Flores et al. 1990), a characteristic apprecia ted by the honey market, as opposed to the honey produced in the La Montaa region that relies more on multiple flowering species of the forest. Production of high water content honey may also be greater at La Montaa because of its reliance on tree speci es whose flowering occurs closer to the rainy season.

PAGE 108

96 Figure 10. Flowering phenology of melliferous and polliniferous species at La Montaa, Campeche, Mxico.

PAGE 109

97 Figure 11. Number of flowering melliferous and polliniferous species and their relation with the beekeeping annual cycle at La Montaa, Campeche, Mxico.

PAGE 110

98 Community Level Flowering Phenology Community level flowering phenology at La Montaa is highly seasonal, with most species flowering in the transition from the dry to rainy seasons, but with many flowe ring through the entire dry period (Figure 12). Although mean annual precipitation in the area is 1223 mm, there are significant variations from year to year that influence phenological patterns. For example, annual rainfall was registered as 648 mm in 1 986 and as 1807 mm in 1999. The rainy season lasts from May/June through August and into September October. Southeastern winds with a strong eastern influence predominate, resulting from the northern shifting of the Intertropical Convergence Zone and the High Pressure Subtropical Zone. Summer rains, although abundant, are characterized as calm, and during the fall rainfall is characterized by strong storms influenced by hurricane events. The dry period extends from November to April or May, during which time the influence of cold winds from the north causes light rains that total 30% of annual precipitation (Flores and Espejel 1994). These rains are locally known as cabauelas Flowering during the dry season at La Montaa (February to April), when 31% of the species may be in flower, contributes the most to annual honey production. Cabauelas at this time are of great importance and flowering or nectar yields of many species are greatly dependent upon them. These rain events are generally very localiz ed, so that flowering episodes for many species vary from site to site. Bullock and Solis Magallanes (1990) also reported the importance of cabauelas for phenology in other tropical deciduous forests in Mxico, where soil water retention capacity is not likely to play a role. As a consequence of yearly variations, honey production in the area may vary greatly from year to year.

PAGE 111

99 Figure 12. Flowering phenology and average monthly precipitation at La Montaa, Campeche, Mxico. A. Flowering phenology of m elliferous and polliniferous species according to data obtained at La Montaa from November 1998 to June 2000. B. Mean monthly precipitation at the meteorological station of Chunchintok (data from 1984 to 1999) During the early rainy season, flowering may contribute to good quality honey production, depending on early rainfall intensity; people report that good quality honey can be harvested as late as June in some years. At the beginning of the rainy period (May) and extending into July, as many as 47% of the species may be in flower. Once wetter conditions in the area are established and during the September and October, only 10% of the species may be in flower. At the beginning of the dry season or the end of the rainy season, flowering increases and g ives way to the bee recovery period. During this time, 12% of the species may be in flower, but many of them yield large quantities of nectar, strengthening bee colonies and contributing to a good honey harvest during the following season. Thus, the gene ral pattern presented in Figure 3 varies from year to year. Furthermore, the contribution of each of the flowering species to honey production

PAGE 112

100 at any given productive unit will also depend on its specific location and site characteristics. The general flow ering pattern of the most important melliferous and polliniferous species at La Montaa resemble the pattern reported by other reproductive phenological studies carried out elsewhere in the dry tropics. Dry tropical lowlands generally peak in flowering ac tivity during the end of the dry season and the early wet season (Bullock and Solis Magallanes 1990, Croat 1975, Daubenmire 1972). However, this does not hold true for all life forms, as was reported by De Steven et al. (1987) for phenological patterns of a palm assemblage in Barro Colorado Island, Panama, where flowering peaked during the mid rainy season. It also is not geographically uniform, as shown by Lieberman (1982), who reported that flowering of most species in a dry tropical forest in Ghana als o occurred in the rainiest periods. Frankie et al. (1974) found that flowering peaked during the dry season but had a less important peak at the onset of the rainy season in a dry forest lowland area in Costa Rica. Although flowering may peak at a certain time in dry tropical lowlands, usually there are other important flowering events that relate either to the dry period or to second and shorter dry wet or wet dry transitions (Borchert 1995, Bullock and Solis Magallanes 1990, Croat 1975, Daubenmire 1972). This pattern also corresponds with observations at La Montaa, where many species flowered either during the transition from the rainy to the dry season, or during the dry season. Generally, phenology in tropical areas correlates with the seasonality of rainfall and in less seasonal rainfall areas, phenological patterns are also less seasonal (Frankie et al. 1974, Hilty 1980, Koptur et al. 1988, Putz 1979,). Nevertheless, phenological patterns

PAGE 113

101 differ widely among species and also among conspecific trees under the same set of climatic conditions (Borchert 1995). These differences are attributed to various factors that may play a role in determining a species phenological pattern, which may be biotic or abiotic and can either be endogenous or exogenous to individuals (Leigh 1990). Natural selection and co evolution with pollinators, dispersers, predators and competitors have been suggested to play a role (Bawa 1974, Janzen 1967, Koptur et al. 1988, Lieberman 1982). Variations in tropical environments suc h as temperature, humidity, rainfall, wind speed and daylight, may also be important in triggering phenological events (Lieberman 1982). Nevertheless, there is strong evidence that the driving force of phenological patterns particularly in dry or seasonal areas of the lowland tropics is water availability, which depends not only on rainfall and its intensity, but also on soil conditions and hydrology (Borchert 1994, Bullock and Solis Magallanes 1990) and on the physiological status and vegetative morpholog y of plant species (Borchert 1994, Reich and Borchert 1984). For any given species, tree water status varies according to factors such as the structure and life span of leaves, time of leaf shedding, wood density and capacity for stem water storage, and d epth and density of root systems (Borchert 1995). According to Borchert (1995), the wide variation of tree phenology found in dry forests within the same climatic regime is caused by variations among these components of the soil plant atmosphere continuum. Differences in flowering patterns of the different populations of the most important melliferous and polliniferous species at La Montaa may be explained by such factors. Flowering patterns of the different species and their contribution to apiculture. O bservations on some species regarding their flowering phenology and their contribution to the beekeeping cycle are given below to exemplify the variation that

PAGE 114

102 exists among the different species. Most of these species have an annual flowering pattern (acco rding to the classification given by Newstrom et al. 1994), with variation among populations where flowering is more closely linked to rainfall events. The four seasons identified for beekeeping are used to order the discussion (honey harvest, high water content honey, dearth period and recovery period). Species that have either subannual or continuous flowering patterns (Newstrom et al. 1994) are described in a separate section as multi flowering species. The descriptions also indicate where (i.e., vege tation type, soil conditions, topography) species are found in most abundance (this latter information was obtained mostly from beekeepers but also from vegetation sampling conducted by the author as described in Chapter 6). Honey Harvest Flowering During the honey harvest season, from mid February to mid or late May, 41 different forest species are considered to be important sources of nectar and/or pollen for beekeepers at La Montaa. They are members of at least 15 families and 29 genera (Annex 1, A). Families that were represented by more than one species are Bignoniaceae (3 species, 2 genera), Bombacaceae (2 species, 1 genera), Boraginaceae (3 species, 2 genera), Euphorbiaceae (5 species, 2 genera), Leguminosae (10 species, 7 genera), Polygonaceae (2 species, 1 genera), Sapindaceae (3 species, 2 genera) and Sapotaceae (2 species, 1 genera). Genera that were represented by more than one species included Tabebuia (Bignoniaceae), Cordia (Boraginaceae), Croton (Euphorbiaceae), Caesalpinia and Haematoxylo n (Leguminosae) and Talisia (Sapindaceae). The species considered to be of most importance for honey production in the Yucatan Peninsula, Viguiera dentata (Compositae), Gymnopodium floribundum (Polygonaceae) and Piscidia piscipula

PAGE 115

103 (Leguminosae) (Chemas an d Rico Gray 1991, Flores 1990, Villanueva 1994), are also among the species considered to be most important at La Montaa. During the honey harvest period, beekeepers usually harvest honey every 15 or 20 days from mid February to mid or late May, dependin g on the timing of the beginning of the rainy season. Furthermore, the honey harvest season is a period of intensive work for beekeepers, because they have to visit their apiaries as often as every eight days. During this time they need to provide provis ions to bee colonies, such as comb foundations, frames and supers, since populations are increasing very rapidly. Beekeepers also need to be aware of swarming and when necessary (usually at a later time) make colony divisions. In addition, beekeepers are often busy capturing feral swarms during this time, in order to increase colony numbers that may have been reduced during the dearth period. The first honey harvest commences in early to mid February and it is honey produced mainly from nectar of the annu al Viguiera dentata ( tajonal), and includes honey from other nectar sources that accumulated during the recovery period. During this time and beginning in December, numerous other annuals and herbaceous plants are also in bloom, many of which are visited by honey bees. Nevertheless, Viguiera dentata is considered the major nectar source at this time, since its massively blooming yellow flowers are abundant in recently disturbed areas, along roads and in abandoned agricultural areas. Usually, the first hon ey harvest is considered by local beekeepers as limpieza honey, generally regarded as low quality. Local beekeepers reported that this honey may have a bitter flavor, as a result of nectar from flowers of the previous season (presumably Verbesina gigant ea ) and sometimes has high water content. Most beekeepers prefer to

PAGE 116

104 discard it before honey has ripened in the combs. This conservative management may also benefit consumers since early honey may contain residues of chemical treatments applied earlier ag ainst varroasis (a sickness produced by a mite) or other parasites Viguiera dentata is a good source of pollen and honey provided there is sufficient humidity. Its flowering period can last from mid January to the end of February, although flowering in t he 2000 season lasted only until the beginning of February because of an unusual drought. Two or three harvests from this nectar source can be obtained, including the limpieza harvest, especially from apiaries found near the most disturbed areas. Never theless, Viguiera dentata is not as good a nectar source as it is in other areas of the Yucatan Peninsula, where agriculture is the major land use type. At La Montaa, activities have been traditionally more linked to forestry and agroforestry and therefo re this species is not very abundant. Viguiera dentata is considered a weed that retards tree regeneration (Turner et al. 2001). Several leguminous species are important nectar and pollen sources at the beginning of the honey harvest season. Caesalpinia gaumeri a nectar yielding tree species found in secondary vegetation and upland forests, flowers massively during January and February. Its conspicuous yellow flowers open after trees have completely shed their leaves regardless of rains, although it may not always yield nectar. Caesalpinia yucatanensis a small tree found in secondary vegetation mainly in lowland areas, produce large yellow and orange flowers during February while bearing leaves. Although this tree produces only small amounts of necta r, it is a major source of pollen and is important at this time because bees are increasing their populations and therefore require large amounts of pollen for feeding their young. Haematoxylon campechianum

PAGE 117

105 and H. brasiletto two trees found in lowland ar eas and savannas that often form large single species stands known as tintales (Flores and Espejel 1994), are important nectar sources in the bajos They produce large amounts of flowers from the end of January to the middle of March, although H. brasilet to the species mainly found in savanna areas, may flower later. Gliricidia sepium is a nectar yielding species found in all types of soils and vegetation ages, and flowers from mid February to April. This species, like C. gaumeri, shows massive white to pink flowers that are very conspicuous among its leafless branches. In areas where it is abundant, it can be an important source of nectar, and beekeepers in the Xnoha area reported having harvested large amounts of honey produced while Gliricidi a was in flower. Nevertheless, its nectar yields are highly rain dependant (Espinosa and Ordetx 1983). Platymiscium yucatanum, a timber species with valuable wood that has been extensively logged in some areas, may sometimes be another good nectar source. It fl owers massively after a rainfall event between mid February and mid March, but lasts only a few days. Although it has large amounts of flowers that produce large quantities of nectar, it may not contribute much to honey production since it is not very abu ndant and its flowering period is very short; in the absence of a rain event it may not flower at all. Flower buds were observed in 1999, but because of drought they seemed to abort before they opened. Another species that has a flowering behavior simila r to P. yucatanum with a short synchronized flowering period after a rainfall event, is the susuyuk tree, a Croton species (Euphorbiaceae) that may flower sometime after the end of February. This species is the first of several Croton species that may co ntribute substantially to honey production, since they yield abundant nectar provided there are sufficient rains.

PAGE 118

106 One of the most important nectar sources in the La Montaa area is the tree Bucida buceras ( pukte) It has an extended flowering period that lasts from mid January to mid May, encompassing the entire harvest period. Pukte is a common tree in the lowland forests where it can grow to be very large, reaching up to 35 m in height and having diameters at breast height (dbh) of up to 1.5 m, someti mes forming single species stands known as puktales (Pennington and Sarukhan 1998). It is reported by the local people that this tree is one of the few species that withstands strong hurricanes. It produces large quantities of nectar yielding flowers for long periods of time and is considered to be a prime nectar yielding tree. Very large individuals Ceiba pentandra ( ceibo ), the sacred tree of the Maya, are found in areas where human settlement is likely to have occurred in the past, such as near water bo dies, or in kakabales or ramonales (vegetation communities characterized by local people as growing in areas with archeological remains). The ceibo tree is also very common in urban areas. This is a bat pollinated tree with high flower production that yi elds abundant nectar and pollen, but which is also known to be visited by birds, small mammals and insects including honey bees (Toledo 1977, Espina and Ordetx 1983). At La Montaa, it flowers during January and February. Other important nectar sources at the beginning of the harvest season are Tabebuia chrysantha and T. roseae These are large trees that grow in secondary vegetation, or quemadales (areas that have experienced wildfires) of lowland areas or near savannas. Their large yellow ( T. chrysanth a ) or whitish pink to dark pink ( T. rosea ) flowers displayed on bare branches, are nectar yielding and have been reported to be adapted for insect pollination (Toledo 1977). Different individuals or stands of

PAGE 119

107 individuals may flower at any time during an e xtended period that lasts from mid February to mid June. Gmez Figueroa and Fournier (1996) report that differences in flowering phenology observed in two populations of T. rosea found in different sites of Costa Rica were produced by differences in tempe rature and soil moisture, since this species requires high temperatures and a dry period to flower. This may explain why trees of T. rosea at La Montaa found in different areas were seen to flower at different times. Crescentia cujete ( giro ), an abundan t species in lowland areas and particularly in savanna communities, is another nectar yielding species that has an extended flowering period lasting as long as from January to May. Few flowers are commonly produced by each individual tree for a long perio d of time, although local people report that flowering greatly increases in intensity after the savanna has burned. Savanna burning is carried out every year according to local people as a management practice, to encourage grass resprouting for cattle and to enhance deer hunting, as well as to increase flower production. This species is characteristically bat pollinated, although it is reported in the literature that other insects, including bees, forage on its abundant nectar loads (Bestmann et al. 1997) as has been observed by local people at La Montaa. Among the nectar sources that flower in March are Talisia olivaeformis and Ehretia tinifolia T. olivaeformis ( guaya) and E. tinifolia ( beek ), are two evergreen species found in the upland forests th at are also widely found in urban settings. These two trees are commonly found in the kakabales or ramonales for which it is alleged that they were cultivated by the Mayas since ancient times. Guayas are appreciated for their

PAGE 120

108 fruits and the beek tree be cause of its shade and because animals forage on its fruits. Beek is also used for medicinal purposes. These trees produce abundant flowers on which many bees can be observed imbibing nectar. The beek tree flowers from March to April. Guaya a dioeciou s species, flowers during March but it was observed by local beekeepers to hold flowers on male trees for a longer period of time. One of the most important species for honey production is Gymnopodium floribundum ( tsitsilche), the species considered to be of most importance for apiculture in the Yucatan state. Honey produced from this species is of very high quality. This small tree is found in abundance in secondary vegetation and to some extent in upland and lowland forests, where it flowers massivel y producing very perfumed nectar yielding greenish flowers. At la Montaa, it may flower from March to June, depending on rains. After a rain event, a burst of flowering occurs that lasts for a period of about one or two weeks. It may have several episo des of flowering in relation to rain events, as was observed in 1999 when the same stand of Gymnopodium floribundum flowered both in April and in June. On the other hand, even where this species is abundant it did not flower until June in 2000, when it st arted to rain. Some of the pollen sources during March are Chrysophyllum mexicanum, Dipholis salicifolia, Pseudobombax ellipticum Chrysophyllum mexicanum, (Sapotaceae), is a forest tree that grows in upland and lowland forests. It flowers in March and a lthough it yields little amounts of nectar, it is a major pollen source. Another pollen yielding member of the Sapotaceae is Dipholis salicifolia a tree that grows in upland forests but that can also be found in secondary vegetation of well drained areas According to local beekeepers, it flowers during March and April, although I observed

PAGE 121

109 flower buds in February. It is not an abundant tree, but is reported to be important for apiculture because of its extended flowering period. Another upland forest t ree, Pseudobombax ellipticum is a bat pollinated species with large flowers that contain several hundred stamens each. This species has been reported to be a pollen source for Africanized honey bees in central lowland Panama and Quintana Roo (Roubik 1991, Villanueva 1994). It is also reported by local beekeepers at La Montaa to be a major pollen source. It flowers during March and April when its conspicuous flowers are visible on bare branches. Metopium brownei is an abundant tree in secondary and uplan d forests. It has an extended flowering period that begins in March and can last until the beginning of May. It produces racimose axillary flowers on leaf bearing branches that produce abundant nectar. This dioecious species is a secure source of nectar at La Montaa. Metopium brownei was observed flowering even in drought periods in which other species had delayed flowering. Flowering of several Croton species is important for honey production. These species are very abundant in secondary vegetation an d yield large amounts of nectar, although their flowering depends on rain events that cause massive synchronous flowering. Yash peres ( Croto sp. ) is found in moderately to well drained sites and flowers sometime between March and early May. This specie s can also be found in more mature vegetation conditions. According to local beekeepers, yash peres trees require only a light rain in order to flower. On the other hand, Sak peres ( Croton glabellus ) flowers after a heavy rain, which occur sometime fro m the end of March to the beginning of June. This species grows mainly in moderately drained areas. Peres de akalche ( C.

PAGE 122

110 guatemalensis ), another abundant species, is only found in secondary vegetation on mainly poorly drained soils. It is reported tha t this species flowers sometime from the end of March to the beginning of May and that it requires a heavy rain to flower. From April until May, the flowering of Piscidia piscipula ( jabin ) constitutes an important nectar source. Actually, many local beeke epers consider it to be the most important nectar source in the region. According to local beekeepers Piscidia piscipula flowers regardless of rains and always yields large quantities of nectar from its massive flowers that are formed on leafless branches It is a leguminous tree found in secondary vegetation and mature forests on all soil types, where it is very abundant. Other forest species that may contribute to honey production are species that flower several times during the year. These will be dis cussed in a different section below as multi flowering species. There are also other species commonly found in agroforestry plots or agricultural areas that yield large amounts of nectar and/or pollen during the honey harvest season. Examples are mango ( Magnifera indica L.), avocado ( Persea americana Miller), many Citrus species, guano ( Sabal yapa C. Wright ex Beccari), cocoyol ( Acrocomia mexicana Karw. Ex Wart) and many others that are found in the diverse traditional Maya homegardens. High Water Content Honey Flowering The beginning of the rainy season is a time when many nectar or pollen yielding species are in flower. At this time, water content in honey is too high to be considered good quality. However, beekeepers may sometimes harvest the honey pr oduced during this time and either sell it to the local market at lower prices, or store it for family use or for feeding bees during critical periods. Also, nectar and pollen produced during this time are important for bee colonies during the dearth peri od. Honey not harvested is

PAGE 123

111 regarded by beekeepers as maintenance honey. Some of these flowers may contribute to good quality honey harvest but only for a short period of time, depending on the intensity of the beginning of the rainy season. From late M ay to July, 39 species are considered to be good sources of nectar and pollen (Annex 1, B). These species are members of at least 16 families and 27 genera. The families that are represented by more than one species are: Anacardiaceae (2 species, 1 gener a), Euphorbiaceae (2 species, 1 genera), Leguminosae (9 species, 7 genera), Malpighiaceae (2 species, 2 genera), Myrtaceae (3 species, 2 genera), Polygonaceae (3 species, 1 genera) and Rubiaceae (4 species, 4 genera). Families represented by one species a re: Ebenaceae, Lauraceae, Meliaceae, Nyctaginaceae, Olacaceae, Rhamnaceae, Sterculiaceae, Ulmifera and Verbenaceae. Enterolobium cyclocarpum ( piich) is one of the species that flowers at the end of the dry season and the beginning of the rainy season and therefore at times may contribute to good quality honey harvest. This is a legume species not very abundant in the natural forests, but whose huge individuals (may grow up to 30 m in height and 3 m in DBH, Pennington and Sarukhan 1998) can be found in upl and forests in sites with rich soils, as well as in ramonales or kakabales A species widely appreciated by people, it is also commonly found in urban areas. The flowering period of the piich tree lasts from May or late April to early June. Each tree pr oduces many nectar yielding flowers. Lysiloma latisiliqua, on the other hand, is a common tree growing in secondary vegetation of well drained areas as well as in upland forests that flowers from May to June. Once this tree begins presenting its abundant high nectar yielding flowers, even at

PAGE 124

112 periods with low intensity rainfall, beekeepers know they will start collecting high water content honey. Important Croton species at this time are C. reflexifolious and C. campechianus C. reflexifolious a small tr ee that grows in abundance in secondary vegetation and is sometimes found in more mature forests, produces large amounts of highly perfumed creamy colored flowers that yield large amounts of nectar. It flowers during June and July. C. campechianus on th e other hand, is not very common and is mainly found in upland sites with deep soils. This species is a source of nectar and pollen for bees during the months of May and June, although they are not produced in large quantities by the flowers. A group of Coccoloba species from the Polygonaceae, an important family for honey production, are nectar sources during this time. C. spicata a tree that can reach large sizes and found both in secondary vegetation and more mature forests in well drained areas and flowers during May and June. C. reflexiflora on the other hand, is a small tree that grows in lowland areas, flowers from the end of May to July. These two species are abundant in the area and their flowers are rich in nectar contributing to high water content honey production. A third Coccoloba species (unidentified) grows in savannas or near the savannas (areas distinguished by local people as specific vegetation communities, named suuts which are described as lowland forests that inundate for long periods of time and border savannas). This is a bush or small tree that produces nectar yielding flowers during July and August. Some of the important pollen sources during this time are Caesalpinia violaceae and Zea mais Caesalpinia violaceae flowers d uring the months of May and June. It is a

PAGE 125

113 tree found mostly in upland forests. It is not very common but its bright yellow flowers produce large loads of pollen. Zea mais is cultivated in the agricultural plots or milpas scattered in different areas of the forest. This wind pollinated species produces high quantities of pollen which is foraged by honey bees generally in August and September. Dearth Period Flowering. After the abundant flowering at the beginning of the rainy season, flowering drops drama tically for a long period of time that lasts from August to November. This is a critical period for beekeepers, since bee populations may decline or colonies may abscond in search of floral sources. The little flowering that occurs is very important at t his time and in its absence beekeepers may need to provide external feed to their bees in the form of diluted honey or sugar, or by providing a pollen substitute. Management during the previous season is also important in determining the extent to which b ee colonies will be affected by this dearth period, since colonies whose last honey production was harvested will be more vulnerable. At this time, beekeepers usually need to visit their apiaries every 15 days. It is also a time when many pests, parasite s and predators may threaten and diminish bee colonies (Echazarreta et al. 1997). A total of nine flowering species were identified as sources of honey or pollen during this extended time (Annex 1, C). The species are members of at least 5 different famil ies. The most represented family is, again, the Leguminosae, contributing three different species. The remaining families are the Araliaceae, Malvaceae, Sapotaceae and Sapindaceae. Although several plants flowering in the previous season may still be in flower in August, there is a noticeable drop (Figure 2). Species that may produce flowers during this time are Colocarpum mammosum Dendropanax arboreus and an unidentified

PAGE 126

114 species commonly known as Inche Calocarpum mammosum a member of the Sapotaceae is a nectar yielding tree found in upland forests, which may only be common in areas with deep soils ( kankabales ). Another nectar yielding tree common in upland forests with good drainage, is the timber species Dendropanax arboreus This species flower s during August, although it did not set flowers during 1999. Another species that flowers during August is Inche a nectar yielding bush or small tree found in the areas bordering the savannas or suuts Hampea trilobata a dioecious species, is consider ed important because of its extended flowering period that lasts from August to September and because it yields large loads of nectar. This tree is common in secondary vegetation and in upland forests. Two leguminous species are also important during the month of September. Lonchocarpus rugosus is a nectar and pollen yielding species found in secondary vegetation and upland forests with a short flowering period in September. Mimosa bahamensis a common bush in secondary vegetation, on the other hand, h as an extended flowering period. It begins flowering during September and lasts until the beginning of December. This bush yields only small quantities of nectar but it produces large loads of pollen. Although this species is considered a good source of pollen for bees during the dearth period, some consider it to affect adversely their health. During the month of October, an unidentified species of tree commonly known as Kaan chan sets nectar yielding flowers. This small tree is found in restricted ar eas of upland forest usually in deep soil areas ( kakabales ). Two important nectar sources at this time are Allophyllus cominia and Bouhinia herrerae Allophyllus cominia a bush member of the Sapindaceae, grows in young secondary vegetation derived from well

PAGE 127

115 drained areas. This is a good source of nectar but it also yields pollen during October and November. Bouhinia herrerae a leguminous climber, grows over secondary vegetation mostly in moderate to poorly drained areas. It is a source of nectar and its flowering period extends from October to early December. Recovery Period Flowering Flowering increases during the transition from the wet to the dry period. Of major importance is the flowering of annual vines and shrubs common in disturbed areas, mo stly from the Compositae and Convolvulaceae families. At this time bee colonies may be weak and their populations greatly diminished. Many beekeepers may have lost many colonies that absconded from their apiaries. Flowering during this time, generally f rom November to mid January (after which many other species start flowering) may contribute to strengthening bee populations, although it depends on weather conditions (e.g., since temperatures may drop during December and January and adversely affect bee colonies), factors such as the flowering of Casimiroa tetrameria that apparently obstructs the digestive tract of the bees (Victor Camara, personal communication), and the attention beekeepers give to their colonies. Eleven species were observed to flowe r during this period, including five families (Annex 1, D). The most important species at this time, as already mentioned, are members of the Compositae and Convolvulaceae. Other families represented are the Rutaceae, Sapotaceae and Tiliaceae. The increa se of flowering in the region begins in the month of November, when Allophyllus cominia and Bouhinia herrerae (two species already mentioned as important in the dearth period) are still in flower. This period is characterized by a noticeable boom of sever al annual shrubs and vines. This includes a Viguiera species very similar to

PAGE 128

116 Viguiera dentata but which grows less abundantly, found also in secondary vegetation and in areas that have undergone a wildfire. This nectar and pollen yielding annual usually flowers for an extended period from November to December. Verbesina gigantean (Compositae), is an annual shrub that produces abundant flowers in its large individuals (can grow up to 5m in height). This shrub grows abundantly in secondary vegetation and is commonly visible along roads and other disturbed areas. Its flowering period extends from December to January and it yields abundant nectar and pollen. Several people in the area consider that the nectar of its flower imparts a bitter flavor to honey Other less abundant herbs, shrubs and vines of the Compositae are seen to start flowering at this time and probably contribute as nectar and pollen sources to bees. Several vine species contribute substantially during this period. Turbina corymbosa lo cally known as Xtabentun is an annual vine very common throughout the Peninsula. It is considered to be one of the most important species for apiculture and sometimes provides a harvest of honey in some areas (Flores 1990). The abundant white flowers of this vine, growing in secondary vegetation and in recently disturbed areas, yield large loads of high quality nectar. At La Montaa, its flowering period lasts from December to January, although it was noted that in the Belha area (the southern most part of the research area) flowering of Xtabentun also occurred during March in the year 2000. According to the people in Belha, Xtabentun may flower two times, but for people of the northern part of the region that was observed as a very strange event. It is important to note that weather conditions may vary much from north to south and in 2000 the Belha area experienced a rain event that people in the north did not.

PAGE 129

117 Another annual vine species is Jacqemontia pentantha locally known as Solen ak. This vine has blue flowers that are present for an extended period of time from December to mid February and yield large quantities of nectar. Solen ak grows in recently disturbed areas and in secondary vegetation. People in the area reported that there is anothe r Solen ak species that has white flowers and also is important as a nectar source. Some of the trees that flower during this time that are reported to provide nectar and pollen to bees are Manilkara zapota and Luehea speciosa Manilkara sapota a tree w hose individuals are sparsely represented in the forests, sometimes attains very large sizes since it is a protected species for the extraction of chicle. It flowers during November and is reported to provide nectar to bees. Luehea speciosa is a common t ree in disturbed areas that may also be found in more mature forests and has large white flowers from November to December. Although local beekeepers were not sure of its importance as a source for bees, beekeepers observed during 2000 that bees were obta ining nectar and pollen from its flowers. Another tree that flowers during the recovery period is the already mentioned Casimiroa tetrameria a tree that is reported to be detrimental to bee colonies. This is readily observed by beekeepers who report th at while in flower, numerous bees are found dead near the apiaries. The flowering of this tree, which occurs from November to December and sometimes January (as observed in 1999), can be a major threat to beekeeping in the area. It is reported that the s pecies may flower more than one time. Local beekeepers treat their bees with salt during this time as a preventive mechanism,

PAGE 130

118 since it seems to lessen the effect that this nectar has on the bees health. Nevertheless they constantly report serious losses resulting from the flowering of this tree. Multiple Flowering Species There are a few other species that contribute to beekeeping in different times but that are considered separately because they flower several times during the year (Annex 1, E). Lonch ocarpus xuul is a Leguminosae tree with a biannual flowering mode. This tree is found in secondary vegetation and in more mature vegetation of poorly drained areas and has some of the largest individual trees in the bajo areas. It is reported by local be ekeepers to flower at two times, in March April, contributing to honey harvest, and in November, as a source of nectar for the recovery period. Another species with a biannual flowering mode is Brosimum alicastrum (Moraceae). This tree is considered to b e one of the most important species in the region for the multiple services it provides, since it is used for feed for horses and cattle and its fruits are edible. It is found in abundance in the areas known as kakabales or ramonales (actually the word ra monales is derived from the common name of B. alicastrum Ramon which is very abundant at these sites and can sometimes be present in large single species stands). This is another tree presumed to have been widely cultivated by people in the past and whi ch is still being cultivated; it is very common in urban areas and agroforestry plots. Although B. alicastrum does not produce nectar, it is a rich source of pollen, as reported by local beekeepers and by Espina and Ordetx (1983). Flowers are produced du ring August and for a second time in November. Species that flower multiple times during the year, sometimes more than three, which could characterize them as continuous flowering species include Thouinia paucidentata and Cissus sicyoides Thouinia paucid entata (Sapindaceae) is a species that

PAGE 131

119 flowers at least three times a year, provided there are strong rainfall events that induce it. It was observed that flower buds were ready most of the time and in the event of a rain, individuals massively flowered w ith abundant nectar and pollen. People reported that this tree may flower up to five times in one year depending on weather conditions. It is a tree found in upland forests and can be locally abundant. It is considered to be a very important species for apiculture in the area since it can contribute both to honey harvest and to maintenance during periods of scarce flowering. Cissus sicyoides a perennial climber of the Vitaceae family flowers at multiple times during the year also. It is abundant in th e area and produces nectar yielding flowers. Conclusions The plant pollinator landscape of La Montaa, where honey production has been an important economic activity from ancient times, can be characterized as highly seasonal. Most plant species important as nectar or pollen sources flower at the end of the dry period or beginning of the rainy season, although many species flower for the entire duration of the dry season. Little flowering occurs during most of the rainy season, and at the transition to th e dry period there is an increase in flowering. For the beekeeping cycle, the dry period represents the good quality honey harvest season, followed by a period (the beginning of the rainy season) when honey has a high water content. Most of the rainy sea son is a dearth period for bees, followed by a recovery period during the transition to the dry season. The list of the most important melliferous/polliniferous species includes more than 100 species from at least 67 genera and 31 families of trees, shrubs vines, woody vines and annuals. These species are present in the different vegetation types found in

PAGE 132

120 the region, including lowland and upland forests, fallow areas and recently disturbed sites (Chapter 6). Several important species are also found in sa vanna communities and in urban and traditional agricultural areas (better described as agroforestry systems). Most of the species have an annual flowering pattern, although some exhibit biannual or continuous flowering. Different species contribute to ho ney production to different extents, but they all may play a role over the annual beekeeping cycle. The soil plant atmosphere continuum that determines tree water status in the different species determines flowering phenology to a large extent. Cabauelas the winter rains that sporadically occur in the dry period, seem to be of great importance for flowering events during the honey harvest season. Therefore honey production is subject to high variation from year to year and from site to site. Timing of the onset of the rainy season also affects the extent to which high water content honey is produced each year. The beekeeping cycle in this forested area of the Maya landscape contrasts to more northern areas where modern agriculture has shaped the landsca pe to a larger degree. At La Montaa, the great diversity of forest species contributes substantially to honey production. The honey harvest season occurs later than in agricultural areas, since it is most dependant on species that flower later in the dr y season and at the beginning of the rainy season (i.e., forest trees). On the other hand, in more agriculturally oriented areas honey is produced mostly from the flowering of shrubs and creepers characteristic of recently disturbed areas. These species flower earlier during the beginning of the dry season. Much of the harvest season in the more agricultural areas is the recovery period in La Montaa.

PAGE 133

121 These characteristics have implications for regional honey production and marketing. Honey from foreste d areas like La Montaa is produced from multiple forest species, as opposed to larger quantities of honey that are produced from single species in agricultural settings, which is a characteristic generally appreciated by consumers. Honey produced from fo rests may also be darker than honey produced from single species, another attribute not favored by the current market that prefers light colored honey. Also, more honey produced at La Montaa has a high water content, since it is dependant mainly on speci es that flower later in the dry season and at the beginning of the rainy season. Nevertheless, there are great benefits to be realized from honey production in forested areas. For example, honey from these areas is produced from plants that have not been treated with pesticides or herbicides, as they often are in agricultural areas. Also, beekeepers at these sites depend less on external sugar feeding, since the diversity of plant species provides a wider array of sources where bees may obtain nectar and pollen. More research is needed on the relative nutritional qualities of honey produced from multiple flowering plants in forests. Honey from La Montaa and similar areas can also be regarded as a non timber forest product obtained from trees that are u sed for other economic activities and that present other complementary benefits. Beekeeping per se does not necessarily ensure a focus on forest conservation, since many important species are found in secondary vegetation, and agricultural areas may be as productive. But there is no doubt that beekeepers benefit from the multiple flowering afforded by forests. External factors, such as the development of alternative markets and increased consumer demands for honey produced in these areas, could contribute significantly to the role of beekeeping in forest conservation. Through market

PAGE 134

122 development, beekeepers could obtain added product value, as well as better conditions for business negotiations, for cooperative organizations, and for obtaining new investme nts. These factors could induce beekeepers to practice forest conservation as part of their management system. Beekeepers are already starting to form cooperatives to commercialize honey through alternative markets. Also, some certified organic honey is being produced through the support of NGOs, which in some cases has led to the establishment of forest reserves as bee foraging zones (Acopa and Boege 1998). Honey production at La Montaa and similar areas represents a promising activity that can favo r both the conservation of natural resources and socio economic development. Nevertheless, for this to occur, there needs to be a directed effort through the implementation of specific management practices. Alternative markets may play a crucial incentiv e role.

PAGE 135

123 CHAPTER 6 BEEKEEPING AND THE LANDSCAPE Introduction The resource landscape in which pollinators live is characterized by a combination of plant and pollinator attributes, such as the timing and synchrony of flowering, the dietary flexibility of pollinators and thei r ability to fly among patches. The plant pollinator landscape therefore, is a function of the response of pollinators to spatial and temporal variation in floral resources (Bronstein 1995). In that way, pollinators, such as the honey bee ( Apis mellife ra L.), adjust their foraging patterns in time and space, tracking the changing mosaic of nectar sources (Sneyd and Camazine 1991). Apis mellifera, both temperate and tropical, can forage in an area up to 300 km 2 However, foraging ranges and flight dir ection are constrained by the distribution of forage patches. Other factors such as competition with other pollinators and density of honey bee colonies also restrain the foraging behavior of honey bees (Roubik 1989). Generalist species such as the honeyb ee have broad diets and forage relatively indiscriminately. As introduced species in many parts of the world, they have become adapted to forage in native floras (Bronstein 1995). At La Montaa, an area near the Calakmul Biosphere Reserve in Campeche, Me xico, where this research took place, beekeepers consider that more than 100 species from at least 67 genera and 31 families of trees, shrubs, vines, woody vines and annuals, are sources of nectar and pollen for honey

PAGE 136

124 bees. These species are present in th e different vegetation types found in the region, including lowland and upland forests, fallow areas and recently disturbed sites. Several important species are also found in savanna communities and in urban and agricultural settings (Chapter V). The obje ctive of this research is to understand the distribution of melliferous species throughout the landscape of La Montaa, and particularly understand the role that forests have for apiculture. Honey production in this area is an important economic activity (Galletti 2000, Porter 1995). Being part of the buffer zone of the Calakmul Biosphere Reserve, beekeeping has been recognized as having great potential for socio economic development while preserving the natural forests (Acopa and Boege 1998, Galletti 200 0). An understanding of the role of forests for honey production will aid in understanding how beekeeping could favor forest conservation in this area. The questions guiding this research were: How are melliferous and polliniferous plant species distri buted in the landscape? How does the structure of the landscape favor or hinder apiculture, and particularly, what is the role of forested areas? How do beekeepers manage vegetation for honey production? And, what are the implications of apiculture with regard to vegetation dynamics? Or in other words, how is beekeeping likely to affect the future composition of forests in the area? A vegetation survey was conducted to understand the distribution and abundance of the important melliferous and pollini ferous woody species in the forest types of the area. Interviews with local beekeepers were conducted to understand management of vegetation for honey production.

PAGE 137

125 Methods Study site La Montaa is an area north of the Calakmul Biosphere Reserve in the stat e of Campeche, Mexico. It is formed by a group of eight ejidos occupying an area of approximately 184,340 ha and having a population of about 2700 inhabitants, mostly of Yucatec Mayan origin (Galletti, 1999). The climate of the area has been determined, according to the Kppen climate classification system as modified by Garcia (1973), to be of the sub humid, warm tropical type (Flores and Espejel 1994). The mean annual temperature is above 22C and the mean annual precipitation is 1,223 mm 1 The most d ominant vegetation type is medium stature subperennifoleus forests, where the mean height for trees is 25 m and where from 25 to 50% of the tree species shed their leaves during the dry season. This type of vegetation is mixed with high stature subperenni foleus forests, low stature forests that are inundated periodically, aquatic vegetation, and secondary vegetation in different successional stages (Ucan et al. 1999). Soils in the area have been formed mostly by the dissolution of calcareous rocks, which o ver time has resulted in a Karst topography (Flores and Espejel 1994). Drainage is subterranean with no superficial flows of water, except in the rainy season during storms and periods of inundation, when rain can be very intense and water moves in tempor ary surface flow channels. Generally, lowlands are nourished by filtration of Karst during the humid season and become dry to varying degrees depending on recharge through out the dry season (Gates 1999). 1 Data for temperature and precipitation from 1984 to 2000 were obtained from the Comision Nacional del Agua for the station of Chunchintok, the closest station to the study area.

PAGE 138

126 Distribution and abundance of melliferous species A stratified random sample to assess the distribution and abundance of melliferous and polliniferous species was conducted according to forest type and to the age of the vegetation (years lapsed since recovering from a disturbance). Forest type was based on the local vegetation/soil classification. Kankab and Tzequel soil types were grouped into the upland forest type, and akalche and yash home soil types were grouped into the lowland forest type. Perez Salicrup and Foster (2000) have also identified t wo major natural forest types in the Calakmul Biosphere Reserve area: lowland or bajo forests in seasonal wetlands possessing deep clay soils, and upland forests on well drained terrain. Vegetation age was divided into three classes: young (less than 19 y ears since abandonment), intermediate (from 20 to 29 years since abandonment) and older (30 and above years since abandonment). The criteria used to differentiate age classes was based on information for forests of the area that indicate that forests reco ver a tree species community indistinguishable from extant mature forest in about 25 to 30 years (Turner et al. 2001) and by information from local beekeepers that consider a 20 year fallow to be ready for slash and burn cultivation. Forty five transects 6 m wide x 30 m long (180 m 2 each) were surveyed. At each transect, all trees with a diameter at breast height greater than 3 cm were recorded and their DBH measured. Botanical voucher specimens for identification of the melliferous and polliniferous spec ies were collected and taken to the Alfredo Marin Herbarium in Mrida, Yucatn. For assessing the importance of the different species in each of the different forest/ vegetation age types (e.g., lowland young, intermediate, and older, and upland young, int ermediate and older) the Importance Value Index (IVI sum of relative density, relative dominance, and relative frequency) was calculated for each species and for all the

PAGE 139

127 trees at each transect. The IVI places the species in a hierarchical order in the co mmunity and the most important species are identified. A species has a high IVI when it has a high density, high basal area and high frequency, or when one or two of these parameters is much higher than those for the other species at a site (Felfili and D a Silva 1993). A Melliferous Importance Value Index (MIVI) for each species and all trees at the different sampled transects was also calculated. The MIVI was obtained by multiplying a melliferous value (MV) to all calculated IVIs. The MV for each specie s was obtained from a flowering phenology study that provided information regarding the flowering periods for each of the important melliferous/polliniferous species (Chapter 5). The flowering periods that make up the beekeeping productive cycle are class ified into four classes: 1) harvest season; 2) high water content honey period; 3) dearth period, and; 4) recovery period. Each class has a different importance in relation to the beekeeping annual cycle according to the amount of species flowering at eac h season or period. Since most species flower during the high water content honey period, each single species flowering during this time contributes only slightly to the beekeeping annual cycle, therefore an arbitrary value of 1 was assigned. Species flo wering during the honey harvest period are the second in number and their contribution to the beekeeping annual cycle is greater, therefore a value of 2 was assigned. During the dearth and the recovery periods not many species flower, therefore their cont ribution is even greater and a value of 3 was used. The MIVI calculation was performed in order to identify the most important species contributing as melliferous and/or polliniferous sources at each forest/vegetation age types.

PAGE 140

128 Dominance (sum of basal ar ea of all species in a transect), melliferous dominance (MD) (sum of basal area of melliferous and polliniferous species in a transect), and the proportion of melliferous dominance (MP, the percentage of dominance occupied by melliferous species in a trans ect) were used to assess differences between forest/vegetation age types regarding melliferous importance. Also, the proportion of the dominance accounted for species flowering at the different period of the beekeeping annual cycle (high water content hon ey period, honey harvest period, and dearth and recovery periods) was calculated to assess differences between forest/vegetation age types. Finally, the dominance of pollen yielding species vs. dominance of nectar yielding species was tested to further as sess differences among forest/vegetation age types regarding their contribution as melliferous or polliniferous yielding vegetation types. Dominance instead of the IVI was used for comparing forest/vegetation age types since the former is an absolute numb er rather than a relative one and therefore can be used for this purpose. Dominance was also chosen after assessing its high correlation with IVI (r= 0.88). Analysis of variances (F test) and multiple comparison tests (Tukey Kramer) were used to assess s tatistical significant differences among forest/vegetation age types. Beehive location Using a hand held Garmin 12 Global Positioning System (GPS), the coordinates of beehives of a sample of beekeepers from La Montaa were obtained and located in a land cover/use classification map made with remotely sensed digital image data from a portion of the Landsat 7 ETM 2000 scene (Chapter 2). Interviews of 32 beekeepers were conducted to understand beehive location criteria as well as information regarding the i mportance of different vegetation types to apiculture as perceived by beekeepers. The interviews were conducted either with individuals or with groups of individuals during a

PAGE 141

129 one and a half year period that lasted from January 1999 to June 2000. The inte rviews were made during visits to the different apiaries. Results from this study were complemented with results obtained from a study carried out in the area during the same time about the important melliferous and polliniferous species and from a flower ing phenology study (Chapter 5). Results Distribution and abundance of melliferous and polliniferous species A total of 135 species were found to occur in the different transects sampled for all forest/vegetation age classes. The number of species found in the lowland forest type was 115 and in the upland forest type 112. Ninety of the species were present in both lowland and upland forests. The total number of important melliferous and/or polliniferous species found in all forest/vegetation age classes was 80. Approximately 59% of the woody species found in the different forest/vegetation age types was found to have a melliferous and/or polliniferous importance. Of these, 69 species were present in lowland forests, and 64 in upland forests. The numbe r of melliferous and/or polliniferous species shared by both forests types was 53. Regarding the structure of the different forest/vegetation age types and its relation to apiculture, the analysis of data showed that there were no significant differences b etween the upland and lowland forest types with respect to the dominance of melliferous species and dominance in general (Table 5). Nevertheless, there were significant differences among vegetation ages. For both forest types, the intermediate vegetation age tended to be lower both in dominance of melliferous species and in dominance in general. As for the proportion of dominance of melliferous species with regards to total

PAGE 142

130 dominance, the analysis of results showed that there was also no significant diff erence between forest types but that there was a significant difference regarding vegetation ages, with the older vegetation ages tending to have a lower proportion of melliferous dominance. Table 5. Average of dominance (D ), dominance of melliferous spe cies (MD ** ) and proportion of MD (MP *** ) for transects sampled for the different lowland young, intermediate and older, and upland young, intermediate and older forest types at La Montaa, Campeche. Values significantly different for each row, according t o an analysis of variance and a multiple comparison test with a 95% confidence interval, are marked with an lowland Upland Average values Young Intermediate Older Young Intermediate Older D (m2 ha) 56.28 33.26 61.64 40.19 25.54 50.50 MD (m2 ha) 54.01 28.10 38.06 37.47 25.17 46.55 MP (%) 95.70 85.95 62.19 93.81 98.71 46.55 *D= Do minance (Sum of basal area for all species) **MD= Sum of basal area of melliferous and polliniferous species ***MP= Percentage of dominance of melliferous and polliniferous species or the proportion of MD with respect to D. Regarding the importance of the different forest/vegetation age types for the different periods that make up the beekeeping annual cycle, results show that there are differences between both forest types and vegetation ages (Figure 13, Table 6). For the proportion of melliferous specie s that flower during the high water content honey period, the older age of the upland forest type was significantly different than the rest, tending to hold lower proportions of dominance of species flowering during that period. For the honey harvest peri od, the older vegetation class for both forest types was significantly different from the rest, tending to have lower dominance of melliferous species flowering at that period. Lowland forests showed more variation than upland forests. For the

PAGE 143

131 dearth and recovery periods no significant difference was found for the proportion of the dominance of melliferous species that flower at that time for any forest or vegetation age types. Figure 13. Percentage of melliferous dominance partitioned according to the important periods identified for the beekeeping annual production cycle: High water content honey period, Honey Harvest Period, and Dearth and Recovery periods, at La Montaa, Campeche. Regarding the contribution of pollen as opposed to nectar yielding sp ecies (Table 7) results showed that there are differences between forest/vegetation age types. The oldest forest type of the upland forest is significantly different from the rest, tending to have higher values of dominance corresponding to species that y ield only pollen.

PAGE 144

132 Table 6. Average proportion of species that flower during the high water content honey period (MP1), of species that flower during the honey harvest period (MP2), and of species that flower during the dearth and recovery periods (MP3) f or transects sampled for the different lowland young, intermediate and older, and upland young, intermediate and older forest types at La Montaa, Campeche. Values significantly different for each row, according to an analysis of variance and a multiple c omparison test with a 95% confidence interval, are marked with an lowland Upland Average values Young Intermediate Older Young Intermediate Older MP1 (%) 26.96 33.68 27.62 33.35 38.71 8.30 MP2 (%) 64.82 41.45 28.45 48.27 54.33 31.61 MP3 (%) 3.92 10.82 6.12 12.19 5.66 6.64 MP* 95.70 85.95 62 .19 93.81 98.71 46.55 *MP = MP1+MP2+MP3 = the proportion of dominance accounted for melliferous and polliniferous species with respect to total dominance. Table 7. Dominance of pollen yielding species (PD), and dominance of nectar yielding species (ND) f or transects sampled for the different lowland young, intermediate and older, and upland young, intermediate and older forest types at La Montaa, Hopelchen, Campeche. Values significantly different for each row, according to an analysis of variance and a multiple comparison test with a 95% confidence interval, are marked with an or lowland Upland Average values Young Intermediate Older Young Intermediate Older PD (m 2 ha) 2.34 3.15 3.00 2.57 0.95 9.26 ND (m 2 ha) 51.67 24.86 35.06 34.90 24.16 37.30 MD* (m 2 ha) 54.01 28.10 38.06 37.47 25.17 46.55 *MD = PD + ND Differences regarding dominance of nectar yielding species are provided by the intermediate vegetation age types that tend to have lower values than young vegetation a ge types, although not significantly different from older types. The pattern of the nectar yielding species resembles the tendency of the melliferous dominance in general as most of the species are nectar yielding.

PAGE 145

133 For ranking the importance of the differ ent species both with regard to the vegetation structure and with regard to their importance as melliferous species, the average of all IVI and MIVI values for each forest/vegetation age type was used (Tables 8 and 9). Generally, species that ranked highe st regarding IVI are important melliferous species as well. Both forest types shared some of the most important species, although their importance regarding IVI or MIVI varied in some cases. Species shared by both forest types that are present among the first 10 ranking species according to the sum of MIVI values for all vegetation ages within each forest type (sum of MIVI) are a group of Croton species Bursera simaruba Piscidia piscipula Gymnopodium floribundum and Hampea trilobata Of these, the fi rst four are among the most important species during the harvest season period, and the latter one is an important species during the dearth period. For the young vegetation age in the lowland forest type the most important species both regarding IVI and MIVI is the Croton spp group, the second is Bursera simaruba, also ranking second in both values. The third MIVI ranking species is Caesalpinia gaumeri another important species during the harvest season, although Coccoloba reflexiflora ranks third regar ding IVI value. This latter species, although melliferous, has a limited value for MIVI since it flowers during the high water content season. The 4, 5 and 6 MIVI ranking species for this forest/vegetation age type are Gymnopodium floribundum Piscidia p iscipula and Metopium brownei having an IVI ranking of 5, 6, and 8, respectively. Fifty percent of all MIVI values are constituted in the mentioned top 6 MIVI ranking species.

PAGE 146

134 For the young vegetation age in the upland forest type eight species account for the first 50% of MIVI values. The first three species are Piscidia piscipula Bursera simaruba and Croton spp all im portant species for the honey harvest period. Their IVI ranking is 2, 3 and 4. Hampea trilobata is the fourth species in importance regarding MIVI and although it ranked 9 for IVI, it has an increased importance for MIVI because it is a dearth period flowering species. Gymnopodium floribundum an important honey harvest period species has a MIVI rank of 5 and an IVI rank of 6. Cocco loba spicata a species that flowers during the high water content honey period, although ranked first regarding the IVI has a MIVI rank of 6. Vitex gaumeri and Metopium brownei were placed 8 and 9 both for MIVI and IVI. For the intermediate lowland forest type nine species account for the first 50% MIVI values. Croton spp and Gymnopodium floribundum ranked 1 and 2 for MIVI, and had a 3 and 4 rank for IVI, respectively. Coccoloba reflexiflora and Eugenia mayana ranked 3 and 4 regarding MIVI. These are s pecies that flower during the high water content honey period but that have important IVI ranking (1 and 2 respectively), enhancing their melliferous importance. Bursera simaruba and Metopium brownei are ranked 5 and 6, with an IVI rank of 6 and 7. Kakaw che (unidentified), and Luehea speciosa although having low IVI ranks of 13 and 12 respectively, have a MIVI rank of 7 and 8 given their importance of flowering during the recovery period. Caesalpinia gaumeri a species flowering during the honey harvest season is the 9 th species that makes up the first 50% of MIVI values and has an IVI rank of 5. For the intermediate upland forest type only four species account for the first 50% of MIVI values. Croton spp. ranked first both for MIVI and IVI. Bursera sim aruba

PAGE 147

135 Table 8. Melliferous importance value index and its ranking (MIVI and MIVI Rank), importance value index and its ranking (IVI and IVI Rank), and the sum of MIVI values for lowland species of young, intermediate and older vegetation ages. Species ar e ordered according to the Sum of MIVI values. Red numbers are the species that account for the top 50% values. SPECIES LOWLAND/ YOUNG LOWLAND/ INTERMEDIATE LOWLAND/ OLDER MIVI (%) MIVI Rank IVI IVI Rank MIVI MIVI Rank IVI IVI Rank MIVI MIVI Rank IVI IVI Rank SUM MIVI Croton spp. 17.4 1 13.8 1 8.3 1 5.8 3 7.6 4 4.1 5 33.2 Coccoloba reflexiflora 4.0 7 6.3 3 5.9 3 9.6 1 13.4 1 14.6 1 23.3 Gymnopodium floribundum 7.5 4 6.0 5 6.3 2 4.7 4 8.3 3 4.5 3 22.1 Bursera simaruba 8.3 2 6.6 2 4.9 5 3.4 6 3.2 7 1.7 13 16.4 Manilkara zapota 3.1 10 1.6 18 1.7 20 0.9 33 11.2 2 4.1 6 15.9 Metopium brownei 4.6 6 3.7 8 4.6 6 3.3 7 6.1 5 3.3 7 15.3 Caesalpinia gaumeri 7.8 3 6.2 4 4.0 9 4.2 5 2.5 10 1.4 21 14.3 Piscidia piscipula 6.9 5 5.5 6 3.8 10 2.6 8 2.0 14 1. 1 32 12.7 Hampea trilobata 3.2 9 1.7 17 2.4 16 1.1 30 3.1 8 1.1 31 8.6 Eugenia mayana 1.2 19 2.0 13 5.1 4 7.0 2 1.9 16 2.1 10 8.3 Vitex gaumeri 2.2 14 1.7 16 2.7 14 1.9 15 2.4 11 1.3 22 7.2 Lonchocarpus rugosus 3.6 8 1.9 14 2.4 15 1.2 26 1.2 23 0.4 5 1 7.2 Croton sp. (susyuk) 2.7 11 2.1 10 3.2 12 2.2 9 0.0 0 0.0 0 5.8 Unidentified (Tojyu) 2.5 12 2.0 11 1.1 25 0.8 39 1.7 17 0.9 33 5.4 Luehea speciosa 0.0 0 0.0 0 4.4 8 2.0 12 0.8 25 0.3 70 5.2 Chrysophyllum mexicanum 0.6 23 0.5 44 1.0 28 0.7 42 3.4 6 1.9 11 5.0 Thouinia paucidentata 1.5 17 0.8 36 3.1 13 1.4 23 0.0 0 0.0 0 4.6 Kakaw che' 0.0 0 0.0 0 4.5 7 2.0 13 0.0 0 0.0 0 4.5 Haematoxylon sp. 0.0 0 0.0 0 1.1 24 2.1 10 2.8 9 1.5 17 3.9 Unidentified (Inche) 0.0 0 0.0 0 3.6 11 1.6 19 0.0 0 0.0 0 3.6 Lysiloma latisiliqua 2.1 15 3.3 9 1.5 23 2.0 14 0.0 0 0.0 0 3.5 Unidentified (Chamalboox) 1.6 16 0.8 34 1.8 17 0.9 32 0.0 0 0.0 0 3.4 Coccoloba spicata 2.3 13 3.7 7 1.0 27 1.4 25 0.0 0 0.0 0 3.3 Psidium sartorianum 0.6 22 0.9 32 1.6 22 2.1 11 1.2 22 1.3 23 3.3 Platymiscium yucatanum 0.0 0 0.0 0 0.8 29 0.6 46 2.3 12 1.3 25 3.2 Cochlospermum vitifolium 0.0 0 0.0 0 1.6 21 1.1 29 1.3 20 0.7 39 2.9 Machaonia lindeniana 1.3 18 2.0 12 1.1 26 1.8 18 0.6 27 0.6 41 2.9 Lonchocarpus xuul 0.0 0 0.0 0 1. 7 19 0.8 37 1.2 21 0.4 50 2.9 Gymnanthes lucida 0.6 24 0.4 48 0.0 0 0.0 0 2.3 13 1.3 26 2.9 Talisia floresii 0.7 21 1.0 29 0.3 32 0.6 43 1.4 19 1.5 18 2.3 Neomillspaughia emarginata 0.0 0 0.0 0 1.7 18 1.2 27 0.6 26 0.3 60 2.3 Talisia olivaeformis 0. 6 25 0.4 49 0.4 31 0.3 63 1.4 18 0.7 38 2.3 Unideintified (Kan chan) 0.0 0 0.0 0 0.3 33 0.2 71 1.9 15 0.7 40 2.2 Dalbergia glabra 0.8 20 0.4 50 0.4 30 0.2 76 1.0 24 0.4 58 2.2 TOTAL PERCENTAGE 0.8 0.4 0.7 0.4 2.9 1.0

PAGE 148

136 Table 9. Melliferous imp ortance value index and its ranking (MIVI and MIVI Rank), importance value index and its ranking (IVI and IVI Rank), and the sum of MIVI values for upland species of young, intermediate and older vegetation ages. Species are ordered according to the Sum of MIVI values. Red numbers are the species that account for the top 50% values. SPECIES UPLAND/ YOUNG UPLAND/ INTERMEDIATE UPLAND/ OLDER MIVI MIVI Rank IVI IVI Rank MIVI MIVI Rank IVI IVI Rank MIVI MIVI Rank IVI IVI Rank Sum of MIVI Croton spp. 8.1 3 6.2 4 26.4 1 21.5 1 6.9 4 5.4 4 41.3 Bursera simaruba 8.5 2 6.6 3 8.5 2 6.9 3 11.9 2 9.4 2 28.9 Piscidia piscipula 8.8 1 6.8 2 7.9 3 6.5 4 5.5 7 4.3 5 22.2 Gymnanthes lucida 0.0 0 0.0 0 0.0 0 0.0 0 15.7 1 12.4 1 15.7 Gymnopodium floribundum 4 .8 5 3.7 6 0.6 22 0.5 37 7.9 3 6.2 3 13.2 Thouinia paucidentata 3.3 11 1.7 19 3.3 8 1.8 13 6.1 6 3.2 9 12.7 Hampea trilobata 5.9 4 3.0 9 4.6 5 2.5 10 1.3 19 0.7 33 11.8 Vitex gaumeri 4.6 7 3.5 7 4.6 6 3.7 8 1.9 13 1.5 20 11.0 Lonchocarpus rugosus 3 .5 10 1.8 18 3.2 9 1.7 14 2.0 12 1.1 23 8.7 Lysiloma latisiliqua 3.0 14 4.6 5 5.0 4 8.2 2 0.2 24 0.4 51 8.2 Luehea speciosa 1.5 22 0.8 35 0.0 0 0.0 0 6.3 5 3.3 8 7.8 Coccoloba spicata 4.7 6 7.2 1 2.9 11 4.8 7 0.2 25 0.3 53 7.8 Manilkara zapota 3.5 9 1.8 17 0.0 0 0.0 0 3.2 9 1.7 19 6.7 Lonchocarpus longistylus 1.0 26 1.5 21 3.8 7 6.2 5 1.9 14 2.9 10 6.6 Metopium brownei 4.5 8 3.5 8 1.7 16 1.4 17 0.0 0 0.0 0 6.2 Caesalpinia gaumeri 0.0 0 0.0 0 0.6 23 0.5 38 5.1 8 4.1 6 5.7 Eugenia yucatanensis 1.8 16 2.8 10 1.2 17 1.9 11 2.5 11 3.9 7 5.5 Nectandra coriacea 1.3 23 2.1 15 3.1 10 5.0 6 0.0 0 0.0 0 4.4 Allophyllus cominia 2.3 15 1.2 26 2.0 13 1.1 20 0.0 0 0.0 0 4.3 Casimiroa tetrameria 3.3 12 1.7 20 0.9 20 0.5 42 0.0 0 0.0 0 4.2 Croton sp. ( susyuk) 3.1 13 2.4 14 0.6 25 0.5 44 0.3 22 0.3 58 4.0 Guettarda combsii 1.8 17 2.8 11 1.9 14 3.2 9 0.3 23 0.4 47 4.0 Neea fagifolia 1.6 21 2.4 13 0.7 21 1.2 18 1.4 17 2.2 13 3.7 Machaonia lindeniana 1.7 18 2.7 12 0.4 26 0.6 25 1.4 18 2.1 14 3.4 Tali sia olivaeformis 0.3 29 0.3 58 0.6 24 0.5 43 2.5 10 2.0 15 3.4 Unidentified (Tam'ay) 0.9 27 0.7 40 2.3 12 1.9 12 0.0 0 0.0 0 3.2 Unidentified (Chamalboox) 0.3 28 0.2 62 0.9 19 0.5 35 1.7 16 0.9 28 3.0 Dipholis salicifolia 1.1 25 0.8 33 1.8 15 1.4 16 0 .0 0 0.0 0 2.8 Psidium sartorianum 0.3 30 0.5 49 0.3 27 0.5 36 1.9 15 2.9 11 2.5 Unidentified (Tojyu) 1.6 20 1.3 25 0.0 0 0.0 0 0.8 20 0.6 36 2.4 Exothea diphilla 1.7 19 1.3 22 0.0 0 0.0 0 0.5 21 0.4 50 2.2 Dendropanax arboreus 1.1 24 0.6 46 0.9 18 0.5 33 0.0 0 0.0 0 2.0 TOTAL PERCENTAGE 89.8 76.3 90.7 85.4 89.3 72.6 270

PAGE 149

137 ranked 2 for MIVI and 3 for IVI. Gymnanthes lucida ranked 3 for MIVI and 4 for IVI. This is a pollen yielding species that flowers during the harvest honey season. Lysil oma latisiliqua ranked 4 for MIVI, although it ranked 2 for IVI. The melliferous value of this latter species is lowered because it flowers during the high water content honey period. For the older vegetation age class of the lowland forest type six speci es account for the first 50% of MIVI values. The 1 st ranking species regarding MIVI is Coccoloba reflexiflora a species that flowers during the high water content honey period but that had a very high IVI value, ranking also 1 st in this category. Manilk ara zapota ranked 2 nd for MIVI, although 6 th for IVI. The latter is a species that yields pollen during the recovery period therefore its melliferous importance is increased. Gymnopodium floribundum ranked 3 rd both for MIVI and IVI and Croton spp. 4 th fo r MIVI and 5 th for IVI. Metopium brownei ranked 5 regarding MIVI and 7 regarding IVI, and Chrysophyllum mexicanum ranked 6 for MIVI and 11 for IVI. For this forest/vegetation age type some of the highly ranking species regarding IVI are not melliferous a nd therefore do not show in Table 7. For the upland older vegetation age type, five species account for the first 50% of MIVI values. Gymnanthes lucida ranks 1 st both for MIVI and IVI. Piscidia piscipula Gymnopodium floribundum and Croton spp ranked 2 n d 3 rd and 4 th respectively also for both values. The latter most important species is Luehea speciosa ranking 5 for MIVI and 8 for IVI. For this species the melliferous value is increased given that it flowers during the recovery period. Beehive locati on The apiaries observed for this study are distributed throughout La Montaa region aggregated according to the different communities where the sampled beekeepers live

PAGE 150

138 (Figure 14). The apiaries shown in Figure 14 do not constitute all apiaries in those a reas but are only a sample. The general distribution of the different apiaries is clumped in areas either near roads or in areas where active agriculture is occurring. According to interviewed beekeepers, the criteria to establish apiaries are based on th e vegetation that surrounds the apiaries and the distance to other apiaries. The distance to other apiaries is considered in order to reduce bee competition to floral sources. Most people consider a minimum of 1 km to be sufficient, although some people place their apiaries at a minimum distance of 3 km from other apiaries. Some people prefer to locate apiaries near water bodies although most people provide this source themselves. Accessibility is also a constraint since beekeepers need to visit apiaries often and transport heavy loads, including water, honey and equipment (e.g., extractors), most often by bicycle or horse. However, people in the area are used to traveling long distances and transporting heavy loads. In most cases, people establish their apiaries close to their agricultural plots to concentrate work in a single place (as all beekeepers also depend on the milpas for their subsistence). Nevertheless, it is often the case that people may have two agricultural plots (a newly felled milpa and a second or third year milpa), very often located in different and distant areas. It is also often the case that the apiary is located where the milpa used to be, as a milpa only lasts from 2 to 4 years. According to interviews, the vegetation surroundin g the apiaries is the most important criterion for locating beehives. People need to make a choice as different vegetation types have different advantages. According to local beekeepers, recently abandoned or disturbed areas are important at the beginni ng of the beekeeping annual

PAGE 151

139 cycle, specially for the recovery period since many species, particularly climbers such as Turbina corymbosa (L.) Raf. and Jacquemontia pentantha (Jacq.) G. Don, bloom at this time, yielding abundant nectar. Viguiera dentata (C av.) is another important species at the beginning of the beekeeping annual cycle that yields abundant nectar and that generates the honey usually harvested in the first period of the honey harvest season. However, the flowering of this species and others found in disturbed and recently abandoned areas ends early. On the other hand, flowering of species present in secondary vegetation of more advanced stages of recovery, as well as in older forests, is also considered by beekeepers to be rich in species a bundance important for apiculture. Flowering in this older vegetation type is known to begin later but to last for a longer period of time. Besides the different advantages presented by the different vegetation ages, people consider different vegetation t ypes to offer differing flowering sources. Lowlands and uplands, and the subtypes identified by local beekeepers, differ to some extent in the sources they offer. Savanna types of vegetation are also important nectar sources. Although all of these type s of vegetation are considered by beekeepers to be rich in melliferous species abundance and very often share some of the most important species, the different forest types may respond differently to dry or very wet years. Beekeepers consider that to a la rge extent, nectar flows are not dependent on the vegetation but rather on the particular rainy season. As a strategy for taking advantage of the different vegetation types, people either locate their apiaries in an area that is close to patches of differe nt types of vegetation, or establish two apiaries, each having access to different types of vegetation types. This

PAGE 152

140 Figure 14. Locations of a sample of apiaries of 4 communities at La Montaa, Campeche, Mexico. latter strategy is considered by some peo ple as a way to make use of the advantages that different vegetation types offer. They would, for example, rely on the harvest provided by an apiary surrounded mostly by recently disturbed vegetation at the beginning of the harvest season, and rely on the harvest provided later in the season by beehives established in an older vegetation type. On the other hand, failure of one apiary could be

PAGE 153

141 offset by relying on the second one. This strategy also buffers against bad seasons, as the rains can vary greatl y from location to location, even if they are only a few kilometers apart. Direct management of the vegetation is very seldom practiced. Only a few people stated that they coppice melliferous trees while felling for milpa with the intention of promoting f uture flowering or to destroy species detrimental to apiculture, such as Casimiroa tetrameria Millsp., a species that affects bees during the recovery season. One of the informants was once seen bringing seeds of the vine Jacquemontia pentantha (Jacq.) G. to his apiary. Nevertheless, vegetation management by beekeepers is mostly selection of the location of the apiary. Different communities follow different landscape management strategies according to the vegetation that predominates in the specific com munities, which favor honey production in different ways. In a study conducted in Panama, the foraging range for Africanized Apis mellifera has been established to be from 7.5 to 10 km with a mean flight range of about 1.7 km, and with most foraging occur ring between 2 to 5 km from the nest (Roubik 1989). Beekeepers must therefore choose the location of their apiary according to the landscape they have available. For example, Xnoha (Figure 15), a very isolated area where only one family has lived for mor e than 50 years, has a landscape very advantageous for honey production. Although there is not much active agriculture, this area contains large patches of savanna type vegetations, as well as patches of lowland and upland forests that local beekeepers co nsider to be very rich in melliferous and polliniferous species. Beekeeping in this area also has the advantage that few people share the landscape and therefore competition between apiaries is not a constraint. In this

PAGE 154

142 area honey production is the most important economic activity, followed by cattle ranching, which is also favored by the extensive savanna areas. Figure 15. A sample of apiaries at Xonha in La Montaa, Campeche, Mexico. Red boxes represent beehive locations. Roubik (1989) has de termined that most foraging by Africanized honey bees occurs between 2 and 5 km from nest, with 1.7 km the average flight distance for foraging. At Xmaben (Fig. 16) on the other hand, people rely more on the flowering of species found in disturbed and agri cultural areas, such as Viguera dentata and other species present in early successional stages and along road sides. People in this area lack the advantage of having a higher diversity of patch types in the landscape so they need to rely mostly on these species. People in this type of landscape must also deal with beehive overcrowding. Some people were also concerned about the problems that may arise if people start using pesticides and herbicides in agricultural production.

PAGE 155

143 Fig ure 16. A sample of apiaries at Xmaben in La Montaa, Campeche, Mexico. Red boxes represent beehive locations. Roubik (1989) has determined that most foraging by Africanized honey beb occurs between 2 and 5 km from nest, with 1.7 km the average flight dis tance for foraging. Discussion Rico Gray et al. (1991) calculated that nearly 30% of total flora found in natural vegetation and homegardens at two different sites in Yucatn is important for apiculture. In this study, the percentage of species important for apiculture was greater, although only woody species in natural vegetation were analyzed. Eighty melliferous species were found in all transects sampled for all forest/vegetation age types, representing 59% of all species recorded. This accounted for about 80 % of all species listed by beekeepers as important for honey production in the area, according to interviews and reported

PAGE 156

144 elsewhere (Chapter 5). The portion of the species not encountered in this sampling are species found in vegetation types not included in the sampling design, such as disturbed areas, recently abandoned agricultural sites and savanna type vegetation, and are mostly represented by vines and other annuals. A few species are also found only in urban areas or agroforestry plots. Th e species that stand out as the most important melliferous or polliniferous species, or that make up the first 50% of the MIVI values (which is a function of the importance value index and their flowering period) and that flower during the honey harvest se ason are a group of Croton spp. that are located in all forest/vegetation age types, Gymnopodium floribundum in all forest/vegetation age types except upland intermediate, Bursera simaruba in the young and intermediate age classes of both forest types, Met opium brounei in all vegetation ages of the lowland forest type and the young vegetation age of the upland forest type, Piscidia piscipula in the young vegetation age of both forest types and the older vegetation age for the upland forest Caesalpinia gaum eri in the young and intermediate age classes of the lowland forest, Gymnanthes lucida in the intermediate and older forests of the upland forest type, Vitex gaumeri in the young vegetation age of the upland forest type, and Chrysophilum mexicanum in the o lder vegetation age of the lowland forest type. The species that stand out as the most important melliferous or polliniferous species, or that make up the first 50% of the MIVI values and that flower during the high water content honey period are: Coccolob a reflexiflora for all vegetation ages in the lowland forest, Coccoloba spicata for the young vegetation age of the upland forest type, Eugenia mayana for the intermediate vegetation age of the lowland forest type, and

PAGE 157

145 Lysiloma latisiliqum for the intermed iate vegetation age of the upland forest type. Their importance regarding MIVI values of these species is given primarily by their high IVI values, since flowering during this period provides little melliferous value according to the criteria used for thi s analysis. The most important melliferous or polliniferous species according to MIVI values and that flower during the recovery period were found to be Luhea speciosa for upland old and lowland intermediate, Kakawche (unidentified) for lowland intermedia te, and Manilkara zapota for lowland old. Only Hampea trilobata was amongst the most important species regarding MIVI values that flower during the dearth season, and only for the young vegetation age of the upland forest. Although the relative values of melliferous species are good indicators of their role for apiculture, this does not consider other factors that may be important such as flowering density or their response to weather. For example, Piscidia picipula is considered to be one of the most imp ortant species and people report that this species flowers regardless of rains, contrary to many species that will flower only with adequate precipitation (Chapter 5). Results show that dominance of melliferous and polliniferous species is high in all vege tation age class and forest types included in the sampling. Most of the dominant species in the vegetation contribute as melliferous and/or polliniferous. Therefore, the importance for each of the forest/vegetation age types regarding the dominance of me lliferous species, parallels the dominance of all species, showing no difference between forest types but differences regarding vegetation ages, with the intermediate vegetation ages showing lower values. Studies by Turner et al. (2001) within the

PAGE 158

146 Calakmu l Biosphere Reserve have shown that both forest types (lowlands and uplands) share 80% of their tree species, but that distinctions can be made regarding the relative abundance of each tree species, and in their structural appearance. Differences in veget ation age can be explained as a result of having included the smaller stems in the sampling, resulting in a high dominance in the younger vegetation age given by a higher number of small stems which in the intermediate ages presumably have died out. Older ages where the trees have had time to grow in size increase their dominance values again. Chemas and Rico Gray (1991) also report that there is a strong relationship between the relative importance values of plants in the forests (young or old) and their importance as melliferous species. The proportion of dominance accounted for melliferous and polliniferous species with regards to dominance in general shows a different pattern. The difference is also provided by vegetation age rather then forest type, and in this case the older vegetation age shows lesser values for the proportion of dominance of melliferous and polliniferous species. The above results indicate that all forests/vegetation age types may be important flowering sources for bees, although the younger vegetation ages are most favorable. Younger vegetation is high in dominance and most of the species are nectar or pollen yielding. In intermediate ages, although most of its components are melliferous and/or polliniferous, dominance values of the vegetation in general are lower. On the other hand, although vegetations of older ages are high in dominance as many of the individuals have grown in size, the proportion of dominance belonging to melliferous or polliniferous species is lower. Many of the species that have become important components of these vegetation types do not have melliferous or polliniferous

PAGE 159

147 importance. The latter is also shown in the MIVI and IVI values, since many of the species that make up the first 50% of IVI have no MI VI value. As for the proportion of dominance of the species according to flowering periods and therefore to their contribution as melliferous or polliniferous species during the different periods identified for the beekeeping annual cycle, older forests al so tend to be less important as nectar and/or pollen yielding during the high water content honey period and the honey harvest period. Nevertheless, all forest/vegetation age types hold the same proportion of dominance of species that flower during the de arth and recovery periods. Flowering during these periods, although minimal, is important for the maintenance of bees, and therefore older forests contribute as important sources during this period. Older vegetation ages are also higher in their dominanc e of species, contributing as pollen yielding species only, particularly for the upland vegetation type. This result can be explained by the high dominance value of Gymnanthes lucida at this forest/vegetation age type, a species that yields only pollen. T his study demonstrated that woody species that flower during the recovery period are as scarce as species flowering during the dearth period. Nevertheless, the recovery period is called so because an increase in flowering strengthens bee populations. Mos t of the species important during this period are species found in other vegetation types, such as disturbed and agricultural areas and include species such as vines and other annuals. Thus, at La Montaa, these latter vegetation types are important for t he recovery period in the beekeeping annual cycles, as are fallow and forest areas for the other periods.

PAGE 160

148 Beekeepers in the area are aware of the advantages presented by the different vegetation types found in the area, as reflected by the choices they ma ke when selecting the beehive location. For beekeepers, a landscape that favors honey production is one where different forest types are present, particularly of different ages. If they are not able to locate an apiary that has access to different vegeta tion types, they choose to have apiaries at different sites, even if this means having to work in two different, distant places. The latter is done as a strategy for taking advantage of what different vegetation types have to offer. Nevertheless, all bee keepers stated that disturbed sites are very important. In a study conducted by Chemas and Rico Gray (1991) in Tixcacaltuyub, Yucatan, it was found that the mixture of plant species present in different vegetation patches ( a combination of young and old v egetation) surrounding the beehive is a guaranty of year round flowering. In that area, patches of older vegetation have become rare, a condition that according to beekeepers has resulted in a set back for apiculture. Although they also report that it is considered that about 80% of the honey produced in the Yucatan comes from young successional stages and road side plants, they consider that the lack of older forests prevents beekeepers from enjoying better production. The high abundance of melliferous a nd polliniferous species in the area may be a relic of past management practices as beekeeping has been an important activity in the region for over 3000 years (Chemas and Rico Gray 1991, Guzmn Novoa 1996). Traditionally, beekeeping by the Mayas was part of an agrosilvicultural system (Remmers and Koeijer 1992). During land clearing for the milpa, some small areas are presumed to have been selected where the melliferous tree species were left undisturbed

PAGE 161

149 and hives found inside logs in the forest were ret ained (Sanabrias 1986). Important areas for keeping the beehives were the tolche special forest tracts of different ages and sizes protected alongside roads, corn fields, and communal land. These areas not only provided resources for bees, but also ser ved as fire protection for the hives during burning of corn fields (Remmers and Koeijer 1992). Chemas and Rico Gray (1991) point out that some authors consider these practices to be a form of silviculture and a key factor in the successional process. Toda y, beekeeping may affect the future structure and composition of forests in the region as the value of forests for beekeeping is perceived by local producers and outside markets. A question that should be raised is how will the flora and fauna composition be affected by an increase in foreign pollinators (Buchmann 1996, Corbet 1996). Not only is there evidence that the Africanized bee displaces native pollinators (Roubik 1978, 1996a, 1996b; Roubik et al. 1986), but also that they may be ineffective at pol linating native flowers and may interact in complex ways with native pollinators, reducing the amount and efficiency of pollen transfer (Kearns & Inouye 1997). For beekeepers in the region, beekeeping with Apis bees has negatively affected management of n ative bee species, since according to local informants, Apis mellifera has displaced the native bees they were used to managing. Conclusions According to the analyses of lowland and upland Maya forests of different ages (e.g., young fallow areas from 8 to <19 years old, intermediate ages from 20 30 years old, and areas >30 years old), 80 species were found to have melliferous and/or polliniferous value. This represents 59% of the woody species encountered in all

PAGE 162

150 forest/vegetation age types sampled and com prises about 80% of the species considered important for beekeepers in the region. The other 20% of the important regional species considered important for beekeepers are found in disturbed areas, recently abandoned agricultural sites, urban settings or i n other vegetation types (e.g., savannas). Important melliferous and polliniferous species, ranked by their dominance and harvest season flowering, are: a group of Croton spp., Gymnopodium floribundum Bursera simaruba Metopium brounei Piscidia piscipul a, Caesalpinia gaumeri Gymnanthes lucida Vitex gaumeri and Chrysophilum mexicanum Important species for the high water content honey period are: Coccoloba reflexiflora C. spicata Eugenia mayana and Lysiloma latisiliqum Hampea trilobata was an impo rtant species that flowers during the dearth season and that has high dominance. Luhea speciosa Kakawche (unidentified) and Manilkara zapota are important dominant species that flower during the recovery period. The dominance of melliferous and pollinif erous species was high in all forest/vegetation age types sampled. In fact, most of the dominant species have melliferous and/or polliniferous value. Therefore, all forests/vegetation age types sampled may be important flowering sources for bees, althoug h the younger vegetation ages are most favorable. In terms of dominance according to flowering periods, older forests also tend to be less important as nectar and/or pollen yielding during the high water content honey period and the honey harvest period. All forest/vegetation age types have similar proportions of dominance of species that flower during the dearth and recovery periods. Flowering during these periods, although minimal, is important for the maintenance of bees and therefore, older forests c ontribute as important sources during

PAGE 163

151 this period. Older vegetation ages are also higher in their dominance of pollen only yielding species, particularly the upland forest type. The patchy nature of the landscape at La Montaa offers diverse sources for b ee foraging. While many of the important melliferous and polliniferous species are found in recently disturbed, agricultural and urban areas, particularly contributing to the recovery and early harvest season periods, forests of different ages contribute in the remaining of the beekeeping annual cycle periods. While forests of older ages are restricted nectar sources, they contribute to increasing the options that bees have during the dearth period and are also important pollen sources. The advantages of having a diverse landscape in which different vegetation types have different melliferous and/or polliniferous values is recognized by local beekeepers, who try to obtain the most of what the landscape offers by carefully selecting their apiary location. Each beekeeper either locates an apiary where bees can have access to patches of different vegetation types, or locate two or more apiaries at different sites that have access to different vegetation types. The latter strategy also helps buffer against se asonal adversities. The location of apiaries is also constrained by the occurrence of other apiaries to avoid over crowding and by accessibility (so that apiaries are usually located near where beekeepers perform their agricultural tasks). Apiary establi shment also depends on the particular community and its surrounding landscape. As a result, beekeepers that live in more isolated, less populated areas rely mostly on flowering from the forest, and beekeepers that live in more agricultural and populated a reas rely mostly on flowering

PAGE 164

152 sources from disturbed sites. The latter have more problems with overcrowding and risks that their honey may be contaminated with pesticides and herbicides. Direct management of the vegetation by beekeepers is minimal; only a few beekeepers reported direct manipulation of vegetation to favor honey yielding species. Since land is not limiting in the area, they have not needed to manage forest patches (e.g., reserve certain areas specifically for bee forage). However, much o f the honey produced in other similar regions comes solely from disturbed areas, so it seems likely that beekeepers can adjust management practices to fit the available vegetation. While beekeepers perceive a benefit of having a diverse landscape where fo rests are present, conservation of forested areas will occur only if it becomes an objective for beekeepers and producers in general. This could occur as a result of market incentives (Chapter 4). The high abundance of melliferous and polliniferous specie s in the area has been proposed to be a relic of past management practices, since beekeeping has been an important activity in the region for over 3000 years (Chemas and Rico Gray 1991, Guzmn Novoa 1996). Today, beekeeping may affect the future structure and composition of forests in the region. Forest conservation could be favored by apiculture, if the value of forests for beekeeping is perceived by local producers and outside markets and directed management is provided (e.g., certain areas are set asid e as bee forage). On the other side, beekeeping as practiced today with Apis mellifera may affect the future composition of forests by favoring the population of exotic pollinators, which could lead to the displacement of natives and result in unknown con sequences to floral and faunal assemblages.

PAGE 165

153 CHAPTER 7 CONCLUSIONS The landscape of La Montaa, a buffer zone to the Calakmul Biosphere Reserve in Campeche, Mxico, has been shaped by natural and human induced disturbances through time. Human activity in the area has been constant, and different episodes of occupation have modified this landscape in different ways. After forests recovered from the high populations and intensive use during the Classic Maya period, forest cover has remained steady, with the people relying mostly on timber and non timber forest products for income and survival. Today, land use activities are changing towards more agricultural production, having potentially important implications for the forested landscape of La Montaa. An analysis of a four year period using remotely sensed i magery indicated that forest conversion may be intensifying, as cattle raising becomes a management goal for people in the area; communication at the regional, national and international levels increases; and the population grows (Chapter 2). In view of th e fact that forest conversion in Mexico is occurring very rapidly and that La Montaa represents a buffer zone to one of the largest protected areas, the Calakmul Biosphere Reserve, it is of the utmost importance that the implementation of management pract ices that favor forest conservation are encouraged. Beekeeping, a long practiced activity, has been recognized as having great potential for bringing greater benefits to the local communities while preserving a forested landscape (Acopa and Boeje 1998, Ga lletti 2000, Porter 1995).

PAGE 166

154 In Mexico, apiculture is an important economic activity and the Yucatan Peninsula is one of the most important areas within the country (Echazarreta et al. 1997, Snchez Vazquez and Colli Ucan 1992). In part, this results from t he rich natural melliferous and polliniferous flora found in the area (Chapter 5) and from the beekeeping tradition of the Maya population. Beekeeping with native bees of the subfamily Melinponinae began with the Mayas over 3000 years ago (Guzman Novoa 19 96). Intensive management of certain species, particularly Mellipona beechei resulted in a bee culture important in religion, medicine, food and trade (Labougle and Zozaya 1986, Sanford 1973). During the 20 th century, beekeeping with Apis mellifera star ted in the Yucatan Peninsula, increasing in importance throughout the century. During the 1960s and 1970s beekeeping with Apis mellifera by the Maya population also gained importance, becoming one of the principal economic activities for rural communities (Echazarreta et al. 1997). During the last half of the 1980s, the arrival of the Africanized honey bee resulted in new challenges and opportunities for beekeepers, as it did throughout most of the American continent (Spivak et al. 1991). At La Montaa, h oney production is currently the number one income generating activity. Although there are traces of traditional beekeeping, as some individuals still keep colonies of native bees, this is becoming less and less common. A large proportion of the populati on manages Apis mellifera bees, with the objective of commercializing the products through the regional, national and international markets. The arrival of the Africanized bee (with its high numbers of feral populations), together with better communicatio n and transportation systems, has brought new prospects for beekeeping in the region.

PAGE 167

155 The use of timber and non timber forest products has been the major land use at La Montaa. Today, agricultural activities are gaining importance as the area has become regionally accessible through better infrastructure and communication systems. Timber resources do not generate much profit, as most of the valuable species (e.g., Cedrela odorata and Swietenia macrophylla ) have already been harvested and markets for othe r species have not developed. Non timber forest products, such as chicle extraction and honey production, represent the current best alternatives for profiting from forest resources. Current market trends for chicle have not been favorable. On the other hand, as beekeeping has recently developed, increasing amounts of the honey are being sold nationally and internationally. Unfortunately, so far profits benefit only a small sector of the population that has better management skills and investment capaci ty. Beekeeping could provide economic benefits to a larger sector of the population. An example of this is provided by Lol kax, a cooperative formed recently by a small number of beekeepers from different communities. Lol kax is beginning to establish market ties with the international fair market which offers better prices. Nevertheless, the experience has not been easy, as Lol kax needs to find mechanisms of self financing and complying with high quality standards at a time when skills and resour ces are still extremely limited. Better technology and skills, as well as the diversification of the activity (e.g., to include the production of organic honey, pollen, royal jelly, propoli, etc.) could provide more benefits to the region. Nevertheless, for this to happen there needs to be external investment from NGOs or other agencies, particularly to increase the organizational, administrative and technical skills of local beekeepers.

PAGE 168

156 The landscape at La Montaa is well suited to apiculture, as the dif ferent vegetation types making up the landscape are rich in nectar and/or pollen yielding species. More than 100 species from at least 67 genera and 31 families of trees, shrubs, vines, woody vines and annuals were found to be important melliferous and/or polliniferous sources by beekeepers (Chapter 5). These species are present in the different vegetation types found in the region, including lowland and upland forests, fallow areas and recently disturbed sites (Chapter 6). Several important species are also found in savanna communities and in urban and traditional agricultural areas (better described as agroforestry systems). Most of the species have an annual flowering pattern, although some exhibit biannual or continuous flowering. Different species contribute to honey production to different extents, but they all may play a role in the annual beekeeping cycle. The plant pollinator landscape where honey bees forage, is a function of their response to spatial and temporal variation in floral resource s (Bronstein 1995). Floral sources are distributed in the landscape both in space and time, according to the distribution and abundance of melliferous and polliniferous species and according to the flowering periods of these species. A flowering phenology study found that La Montaa is highly seasonal and that most plant species important as nectar or pollen sources flower at the end of the dry period or beginning of the rainy season, although many flower for the entire duration of the dry season. Little flowering occurs during most of the rainy season and at the transition to the dry period there is an increase in flowering. For the beekeeping cycle, the dry period represents the best quality honey harvest season, followed by a period (the

PAGE 169

157 beginning of t he rainy season) when honey has a high water content and is therefore of lesser value. Most of the rainy season is a dearth period for bees, followed by a recovery period during the transition to the dry season. The soil plant atmosphere continuum that de termines tree water status in the different species determines flowering phenology to a large extent. Cabauelas the winter rains that sporadically occur in the dry period, seem to be of great importance for flowering events during the honey harvest seas on. Therefore, honey production is subject to high variation from year to year and from site to site. Timing of the onset of the rainy season also affects the extent to which high water content honey is produced each year. The beekeeping cycle in this fo rested area of the Maya region contrasts to more northern areas where modern agriculture has shaped the landscape to a larger degree. At La Montaa, the great diversity of forest species contributes substantially to honey production. The honey harvest se ason occurs later than in agricultural areas, since it is most dependent on species that flower later in the dry season and at the beginning of the rainy season (i.e., forest trees). On the other hand, honey in more agriculturally oriented areas is produc ed mostly from the flowering of shrubs and creepers characteristic of recently disturbed areas. These species flower earlier during the beginning of the dry season. Much of the harvest season in the more agricultural areas is the recovery period in La Mo ntaa. These characteristics have implications for regional honey production and marketing. On the negative side, honey from forested areas like La Montaa is produced from multiple species, as opposed to larger quantities of honey that are produced from single species in agricultural settings, the latter a characteristic generally favored by

PAGE 170

158 consumers. Honey produced from forests may be darker than honey produced from single species, another attribute not favored by the current market that prefers light colored honey. Also, more honey produced at La Montaa has a high water content, since it is dependent mainly on species that flower later in the dry season and at the beginning of the rainy season. On the positive side, there are also benefits to be rea lized from honey production in forested areas. For example, honey is produced from plants that have not been treated with pesticides or herbicides, as they often are in agricultural areas, so that contaminant risk is low. Beekeepers at these sites depend less on external sugar feeding, since the diversity of plant species provides a wider array of sources where bees may obtain nectar and pollen. More research is needed on the relative nutritional qualities of honey produced from multiple flowering plants in forests. Honey from La Montaa and similar areas can also be regarded as a non timber forest product obtained from trees that are used for other economic activities and thus present other complementary benefits. According to my analyses of lowland an d upland forests of different ages (e.g., young fallow areas from 8 19 years old, intermediate from 20 30 years old, and >30 years old), 80 species were found to have melliferous and/or polliniferous value. This represents 59% of the woody species encount ered in all forest/vegetation age types sampled and comprises about 80% of the species considered important for beekeepers in the region. Other important melliferous and polliniferous species are found in disturbed areas, recently abandoned agricultural s ites, urban settings or other vegetation types such as savannas.

PAGE 171

159 Important melliferous and polliniferous species ranked by their dominance and that flower during the honey harvest season are: a group of Croton spp., Gymnopodium floribundum Bursera simaru ba Metopium brounei Piscidia piscipula, Caesalpinia gaumeri Gymnanthes lucida Vitex gaumeri and Chrysophilum mexicanum Important species for the high water content honey period are Coccoloba reflexiflora C. spicata Eugenia mayana and Lysiloma latis iliqum Hampea trilobata was an important species that flowers during the dearth season and that has high dominance. Luhea speciosa Kakawche (unidentified) and Manilkara zapota are important dominant species that flower during the recovery period. Acco rding to results, the dominance of melliferous and polliniferous species was high in all forest/vegetation age types sampled. In fact, most of the dominant species have melliferous and/or polliniferous value. Therefore, all forests/vegetation age types s ampled may be important flowering sources for bees, although the younger vegetation ages are most favorable. In terms of the proportion of dominance according to flowering periods, older forests tend to also be less important as nectar and/or pollen yield ing during the high water content honey period and the honey harvest period. All forest/vegetation age types have similar proportions of dominance of species that flower during the dearth and recovery periods. Flowering during these periods, although min imal, is important for the maintenance of bees and, therefore, older forests contribute as important sources during this period. Older vegetation ages are also higher in their dominance of pollen only yielding species, particularly the upland forest type. The patchy nature of the landscape at La Montaa offers diverse sources for bee foraging. While many of the important melliferous and polliniferous species are found in

PAGE 172

160 recently disturbed, agricultural and urban areas, particularly contributing to the re covery and early harvest season periods, forests of different ages contribute in the remaining of the beekeeping annual cycle periods. While forests of older ages are restricted nectar sources, they contribute to increasing the options that bees have duri ng the dearth period and are also important pollen sources. The advantages of having a diverse landscape in which different vegetation types have different melliferous and/or polliniferous values is recognized by local beekeepers, who try to obtain the mos t of what the landscape offers by carefully selecting their apiary location. Each beekeeper either locates an apiary where bees can have access to patches of different vegetation types, or locate two or more apiaries at different sites that have access to different vegetation types. The latter strategy also helps buffer against seasonal adversities. The location of apiaries is also constrained by the occurrence of other apiaries to avoid over crowding and by accessibility (so that apiaries are usually lo cated near where beekeepers perform their agricultural tasks). Apiary establishment also depends on the particular community and its surrounding landscape. As a result, beekeepers that live in more isolated, less populated areas rely mostly on flowering from the forest, and beekeepers that live in more agricultural and populated areas rely mostly on flowering sources from disturbed sites. The latter have more problems with overcrowding and risks that their honey may be contaminated with pesticides and he rbicides. Direct management of the vegetation by beekeepers is minimal; only a few beekeepers reported direct manipulation of vegetation to favor honey yielding species. Since land is not limiting in the area, they have not needed to manage forest patch es (e.g.,

PAGE 173

161 reserve certain areas specifically for bee forage). However, much of the honey produced in other similar regions comes solely from disturbed areas, so it seems likely that beekeepers can adjust management practices to fit the available vegetatio n. While beekeepers perceive a benefit of having a diverse landscape where forests are present, conservation of forested areas will only occur if it becomes an objective for beekeepers and producers in general. This could occur as a result of market ince ntives (Chapter 4). Beekeeping per se does not necessarily ensure a focus on forest conservation, since many important species are found in secondary vegetation and agricultural areas. However, there is no doubt that beekeepers benefit from the multiple f lowering afforded by forests. External factors, such as the development of alternative markets and increased consumer demands for honey produced in these areas, could contribute significantly to the role of beekeeping in forest conservation. Through mark et development, beekeepers could obtain added product value, as well as better conditions for business negotiations, for cooperative organizations, and for obtaining new investments. These factors could induce beekeepers to practice forest conservation as part of their management system. Beekeepers are already starting to form cooperatives to commercialize honey through alternative markets. Also, some certified organic honey is being produced through the support of NGOs, which in some cases has led to t he establishment of forest reserves as bee foraging zones (Acopa and Boege 1998). Honey production at La Montaa and similar areas represents a promising activity that can favor both the conservation of natural resources and socio economic development. Nevertheless, for this to occur, there needs to be a directed effort through

PAGE 174

162 the implementation of specific management practices. Alternative markets may play a crucial incentive role. It is important to keep in mind that beekeeping as practiced today fav ors the increase of an exotic bee species that has proven to displace natural pollinators, with unknown consequences for floral and faunal composition (Buchmann 1996, Roubik 1996a, Sugden 1996). More efforts should be made to promote beekeeping with sting less bees. Future studies should address the implication that beekeeping with Africanized Apis mellifera may have on native pollinators, and ultimately on the structure and composition of the flora and fauna. Nevertheless, beekeeping at La Montaa and ot her similar areas in the Yucatan Peninsula presents an economic alternative to other land uses such as ranching, that surely produce a greater impact on forest composition.

PAGE 175

163 APPENDIX A LIST OF THE IMPORTAN T MELLIFEROUS AND PO LLINIFEROUS SPECIES ACCORDING TO THEIR F LOWERING PERIODS IN RELATION TO THE ANNUAL BEEKEEPING CY CLE AT LA MONTAA, C AMPECHE, MEXICO,

PAGE 176

164 Table A 1. Harvest season flowering species. Scientific Name Family Common Name N/P Flowering period Habit Habitat Bucida buceras L. Combretaceae Pukte' N mid Jan. to mid May tree lowland forest Bursera simaruba (L.) Sarg. Burseraceae Chaka N April to May tree secondary veg./upland forest Caesalpinia gau meri Greenman Leguminosae Kitiinche' N Jan. to Feb. tree upland forest/secondary veg. Caesalpinia yucatanensis Greenman Leguminosae Taak'inche' P February tree secondary veg., lowlands Ceiba pentandra (L.) Gaertn Bombacaceae Ceibo/Ya'ax che' N Jan to Feb. tree upland forest and urban areas Chrysophyllum mexicanum Brandegee ex Standley Sapotaceae Chii keej/Caimito de monte P March tree upland forest Cochlospermum vitifolium (Willd.) Spergel Cochlospermaceae Chu'um P Feb. to March tree secon dary veg./upland forest Cordia dodecandra A. DC. Boraginaceae Siricote P April to May tree lowland forests/urban areas Cordia gerascanthus L. Boraginaceae Bojom N April to May tree upland forest Crescentia cujete L. Bignoniaceae Giro/Was N Jan. to May tree lowland forests and savannas Croton glabellus L. Euphorbiaceae Sak p'eres N end March to big. June tree secondary veg. Croton guatemalensis Lotsy Euphorbiaceae P'eres de akalche' N end March to big. May tree secondary veg./lowlands Cro ton sp. Euphorbiaceae susuyuk N end Feb to mid March a med.marzo tree secondary veg./upland forest Croton sp. Euphorbiaceae Yash p'eres N March to big. May tree secondary veg./upland forest Dipholis salicifolia Sapotaceae Tsiists yaj P March to Apri l tree secondary veg./upland forest Ehretia tinifolia L. Boraginaceae Beek/Roble N March to April tree upland forest and urban areas Exothea diphylla (Standley) Lundell Sapindaceae Wayum koox N February tree upland forest Gliricidia sepium (Jacq.) Steudel Leguminosae Balche kej N mid Feb. to April tree secondary veg./upland forest Gymnanthes lucida Sw. Euphorbiaceae Bak'ayum/Yaytil/Xpij P April tree secondary veg./upland forest Gymnopodium floribundum Rolfe Polygonaceae Tsitsilche' N Mar ch to June tree secondary veg./upland forest Haematoxylon brasiletto Karsten Leguminosae Sak tinto/Tinto de Sabana N end Jan. to mid March tree lowland forests Haematoxylon campechianum L. Leguminosae Chak tinto N end Jan. to mid March tree lowland f orests/savannas Lonchocarpus hondurensis Benth. Leguminosae Burro che'/Balche k'aax N March tree near water bodies Metopium brownei (Jacq.) Urban Anacardiaceae Chechem N March to April tree secondary veg./upland forest Neomillspaughia emarginata ( Gross.) Blake Polygonaceae Sa'itsab N March tree secondary veg./ upland forest No identificada Tojyu N March to April tree secondary veg./upland forest No identificada Chik che' N/P April to May tree upland forest No identificada Ta'amay N M ay tree secondary veg. y upland forest No identificada Sihum N January to February tree near water bodies Piscidia piscipul a (L.) Sarg. Leguminosae Ja'abin N April to May tree secondary veg. y upland forest

PAGE 177

165 Table A 1.(continued) Scientific Name Family Common Name N/P Flowering period Habit Habitat Pisonia acuelata L. Nyctaginaceae Be'eb N February climber secondary veg./lowland forest Pithecellobium platylobum (DC.) Urb Leguminosae Sak muk N mid April to mid June climber upland forest Platymiscium yucatanum Standley Leguminosae Granadillo N mid Feb. to mid March tree upland forest/lowland forest Pseudobombax ellipticum (Kunth) Dugand Bombacaceae Chulte/Amapola P March to April tree upland forest Swartzia cubensis (Brit ton & Wilson) Standley Leguminosae K'ataloox P mid March to mid May tree upland forest Tabebuia chrysantha (Jacq.) Nichols. Bignoniaceae Jo'ok'ab/makulis N Feb. to mid june tree secondary veg./lowland. savannas Tabebuia roseae (Bertol.) Donn Smith Bignoniaceae Jo'ok'ab/makulis N Feb. to mid June tree secondary veg. Talisia floresii Standley Sapindaceae Kolok N mid. April to mid May or June tree upland forest Talisia olivaeformis (HBK.) Radik. Sapindaceae Guaya/guayum N March tree upland f orest/urban areas Vitex gaumeri Greenman Verbenaceae Ya'ax nik N mid. April to mid July tree secondary veg./upland forest Viguiera dentata (Cav.) Sprengel Compositae Tajonal/Tah N/P January to February annual recently disturbed areas Table A 2 High water content honey period flowering species Scientific Name Family Common Name N/P Flowering period Habit Habitat Acacia angustissima (Miller) Blake Leguminosae K'aantemo' N May tree upland forest Acacia argustifolias Leguminosae Subint ul N May to June tree secondary veg./upland forest Acacia gaumeri Blake Leguminosae Box kaatsim P June to July shrub secondary veg. Byrsonima bucidaefolia Standley Malpighiaceae Sak paj/Nance agrio N end May to mid August tree secondary veg./upland forest Caesalpinia violaceae (Miller) Standley Leguminosae Chakte viga P May to June tree upland forest Cedrela odorata L. Meliaceae Cedro/K'uj che N May to June tree upland forest /urban areas or agroforestry plots Coccoloba reflexiflora Sta ndley Polygonaceae Boob chich N end May to July tree lowlands Coccoloba sp. Polygonaceae Boob de sabana N July to August shrub savanna or suuts Coccoloba spicata Lundell Polygonaceae Boob N May to June tree secondary veg./upland forest Colubri na greggi S. Watson Rhamnaceae Puk'in N end June to July arbusto secondary veg./upland forest Cornutia pyramidata L. Verbenaceae Lot che' N April to May or June tree secondary veg. /upland forest

PAGE 178

166 Table A 2 (continued) Scientific Name Family Commo n Name N/P Flowering period Habit Habitat Croton campechianus Standley Euphorbiaceae Kushu che' N May to June tree upland forest (secondary veg.?) Dalbergia glabra (Miller) Standley Leguminosae Chak muk N July to August climber lowlan ds and savannas Diospyros cuneata Standley Ebenaceae Siliil N May to June tree secondary veg. /upland forest Enterolobium cyclocarpum (Jacp.) Griseb. Leguminosae Piich N end April or May to big.Juneio tree upland forest /urban areas Eugenia mayan a Standley Myrtaceae Sak loob N June to July tree lowlands Eugenia yucatanensis Standley Myrtaceae Wirich N May tree secondary veg. Exostema caribaeum (Jacq.) Roem. & Schultes Rubiaceae Chak sabak che' N June tree lowlands near savannas or suu ts Guazuma ulmifolia Lam. Sterculiaceae Piixoy N May to June tree secondary veg. Guettarda combsii Urban Rubiaceae Tasta'ab N June to July tree secondary veg./upland forest Havardia albicans Leguminosae Chukum P June to July tree upland forest Lonchocarpus longistylis Pittier Leguminosae Xu'ul N end June to July tree upland forest Lysiloma latisiliqua (L.) Benth. Leguminosae Tsalam N May to June tree upland forest Machaonia lindeniana Baillon Rubiaceae Kuchel N June to July tree sec ondary veg./upland forest Nectandra coriacea (Sw.) Griseb. Lauraceae Xochyuc/Laurelillo N end May to July tree secondary veg./upland forest Neea fagifolia Heimerl. Nyctaginaceae Ta'tsi N mid May to July tree secondary veg. /upland forest No identif icada Kiis yuk N April to May tree lowlands No identificada Kan chulul/Palo amarillo N May to June tree upland forest No identificada Malpighiaceae Chaka ni' ka' N May to June trepadora recently disturbed areas/ secondary veg./upland forest No identificada Chok che' P May to July tree secondary veg./ lowlands/ savannas No identificada Pakal che' N mid May to July tree upland forest Psidium sartorianum (Berg.) Niedenzu Myrtaceae Pichiche/Guayabillo N June to July tree older f orests Shoepfia schereberi Sak beek N June to July tree secondary veg./upland forest Sickingia salvadorensis Standley Rubiaceae Chakte kok N En May to July tree secondary veg./upland forest Spondias mombin. Anacardeaceae Kiinim N May to July tree lowlands and savannas Spondias sp Anacardeaceae Ju'ujub/Jobo N June to July tree lowlands/savannas/suuts

PAGE 179

167 Table A 3 Dearth period flowering species Scientific Name Family Common Name N/P Flowering period Habit Habitat Trema mycran tha (L.) Blume Ulmifera Sak piixoy/capulin N June to July Shrub Recently disturbed and secondary vegetatopm Allophyllus cominia (L.) Sw. Sapindaceae Ik bach N October to November shrub secondary veg. Bouhinia herrerae (Britton & Rose) Standley & S teyerm. Leguminosae Kibix N Oct. to big. December Annual climber secondary veg., lowlands Dendropanax arboreus (L.) Planch. Et Decne. Araliaceae Sak chaka/Shunan che' N August tree upland forest Hampea trilobata Standley Malvaceae Puk/Majagua/Jool N August to September tree secondary veg./upland forest Lonchocarpus rugosus Benth Leguminosae Kanasin N September tree secondary veg./upland forest Mimosa bahamensis Benth. Leguminosae Sak katsiim N Sept. to big. December shrub secondary veg. No identificada Inche' N August tree near savanna (suuts) No identificada Kan chan N October tree upland forest Pouteria campechiana (H.B. & K.) Baehni ; Calocarpum mammosum Sapotaceae K'aaniste' N August tree upland forest Table A 4. Recove ry period flowering species Scientific Name Family Common Name N/P Flowering period Habit Habitat Casimiroa tetrameria Millsp. Rutaceae Yu'uy N November to December tree secondary veg./upland forest Jacquemontia pentantha (Jacq.) G. Don C onvolvulaceae Solen ak N December to mid Feb. annual climber recently disturbed and secondary veg. Luehea speciosa Willd. Tiliaceae Kas kaat N November to December tree secondary veg./upland forest Manilkara zapota (L.) Van Royen Sapotaceae Zapote or Ya' N November tree upland forest Unidentified Kakaw che' N December tree upland forest Unidentified Chakte k'aax N November tree upland forest Unidentified Sak oox N December tree upland forest unidentified Chamalbox N January tree u pland forest Turbina corymbosa (L.) Raf. Convolvulaceae Xtabentun N December to January Annual climber recently disturbed and secondary veg. Verbesina gigantea Jacq. Compositae Chulke N December to January annual secondary veg. Viguiera sp. Compo sitae Sak tah NN November to December Annual secondary veg.

PAGE 180

168 Table A 5 Multiple flowering species Scientific Name Family Common Name N/P Flowering Period Habit Habitat Brosimum alicastrum Swartz Moraceae Ramn/Oox P twice a year tree upland fore st and urban Cissus sicyoides L. Vitaceae Ta'ka'nil N several times climber secondary veg./upland forest Lonchocarpus xu'ul Lundell Leguminosae Kan chikin che'/Kan xu'ul N twice a year tree lowland forest Thouinia paucidentata Radlk. Sapindaceae Kaanchunup N several times tree upland forest

PAGE 181

169 APPENDIX B FLOWERING EPISODES OF SELECTED SPECIES OBSERVED FROM JANUARY 1999 TRHOUGH MARCH 2000 AT LA MONTAA, CAMPECHE, MXICO

PAGE 182

170

PAGE 183

171

PAGE 184

172 LIST OF REFERENCES Acopa, D., and E. Boege. 1998. The Maya forest in Campeche, Mexico: experiences in forest management at Calakmul. In: R. B. Primack, D. Bray, H. A. Galletti and I. Ponciano, editors. Timber, Tourists, and Temples. Island Press, Washin gton, D.C. Pages 81 97. Adams, R. E. W., W. E. Brown, Jr., and T. P. Culbert. 1981. Radar mapping, archeology, and ancient Maya land use. Science 213(4515):1457 1463. Alencar da Cruz, J. 1998. Fenologia de especies arboreas tropicais na Amazonia centra l.In: C. Gascon and P. Mounthinho, editors. Floresta Amazonica: dinamica, regeneracao, e manejo. Ministerio de Ciencia e Technologia. Instituto Nacional de Pesquisa de Amazonia. Manaos. Pages 25 40. Angulo Carrera, A. 1992. La situacin de la apicultura mexicana. In: FederacinEditorial Mexicana (editor). Agricultura y Tratado de Libre Comercio. Cuadernos Agrarios: Nueva poca, Mxico, D. F. Pages 104 116. Arellano R. J. A., R. Rodrguez R., and P. Uuh C. 1992. Glosario de terminus agrcolas Maya Espa ol. Etnoflora Yucatanense, Fascculo 7 Universidad Autnoma de Yucatn/Sostenibilidad Maya. Yucatn, Mxico. Asquith, N., and R. J. Whittaker. 1992. Rain forest disturbance and recovery. Geography Review. May :11 15. Bastiaan, M. 1984. The sex life of flowers. Facts on File, New York, New York. Basualdo, M. 1996. Polinizacin entomfila: importancia econmica y su implicancia enel sistema agronmico. V Congreso Ibero Latinoamericano de Apicultura. Intendencia Municipal de Soriano, Central Apicola Coope rativa Calmer. Mercedes, Uruguay. Pages 159 161. Barrera A., A. Gmez Pompa, and C. Vzquez Yaes. 1977. El manejo de las selvas por los Mayas: sus implicaciones silvcolas y agrcolas. Biotica 2(2):47 61. Bawa, K. S. 1995. Pollination, seed dispersal an d diversification of angiosperms. Trends in Ecology and Evolution 10(8):311 312.

PAGE 185

173 Bestman, I. J., L. Winkler, and O. Von Helversen. 1997. Headspace analysis of volatile flower scent constituents of bat pollinated plants. Phytochemistry. 46 (7):1169 1172. Binford, M. W., M. Brenner, T. J. Whitmore, A. Higuera Gundy, E. S. Deevey, and B. Lwyden. 1987. Ecosystems, paleoecology and human disturbance in subtropical and tropical America. Quaternary Science Reviews 6:115 128. Borchert, R. 1994. Soil and Stem wat er storage determine phenology and distribution oftropical dry forest trees. Ecology 75 (5):1437 1449. Borchert, R. 1995. Phenology and flowering periodicity of neotropical dry forestspecies: evidence from herbarium collections. Journal of Tropical Ecolog y 12:65 80. Breed, M. D. 1991. Defensive behavior. In: Spivak, M., D. J. C. Fletcher, and M. D.Breed (eds.). The African honey bee. Westview Studies in Insect Biology. Westview Press, Inc. Boulder, Colorado. Pages: 299 308. Brondizio, E. S., E. F. Mora n, P. Mausel, and Y. Wu. 1994. Land use change in the Amazon Estuary: patterns of Caboclo settlement and landscape management. Human Ecology 22(3):249 278. Bronstein, J. L. 1995. The plant pollinator landscape. In: L. Hansson, L. Fahrig, and G.Merriam, ed itors. Mosaic landscapes and ecological processes. Chapman and Hall, London. Pages 257 288. Browder, J.O. 1996. Reading colonist landscapes: social interpretations of tropical forest patches in an Amazonian agricultural frontier. In: Schelhas J., and R. Gr eenberg, editors Forest patches in tropical landscapes. Island Press, Washington, D. C. Pages 285 299. Buchmann, S. L. 1996. Competition between honey bees and native bees in the Sonoran Desert and global bee conservation issues. In: A. Matherson, S. L. B uchmann, C. OTole, P. Westrich, and I.H. Williams (eds.). The conservation of bees. Academic Press, Harcourt Brace, London. Pages 125 142. Bullock, S. H. and J. A. Solis Magallanes. 1990. Phenology of canopy trees of a tropicaldeciduous forest in Mexico. Biotropica 22:22 35. Caballero, J. 1992. Maya homegardens: past, present and future. Etnoecologa 1:35 54. Cairns, M. A., R. Dirzo, and F. Zadroga. 1995. Forests of Mexico: a diminishing resource? Journal of Forestry. July: 21 24.

PAGE 186

174 Carabias Lillo J., an d S. Guevara Sada. 1985. Fenologa en una selva tropical hmeda en una comunidad derivada, Los Tuxtlas, Veracruz. In: A. Gomez Pompa, and S. del Amo, editors. Investigaciones sobre la regeneracin de selvas altas en Veracrz, Mexico. Instituto Nacional de Investigaciones sobre Recursos Bioticos, Xalapa, Veracruz. Pages 27 66. Chemas, A., and V. Rico Gray. 1991. Apiculture and management of associatedvegetation by the Maya of Tixcacaltuyub, Yucatn, Mxico. Agroforestry Systems 13:13 25. Corbet, S. A. 1996 Which bees do plants need? In: A. Matherson, S. L. Buchmann, C. OTole, P. Westrich, and I.H. Williams, editors. The conservation of bees. Academic Press, Harcourt Brace, London. Pages105 113. Corlett, R. T. 1990. Flora and reproductive phenology of the rain forest at Bukit Timah, Singapore. Journal of Tropical Ecology 6:55 63. Corlett, R. T. 1993. Reproductive phenology of Hong Kong shrubland. Journal of Tropical Ecology 9:501 510. Crane, E. 1990. Bees and beekeeping:science, practice and world resour ces. Oxford Heinemann Newnes. 614 pages. Croat, T. B. 1975. Phenological behavior of habit and habitat classes on Barro Colorado island (Panama Canal Zone). Biotropica 7:270 277. Daly, H. V. 1991. Systematics and identification of Africanized honey bees. In: Spivak, M., D. J. C. Fletcher, and M. D. Breed, editors. The African honey bee. Westview Studies in Insect Biology. Westview Press, Inc. Boulder, Colorado. Pages 13 44. Daubenmire, R. 1972. Phenology and other characteristics of tropical semi decid uous forest in north western Costa Rica. Journal of Ecology 60:147 170. Deevey, E. S., D. S. Rice, P. M. Rice, H. H. Vaughan, M. Brenner, and M. S. Flannery. 1979. Mayan urbanism: impact on a tropical karst environment. Science 206(19):298 306. De Steven D., D. M. Windsor, F. E. Putz, and B. de Len. 1987. Vegetative and reproductive phenologies of a palm assemblage in Panama. Biotropica 19:342 356. Dumond, E. 1997. Breve historia de los pacficos del sur. In: Calakmul: volviendo al sur. Gobierno del Est ado Libre y Soberano de Campeche. Campeche, Campeche. Pages 33 49.

PAGE 187

175 Echazarreta, C. M., J. J. G. Quezada Euan, L. M. Medina, and K. L. Pasteur. 1997. Beekeeping in the Yucatan peninsula: development and current status. Bee World 78(3):115 127. ECOMAT. 19 95. Calakmul. Inventario Biolgico. Boletin informativo, UAC, Campeche, Mxico, 15 pp. Ericson, J. 1997. Regional assessment: Calakmul population environment initiative. Unpublished manuscript. Espina, D., and G. Ordetx. 1983. Flora apcola tropical. Edi torial Tecnolgica de Costa Rica. Cartago, Costa Rica. Faust, B. B. 1998. Mexican rural development and the plumed serpent. Technology and Maya cosmology in the tropical forests of Campeche, Mxico. Page 190. Fedick, S. L., and A. Ford. 1990. The prehist oric agricultural landscape of the central Maya lowlands: an examination of local variability in a regional context. Soils and Early Agriculture 22(1): 18 33. Ferr Damar. 1997. Marcos Canul, libertador del sur de Campeche. In: Calakmul: volviendo al su r. Gobierno del Estado Libre y Soberano de Campeche. Campeche, Campeche. Pp. 33 49. Flenley, J. R. 1992. Palynological evidence relating to disturbance and other ecological phenomena of rain forests. In: J. G. Goldammer (editor) Tropical forests in tran sition. Birkhauser Verlag Basel/Switzerland. Pp. 17 24. Fletcher, D. J. C. 1991. Interdependence of Genetics and eclogy in a solution to the African bee problem. In: Spivak, M., D. J. C. Fletcher, and M. D. Breed, editors. The African honey bee. Westvie w Studies in Insect Biology. Westview Press, Inc. Boulder, Colorado. Pages 77 94. Flores, J. S., 1990. The flowering periods of leguminosae in the Yucatan Peninsula in relation to honey flows. Journal of Apicultural Research 29 (2):82 88. Flores J. S., a nd I. Espejel. 1994. Tipos de vegetacin de la Pennsula de Yucatn. Etnoflora Yucatanense, Fascculo 3. Universidad Autnoma de Yucatn/Sostenibilidad Maya. Yucatn, Mxico. Folan, W. 1999. Patrimonio histrico cultural. In: W. Folan, M. C. Snchez, y J. M. Ortega. Naturaleza y cultura en Calakmul, Campeche. CIHS. Universidad Autnoma de Campeche. Pages 71 78. Forman, R. T. T. 1995. Land mosaics. The ecology of landscapes and regions. Cambridge University Press. Great Britain. P. 632.

PAGE 188

176 Frankie, G. W., H. G. Baker, and P. A. Opler. 1974. Comparative phenological studies of trees in tropical wet and dry forests in the lowlands of Costa Rica. Journal of Ecology 62:881 919. Galletti, H. 1999. Plan de Manejo Forestal para La Montaa. Propuesta presentada al I AF por Oxfam/Comadep. Unpublished. Galletti, H. 2000. Evaluacin social del rea focal Nm. 2 La Montaa. Unpublished manuscript. Gates, G. 1992. Fisiografa, geologa, e hidrologa. In: Folan, W.J., Garca O.,J.M., Snchez G.,M.C., (coordinators), Pro grama de Manejo Reserva de la Bisfera de Calakmul, Campeche. UAC, SEDESOL. Unpublished manuscript, Campeche. Gates, G. 1999. Fisiografa, geologa e hidrologa. In: W. Folan, M. C. Snches, and J. M. Ortega, editors. Naturaleza y cultura en Calakmul, Cam peche. CIHS, Universidad Autnoma de Campeche, Campeche. Pages 31 39. Gates, M. 1999a. Uso humano de los recursos en Calakmul: conservacin o regeneracin de un rea perturbada. In: W. Folan, M. C. Snchez, y J. M. Ortega. Naturaleza y cultura en Calakmul, Campeche. CIHS. Universidad Autnoma de Campeche. Pages 123 137. Gates, M. 1999b. Histria econmica. In: W. Folan, M. C. Snchez, y J. M. Ortega. Naturaleza y cultura en Calakmul, Campeche. CIHS. Universidad Autnoma de Campeche. Pages 107 109. Goldeme r, J. G. 1992. Tropical forests in transition. Ecology of natural and anthropogenic disturbance processes an introduction. In: J.G. Doldammer (editor) Tropical forests in transition. Birkhauser Verlag. Basel/Switzerland. Pp. 1 16. Gmez Figueroa, P., an d L. A. Fournier O. 1996. Fenologa y ecofisiologa de dospoblaciones de Tabebuia rosea (Roble de Sabana) en Costa Rica (Bignoniaceae). Revista Biolgica Tropical 44(1):61 70. Gmez Pompa, A. 1987. On Maya silviculture. Mexican Studies 3(1):1 17. Gmez P ompa, A. 1991. Learning from traditional ecological knowledge: insights from Mayan silviculture. In: Gmez Pompa, A., T. L. Whitmore, and M. Hadley (eds.) Rainforest regeneration and management. Man and the Biosphere Series. Pp. 335 341. Gmez Pompa, A. a nd D. A. Bainbridge. 1995. Tropical forestry as if people mattered. In: Lugo, A. E., and C. Lowe, editors. Tropical forests: management and ecology. Springer Verlag, New York. Pages 408 422.

PAGE 189

177 Gmez Pompa, A., and A. Kaus, 1992. Taming the wilderness myth. BioScience 42 (4): 271 279. Gonalves, L. S., A. C. Stort, and D. De Jong. 1991. Beekeeing in Brazil. In: Spivak, M., D. J. C. Fletcher, and M. D. Breed, editors. The African honey bee.Westview Studies in Insect Biology. Westview Press, Inc. Boulder, Co lorado. Pages 359 372. Guzman Novoa, E. 1996. La apicultura en Mexicoy Centro America. V Congreso Ibero Latinoamericano de Apicultura. Intendencia Municipal de Soriano, Central Apicola Cooperativa Calmer. Mercedes, Uruguay. Pages 14 17. Hellmich II, R.L. and T. E. Rinderer. 1991. Beekeeping in Venezuela. In: Spivak, M., D.J. C. Fletcher, and M. D. Breed, editors. The African honey bee. Westview Studies in Insect Biology. Westview Press, Inc. Boulder, Colorado. Pages 399 411. Hilty, S. L. 1980. Floweri ng and fruiting periodicity in a premontanine rainforests in pacific Colombia. Biotropica 12 (4): 292 306. Holling, C. S. 1986. The resilience of terrestrial ecosystems: local surprise and global change. In: W. Clark, and R.E. Munn, editors. Sustainable d evelopment of the biosphere. Pages 292 316. Holling, C. S. 1992. Cross scape morphology, geometry, and dynamics of ecosystems. Ecological Monographs 62(4): 447 502. Holling, C. S., G. Peterson, P. Marples, J. Sendzimir, K. Redford, L. Gunderson, and D. L ambert. 1996. Self organization in ecosystems: lumpy geometries, periodicities and morphologies. In: B. L. Walker, and W. L. Steffen (eds.). Global change and terrestrial ecosystems. Cambridge University Press, Cambridge. Pages 346 384. Janzen, D. H. 1967 Synchronization of sexual reproduction of trees within the dry season ikn Central America. Evolution 21:620 637. Jensen, J. R. 1996. Introductory digital image processing. A remote sensing kperspective. Prentice Hall. New Jersey. 318 pages. Kearns, C. A ., and D. W. Inouye. 1997. Pollinators, flowering plants, and conservation biology. BioScience 47(5): 297 307. Kent, R. B. 1991. The Africanized honey bee in Peru. In: Spivak, M., D. J. C. Fletcher,and M. D. Breed, editors. The African honey bee.Westvie w Studies in Insect Biology. Westview Press, Inc. Boulder, Colorado. Pages 373 398.

PAGE 190

178 Konrad, H. 1999. Histria de la regin. In: W. Folan, M. C. Snchez, and J. M. Ortega. Naturaleza y cultura en Calakmul, Campeche. CIHS. Universidad Autnoma de Campeche. Pages 91 106. Koptur, S., W. A. Haber, G. W. Frankie, and H. G. Baker. 1988. Phenological studies of shrubs and treelet species in tropical cloud forests of Costa Rica. Journal of Tropical Ecology 4:323 346. Labougle, J.M., and Zozaya, J. A. 1986. La ap icultura en Mxico. Ciencia y Desarrollo 69: 17 36. Leigh, Jr,. E. G. 1990. Introduccin: la seleccin natural y los ciclos del bosque. In: E. G. Leigh, Jr. et al., editors. Ecologa de un bosque tropical. Pages 1755 178 Lieberman, D. 1982. Seasonality a nd phenology in a dry tropical forest in Ghana. Journal of Ecology 70:791 806. Lpez Palacios, S. 1986. Catlogo para una flora apcola Venezolana. Consejo de Desarrollo Cientfico y Humanstico de la Universidad de Los Andes. Mrida, Venezuela. Lyon J. and R. H. Horwich. 1996. Modification of tropical forest patches for wildlife protection and community conservation in Belize. In: Schelhas J., and R. Greenberg, editors. Forest patches in tropical landscapes. Island Press, Washington, D. C. Pages 205 230. Marcus, J. 1982. Plant world of the sixteenth and seventeenth century lowland Maya. In: Flannery, K.V., editor. Maya subsistence: studies in Memory of Dennis E. Puleston. Academic Press, New York. Pages 239 273. Masera, O. R., M. J. Ordoes, and R. Dirzo 1997. Carbon emissions from Mexican forests: current situation and long term scenarios. Climatic Change 35:265 1997. Matheny, R. T. 1978. Northern Maya lowland water control systems. In: Harrison, P. D., and Turner II, B. L., editors. Prehispanic Maya a griculture. University of New Mexico Press, Aluquerque. Pages 185 210. McIntosh, 1981. Succession and ecological theory. In: D. C. West, H. H. Shugart, and D. B. Botkin, editors. Forest succession. Concepts and application. Springer Verlag New York, New Y ork. Pages 10 23. Mendes P., S.R., and J.M.F. de Camargo. 1991. Interactions on floral resources between the Africanized honey bee Apis mellifera L. and the native bee community (Hymenoptera: Apoidea) in natural cerrado ecosystems in southeast Brazil. A pidologie 22:397 415.

PAGE 191

179 Michener, C. D. 1973. The Brazilian honeybee. BioScience 23:523 527. Michener, C. D. 1975. The Brazilian bee problem. Annual Review of Entomology 20:399 416. Morales, 1992. Suelos. In: Folan, W.J., Garca O., J.M., Snchez G., M.C. (coordinators). Programa de Manejo Reserva de la Bisfera de Calakmul, Campeche. UAC, SEDESOL, unpublished manuscript. Campeche, 27 34. Newstrom L. E., G. W. Frankie, and H. G. Baker. 1994. A new classification for plant phenology based on flowering patt erns in lowland tropical rain forest trees at La Selva, Costa Rica. Biotropica 26:141 159. Otis, G. W. 1991. Population Biology of the Africanized Honey Bee. In: Spivak, M., D. J. C. Fletcher, and M. D. Breed, editors. The African honey bee.Westview Stud ies in Insect Biology. Westview Press, Inc. Boulder, Colorado. Pages 213 234. Pennington, T. D., and J. Sarukhn. 1998. rboles tropicales de Mxico. Segunda Edicin. Universidad Nacional Autnoma de Mxico, Fondo de Cultura Econmica, Mxico, D. F. Port er, L. 1995. Analysis of a Natural Resources Management System in the Calakmul Biosphere Reserve. Master of Science Thesis. State University of New York, College of Environmental Science and Forestry. Syracuse, New York. Unpublished. Puleston, D. E. 1982. The role of ramn in Maya subsistence. In: Flannery, K.V., editors. Maya subsistence: studies in Memory of Dennis E. Puleston. Academic Press, New York. Pages 353 366. Putz, F. E. 1979. Aseasonality in Malaysian tree phenology. The Malaysian Forester.42 (1):1 24. Quezada Euan, J. J. G., and J. Gonzlez Acereto. 1994. A preliminary study on the development of colonies of Melipona beecheii in traditional and rational hives. Journal of Apicultural Research 33(3): 167 170. Ramayo Lanz, T. 1996. Los mayas pa cficos de Campeche. Universidad Autnoma de Campeche. Campeche, Campeche. 259 pages. Ratnieks, F. L. W. 1991. Africanized bees: natural selection for colonizing ability. In: Spivak, M., D. J. C. Fletcher, and M. D. Breed, editors. The African honey bee Westview Studies in Insect Biology. Westview Press, Inc. Boulder, Colorado. Pages 119 135.

PAGE 192

180 Reich, P. B., and R. Borchert. 1984. Water Stress and tree phenology in a tropical dry forest in the lowlands of Costa Rica. Journal of Ecology 72:61 74. Remmers G. G. A., and H. Koeijer. 1992. The tolche, a Maya system of communally managed forest belts: the causes and consequences of its disappearance. Agroforestry Systems 18:149 177. Rico Gray, V., A. Chemas, and S. Mandujano. 1991. Uses of tropical deciduo us forest species by the Yucatecan Maya. Agroforestry Systems 14:149 161. Rinderer, T. E., and A. M. Collins. 1991. Foraging behavior and honey production. In: Spivak, M., D. J. C. Fletcher, and M. D. Breed, editors. The African honey bee. Westview Stud ies in Insect Biology. Westview Press, Inc. Boulder, Colorado. Pages 235 257. Rinderer, T. E., and R. L. Hellmich II. 1991. The processes of Africanization. In: Spivak, M., D. J. C. Fletcher, and M. D. Breed, editors. The African honey bee. Westview Stu dies in Insect Biology. Westview Press, Inc. Boulder, Colorado. Pages 95 117. Roubik, D. W. 1978. Competitive interactions between neotropical pollinators and Africanized honey bees. Science 201:1030 1032. Roubik, D.W. 1989. Ecology and natural history o f tropical bees. Cambridge University Press. New York, New York. 514 p. Roubik, D. W. 1991. Aspects of Africanized honey bee ecology in tropical America. In: M. Spivak, D. J.C. Fletcher, and M. D. Breed, editors. The African honey bee. Boulder Westview P ress. Pages 259 281. Roubik, D. W. 1996a. Measuring the meaning of honey bees. In: A. Matherson, S. L. Buchmann, C. OTole, P. Westrich, and I.H. Williams, editors. The conservation of bees. Academic Press, Harcourt Brace, London. Pages 164 172. Roubik, D. W. 1996b. African honey bees as exotic pollinators in French Guiana. In: A. Matherson, S. L. Buchmann, C. OTole, P. Westrich, and I.H. Williams, editors. The conservation of bees. Academic Press, Harcourt Brace, London. Pages 173 182. Roubik, D. W., a nd S. L. Buchmann. 1984. Nectar selection by Melipona sp. and Apis mellifera (Hymenoptera: Apidae) and the ecoloy of nectar intake by bee colonies in a tropical forest. Oecologia 61:1 10.

PAGE 193

181 Roubik, D. W., J. E. Moreno, C. Vergara, and D. Wittmann. 1986. Spo radic food competition with the African honey bee: projected impact on neotropical social bees. Journal of Tropical Ecology 2:97 111. Sader, S. A., C. Reining, T. Sever, and C. Soza. 1997. Human migration and agricultural expansion. A threat to the Maya t ropical forests. Journal of Forestry. December: 27 32. Sader, S. A., T. Sever, J. C. Smoot, and M. Richards. 1994. Forest change estimates for the northern Petn region of Guatemala 1986 1990. Human Ecology 22(3):317 332. SAGAR. 1998. Flora nectarfera y polinfera en la Pennsula de Yucatn. Secretara de Agricultura, Ganadera y Desarrollo Rural Subsecretara de Agricultra y Ganadera. Direccin General de Ganadera. SAGAR, Alianza para el Campo. Mxico, D. F. Sanabria, O.L. 1986. El uso y manejo for estal en la comunidad Xul, en el sur de Yucatn. Etnoflora Yucatense. Fascculo 2. INIREB, Xalapa, Veracrz, 191 pp. Snchez, M. C. 1999. Poblacin actual. In: W. Folan, M. C. Snchez, y J. M. Ortega, editors. Naturaleza y cultura en Calakmul, Campeche. CIHS. Universidad Autnoma de Campeche. Pages 111 116. Snchez Vazquez, A. M., and W. Colli Uan. 1992. Apicultura y africanizacin. Los retos de una nueva explotacin. In: A. C. Dacary, D. Navarro Lpez, S. M. Arnaiz Burne, editors. Quintana Roo: los reto s del fin de siglo. Centro de Investigaciones de Quintana Roo (CIQROO). Chetumal, Quintana Roo, Mxico. Pages:169 184. Sanford, M. T. 1973. A geography of apiculture in Yucatan, Mexico. Master of Science Thesis. University of Georgia. Sohn, Y., E. Moran, and F. Gurri. 1999. Deforestation in north central Yucatan (1985 1995): mapping secondary succession of forest and agricultural land use in Sotuta using the cosine of the angle concept. Photogrammetric Engineering & Remote Sensing. 65 (8):947 958. Sneyd J., and S.Camazine. 1991. Selection of high quality nectar sources by honeybee colonies: a self organizing system. In: L.J. Goodman, and R. Fisher, editors. The behavior and physiology of bees. Wallingford, Oxon, UK. Pages 102 115 Snook. L. 1996. Catastro phic disturbance, logging and the ecology of mahogany ( Swietenia macrophylla King): grounds for listing a major tropical timber species in CITES. Botanical Journal of the Linnean Society 12:1 11.

PAGE 194

182 Sosa, V., J. S. Flores, V. Rico Gray, R. Lira, and J. J. Or tz. 1985. Lista florstica y sinonimia Maya. Etnoflora Yucatanense. Fasciculo 1. Instituto Nacional de Investigaciones sobre Recursos Biticos. Xalapa, Veracrz, Mxico. Souza Novelo, N. 1981. Plantas melferas y polinferas que viven en Yucatn. Fondo E ditorial de Yucatn. Mrida, Yucatn, Mxico. Spivak, M. 1991. The Africanization process in Costa Rica. In: Spivak, M., D. J. C. Fletcher, and M. D. Breed (eds.). The African honey bee.Westview Studies in Insect Biology. Westview Press, Inc. Boulder, C olorado. Pages 137 155. Sugden, E. A. 1996. Toward an ecological perspective of beekeeping. In: A. Matherson, S. L. Buchmann, C. OTole, P. Westrich, and I.H. Williams (eds.). The conservation of bees. Academic Press, Harcourt Brace, London. Pages 154 162 Sugden, E. A., R. W. Thorp, and S. L. Buchmann. 1996. Honey bee native bee competition: focal point for environmental change and apicultural response in Australia. Bee World 77:26 44. Toledo, V. M. 1991. Patzcuaros lesson: nature, production, and cult ure in an indigenous region of Mxico. In: Oldfield, M.L., and Alcorn, J.B., editors. Biodiversity, culture, conservation, and ecodevelopment. West View Press. Boulder, Colorado. Pages 147 171. Toledo, V. M. 1977. Pollination of some rain forest plants by non hovering birds in Veracruz, Mexico. Biotropica 9(4):262 267. Turner II, B. L., S. Cortina Villar, D. Foster, J. Geoghegan, E. Keys, P. Klepeis, D. Lawrence, P. M. Mendoza, S. Manson, Y. Ogneva Himmelberger, A. B. Plotkin, D. Prez Salicrup, R. R. Cho wdhury, B. Savitsky, L. Schneider, B. Schmook, C. Vance. 2001. Deforestation in southern Yucatn peninsular region: an integrarive approach. Forest and Ecosystem Management, forthcoming. Turner II, B. L., and Harrison, P. D. 1981. Prehistoric raised fiel d agriculture in the Maya lowlands. Science 213:399 405. Ucan, E., L. Ortega, J. Ortiz, J. Tun, and S. Flores. 1999. Vegetacin y flora. In: Folan, W., M.C. Snchez and J. M. Ortega, editors. Naturaleza y cultura en Calakmul, Campeche. CIHS. Universidad A utnoma de Campeche. Pages 139 155.

PAGE 195

183 Villanueva G., R. 1994. Nectar sources of European and Africanized honey bees ( Apis mellifera L .) in the Yucatan Peninsula, Mexico. Journal of Apicultural Research 33(1): 44 58. Waide, R. B., and A. E. Lugo. 1992. A re search perspective on disturbance and recovery of a tropical montane forest. J. G. Golammer, editor. In: Tropical forests in transition. Birkhauser Verlag Basel/Switzerland. Pages 173 190. Weaver, N., and E. Weaver. 1981. Beekeeping with the stingless be e Melipona beechei by the Yucatecan Maya. Bee World 62(1):7 19

PAGE 196

184 BIOGRAPHICAL SKETCH Luciana Porter Bolland was born and raised in Mexico City. She conducted her bachelors degree studies with a major in biology at the Universidad Autnoma Metropolitana Unidad Xochimilco in Mexico City, and graduated in 1992. In 1993 she moved to Syracuse, New York to conduct a Master of Science program at the State University of New York College of Environmental Science and Forestry. In 1995 she received her degree and returned to Mexico to continue working in the Maya area, where s he had conducted research for her thesis. In 1996 she returned to the U.S. to pursue a Ph.D. program at the School of Forest Resources and Conservation at the University of Florida. After graduation she plans to continue working in the management and con servation of tropical resources area.


Permanent Link: http://ufdc.ufl.edu/UFE0000343/00001

Material Information

Title: Landscape ecology of apiculture in the Maya area of La Montana, Campeche, Mexico
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000343:00001

Permanent Link: http://ufdc.ufl.edu/UFE0000343/00001

Material Information

Title: Landscape ecology of apiculture in the Maya area of La Montana, Campeche, Mexico
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000343:00001


This item has the following downloads:


Full Text












LANDSCAPE ECOLOGY OF APICULTURE IN THE MAYA AREA OF LA
MONTANA, CAMPECHE, MEXICO

















By

LUCIANA PORTER-BOLLAND


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


2001



























Dedicado a los apicultores y apicultoras de La Montafia.















ACKNOWLEDGMENTS

This work was possible thanks to the collaboration and assistance of many people

of the La Montafia area; particularly the Montoy Koh family, especially Jorge and Paula.

Don Juan Montoy and family, Gregorio Montoy and family, Don Beto Camara,

Guadalupe May Pacheco, Gabriela Martin Ek, the Gongora Cauich family at Xnoha, and

many others, who taught me much of what is in these pages and welcomed me into their

homes. I am also thankful to the people of Oxfam/Comadep (particularly Leticia Coello,

Bernard Vanderweerd, and Lamberto Huh) who facilitated my stay at Xmaben and

Merida.

I learned a lot about beekeeping from Victor Camara, who also provided me with

insights and suggestions throughout the field work and the writing phase. Juan Jimemez

Osornio was very helpful by assisting me at the Universidad Aut6noma de Yucatan in

Merida. Juan Tun Garrido helped me with the botanical identification at the Alfredo

Barrera Marin herbarium of the Universidad Aut6noma de Yucatan, making this work

possible. Eddie Ellis, while at work at the Universidad de Quintana Roo, was very

helpful by assisting me in the satellite data analysis and image classification. Hugo

Galletti Busy also provided me with insights during different points of my field work. I

am forever grateful to all of them.

I am very happy to have had Henry Gholz and Mary Duryea as advisors, as they

have always been patient and helpful in many aspects. I am forever indebted to them. I

thank Hugh Popenoe, Marianne Schmink and Michael Binford for serving on my









committee. I am grateful to Jack Putz, Malcom Sanford, Martin Barker, and Michael

Valki for reading drafts of some of the chapters and giving me suggestions. I am also

grateful to the statistics department (particularly Ziyad Mahfoud, Bernhard Klingenberg

and David Hitchcock) for helping me with analysis.

My friends in Gainesville have helped make it a home and are to a great extent

responsible for my having achieved my study goals. I am particularly thankful to

Mercedes Prieto, Jorge Aragon, Mariana Varese, Gabriella Scollo, Michael Valki,

Paulina and Raul Aragon, Victoria Reyes, Vincent Vades, Monica Espinosa, Victoria

Saenz, and many others. I am also thankful to my family for always being there for me.

I am specially thankful to my father, Luis Porter, as he has always been a source of

inspiration to me in many aspects of my life and particularly in the academic one.

The research and writing phase of this work were made possible by funding from

Fulbright/Garcia Robles/CONACYT scholarship, a TCD field grant, a Compton

Foundation grant through the PSTC, an Inter-American Foundation field grant, a TCD

fellowship, and support from the School of Forest Resources and Conservation..
















TABLE OF CONTENTS
page
A C K N O W L E D G M E N T S ............................................... ............................. ...............iii

LIST OF TABLES .................... .......................................viii

LIST O F FIG U RE S .......................... ...................... .. .. ........... ..... .. ....ix

A B ST R A C T .................................................................. ................................ xi

CHAPTERS

1 IN TRODUCTION .................................................. ................... 1

2 THE MAYA LANDSCAPE AT LA MONTANA: PAST AND FUTURE ....................5

In tro d u ctio n ....................................................................... 5
S tu d y A rea ....................................................................... ................................. . 7
L landscape H istory............................ .................... 12
M aya Period ................................................................. .............. 13
Colonial and post-Colonial period........................................... .................... 19
Modern period........................................................... 24
Current situation ...................................................................... 26
M ethods.............................. .................... 27
Surface A naly sis ...................................................... .............. 29
R e su lts ....................................................................... 3 1
D iscu ssio n ..................................................... 4 0
C learn g R ate .............................................................. 4 0
Land Use Change at La M ontafia............................................... .................... 42
C o n c lu sio n s ....................................................................................................... 4 5

3 BEEKEEPING IN THE MAYA AREA: HISTORY AND CURRENT STATUS .......47

In tro d u ctio n ........................................... ............. .......... ... .. .. .. .............. 4 7
Beekeeping Industry in Context: Mexico's Place in World Honey Production and
Relevance of the Yucatan Peninsula ....................... ............................................... 48
History of Beekeeping in the Maya Area ......... ............. ...... ......... 51
Beekeeping with the Africanized Bee.............................................. .................... 53
A fricanized B ee........................................... .................... .............. 53
New Challenges and Opportunities.......................................... .................... 55
Conservation Concerns .. ... ........ .................. .............. 59
Conclusions .......... ......... ........ .............. .. .... ......... .... 60


V









4 SOCIO-ECONOMIC IMPORTANCE OF BEEKEEPING IN THE MAYA FORESTS:
THE EXAMPLE OF THE LA MONTANA AREA .........................................62

Introduction .......................................... 62
M ethods.............................. .............. ...... 63
S tu d y S ite ........................................................................... 6 3
R research Conducted ................................... .............. 64
R esu lts ................... ................................................... 6 5
Beekeeping at La Montafia ............................................................... 65
Socioeconomic Importance of Beekeeping..................................... 67
Honey Production and M marketing .......................................................... ........ 71
The Lol-K'ax Cooperative Example.................. ...... .................... 75
C o n clu sio n s ................................................................. 8 3

5 FLOWERING PHENOLOGY OF MELLIFEROUS SPECIES AND ITS RELATION
TO THE BEEKEEPING CYCLE ................................................................... 86

Introduction................................... ..................... 86
Study Area and M ethods....................................................... ......................... 89
R results and D iscussion...................... .... ......... ................. ............... 93
Melliferous/Polliniferous Species and Their Relation to the Beekeeping Cycle...... 93
Community Level Flowering Phenology .............................................................. 98
Flowering patterns of the different species and their contribution to apiculture. ... 101
H oney-H harvest Flow ering .............................................................. .............. 102
High-Water-Content Honey Flowering ....................................................... 110
Dearth Period Flowering. ............. ........... ................... 113
Recovery Period Flowering ............................................. 115
M multiple Flow ering Species ........................................................ .......... ..... ..... 118
C o n c lu sio n s ...................................................................................................... 1 1 9

6 BEEKEEPING AND THE LANDSCAPE...................... ...................................... 123

Introduction ............................ .............. ...... 123
M ethods....................... ................. 125
Study site ............... .... .... .... .. ........... .......................... .... .. 125
Distribution and abundance of melliferous species ............................................ 126
B eeh iv e lo cation ................................ .................... ................................... 12 8
R esu lts .............. .. .... ......... ... ... ... ... ............................................ 12 9
Distribution and abundance of melliferous and polliniferous species .................... 129
Beehive location........................................................................................... 137
D discussion .............................................. 143
C o n clu sio n s ...................... .... ......... .. .. ........................................ 14 9

7 C O N CLU SIO N S ..........................................................................153






vi









APPENDICES


A LIST OF THE IMPORTANT MELLIFEROUS AND POLLINIFEROUS SPECIES
ACCORDING TO THEIR FLOWERING PERIODS IN RELATION TO THE
ANNUAL BEEKEEPING CYCLE AT LA MONTANA, CAMPECHE, MEXICO, 163

B FLOWERING EPISODES OF SELECTED SPECIES OBSERVED FROM
JANUARY 1999 THOUGH MARCH 2000 AT LA MONTANA, CAMPECHE,
M E X IC O ........................................................................................................... 1 6 9

L IST O F R E FE R E N C E S ........................................................................ ...................172

BIOGRAPHICAL SKETCH ............................................................................... 184















LIST OF TABLES


Table Page
1. Major soil/vegetation types identified by local people at La Montafia, Campeche. ....14

2. Change detection matrix obtained by combining classifications made of the La
Montafia portion of Landsat 5 TM, 02/05/96 and Landsat 7 ETM 03/27/00
sc en e s.. ................................................................................ 3 5

3. Changes in pixel count, area, and percentage of total area of land use/ cover types
obtained from a post-classification analysis of Landsat TM, 02/05/96 and
Landsat 7 ETM 03/27/01 ............................................................................ 37

4. Some production practices taught at the Technical Training Program for Lol-K'ax
members and used in the cooperative's apiaries at La Montafia..........................81

5. Average of dominance (D*), dominance of melliferous species (MD**) and proportion
ofMD (MP***) for transects sampled for the different lowland young,
intermediate and older, and upland young, intermediate and older forest types
at La M ontafia, Cam peche. ...........................................................................130

6. Average proportion of species that flower during the high water-content honey period
(MP 1), of species that flower during the honey harvest period (MP2), and of
species that flower during the dearth and recovery periods (MP3) for transects
sampled for the different lowland young, intermediate and older, and upland
young, intermediate and older forest types at La Montafia, Campeche.............132

7. Dominance of pollen yielding species (PD), and dominance of nectar yielding species
(ND) for transects sampled for the different lowland young, intermediate and
older, and upland young, intermediate and older forest types at La Montafia,
H opelchen, C am peche ............................................... .............................. 132

8. Melliferous importance value index and its ranking (MIVI and MIVI Rank),
importance value index and its ranking (IVI and IVI Rank), and the sum of
MIVI values for lowland species of young, intermediate and older vegetation
ag es. ............................................................................. 13 5

9. Melliferous importance value index and its ranking (MIVI and MIVI Rank),
importance value index and its ranking (IVI and IVI Rank), and the sum of
MIVI values for upland species of young, intermediate and older vegetation
ag es. ............................................................................. 136
















LIST OF FIGURES


Figure Page
1. La M ontafia area at Hopelchen, Campeche, M exico ................................... ............... 11

2. Soil/vegetation types identified by local people at La Montafia, Campeche and their
relation to topography. See Table 1. ................................... ........................... 16

3. Conceptual diagram of procedures for conducting a post-classification comparison
change-detection .............................................. .. ........ .. ........ .... 30

4. Image classification of La Montafia portion of a) Landsat 5 TM image02/05/96; and b)
Landsat 7 ETM 03/27/00 ................................. .....................................33

5. Change map created by combining classification data of the La Montafia portion of
Landsat 5 TM, 02/05/96, and Landsat 7 ETM 03/27/00 scenes. ......................... 36

6. La Montafia area, Campeche, Mexico. Buffers made at: A. 1, 3 and 5 km from roads;
B. At 3, 5 and 8 km from roads; and C. At 0.5 and 1.0 km from lowlands.........39

7. Number of beekeepers that gained, lost, or had no change in their colony numbers from
1988 to 1999, according to a survey of 64 beekeepers of the communities of
Chunek, Pachuitz, Ukum and Xmaben conducted during 1999 ........................73

8. Honey produced by Lol-K'ax during 2000. Values are given for 18, 19 and 20 percent
water content and for total production. Data obtained from Lol-k'ax financial
sum m ary of 2000. ................. .................................................... ..................... 78

9. Metric tons of honey produced in 2000 by each of the 74 producers belonging to the
Lol-K'ax cooperative. Data obtained from Lol-k'ax financial summary of
2 0 0 0 ......................................................................... 8 2

10. Flowering phenology of melliferous and polliniferous species at La Montafia,
C am peche, M exico......... .......................................... .............. .. .... ........96

11. Number of flowering melliferous and polliniferous species and their relation with the
beekeeping annual cycle at La Montafia, Campeche, Mexico...........................97

12. Flowering phenology and average monthly precipitation at La Montafia, Campeche,
Mexico. A. Flowering phenology of melliferous and polliniferous species
according to data obtained at La Montafia from November 1998 to June 2000.









B. Mean monthly precipitation at the meteorological station of Chunchintok
(data from 1984 to 1999) ................................. ......................................99

13. Percentage of melliferous dominance partitioned according to the important periods
identified for the beekeeping annual production cycle: High water-content
honey period, Honey Harvest Period, and Dearth and Recovery periods, at La
M ontafia, Campeche................ ........ ............. ..... ....... 131

14. Locations of a sample of apiaries of 4 communities at La Montafia, Campeche,
M ex ico ........................................................................ 14 0

15. A sample of apiaries at Xonha in La Montafia, Campeche, Mexico.. ...........................142

16. A sample of apiaries at Xmaben in La Montafia, Campeche, Mexico...........................143















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

LANDSCAPE ECOLOGY OF APICULTURE IN THE MAYA
AREA OF LA MONTANA, CAMPECHE, MEXICO

By

Luciana Porter-Bolland

December 2001


Chair: Henry L. Gholz
Co-Chair: Mary L. Duryea
Major Department: School of Forest Resources and Conservation

Apiculture is one of the most important economic activities for rural communities

in the Yucatan Peninsula. In part, this results from the rich natural melliferous and

polliniferous flora found in the area and from the beekeeping tradition of the Maya

population. At La Montafia, Campeche, Mexico, beekeeping is currently the number one

income-generating activity. Honey production is currently carried out with Africanized

bees (Apis mellifera L.), for marketing to national and international markets. An analysis

of some economic aspects showed that profits are benefiting only a small sector of the

population that has better management skills and investment capacity. Better technology

and technical training, as well as the diversification of the activity could provide more

benefits to the region. Nevertheless, for this to happen external investment is needed,

particularly to increase the organizational, administrative and technical skills of

beekeepers.









The landscape at La Montafia greatly benefits apiculture as the different

vegetation types are rich in nectar- and pollen-yielding species. More than 100 species

from at least 67 genera and 31 families of trees, shrubs, vines, woody-vines and annuals

were found to be important for honey production. These species are present in the

different vegetation types found in the region, including lowland and upland forests,

fallow areas, recently disturbed sites, and savanna communities.

The "plant-pollinator landscape" where honey bees forage, is a function of the

bees' response to spatial and temporal variation in floral resources (Bronstein 1995).

Floral resources are distributed in the landscape both in space and time, according to the

distribution and abundance of melliferous and polliniferous species, and to their

flowering periods. The patchy nature of the landscape at La Montafia offers diverse

resources for bees to forage in, favoring apiculture.

An analysis of a 4-year period of Landsat Thematic Mapper remotely-sensed

imagery showed that forest conversion may be intensifying at La Montafia, as cattle

ranching becomes a management goal for people in the area; communication at the

regional, national and international levels increases; and the population grows. As a

buffer zone for the largest reserve in Mexico, the Calakmul Biosphere Reserve,

management practices for La Montafia that favor forest conservation are of the utmost

importance. Honey production at La Montafia and similar areas represents a promising

activity that can favor both natural resources conservation and socio-economic

development. Nevertheless, for this to occur, there needs to be a directed effort through

the implementation of specific management practices. Alternative markets may also play

a crucial incentive role.














CHAPTER 1
INTRODUCTION

Beekeeping in the Yucatan Peninsula has been a traditional activity since ancient

times. Maya people are known to have practiced intensive management of stingless bee

species, including Melipona beecheii Benneth (Chemas and Rico-Gray 1991). Although

meliponiculture (beekeeping with native stingless bees of the subfamily Meliponinae) is

now seldom practiced, beekeeping with Apis bees has become an important activity in the

region. Beekeeping is now carried out with modem techniques (movable frames) and

with bees of the species Apis mellifeara L. Over 40 % of Mexico's honey production

comes from this region, contributing to Mexico's importance as one of the world's

leading producers and exporters of honey (Angulo-Carrera 1992, Crane 1990,

Echazarreta et al. 1997, Guzman-Novoa 1996). Although comprising only 8% of the

Mexican territory, the Yucatan Peninsula contributes nearly 40% of the national honey

production (Echazarreta et al. 1997). Given that nearly 80% of the beehives are owned

by small farmers, beekeeping constitutes an important activity for rural communities

(Guzman-Novoa 1996).

The landscape in the Yucatan Peninsula is known to be rich in nectar

(melliferous) and pollen polliniferouss) yielding species (Rico-Gray et al. 1991). This

characteristic favors beekeeping, as nectar and pollen represent the major food sources

for bees. Floral resources are distributed in the landscape both in space and in time,

according to the distribution and abundance of melliferous and polliniferous species, and

to their flowering periods. Foraging bees are favored by a diverse landscape where









patches of different types of vegetation offer flowering sources through time and space

(Chemas and Rico-Gray 1991). In areas where forests are still a major component of the

landscape, beekeeping has been recognized as having great potential for socio-economic

development while preserving the natural forests (Acopa and Boege 1998, Galletti 2000).

The objective of this research is to analyze the social and ecological implications

of beekeeping at the landscape level of the La Montafia area, in Campeche, Mexico. This

region is found north of the Calakmul Biosphere Reserve, and is a buffer zone for this

protected region. At La Montafia, beekeeping represents one of the most important

economic activities (Galletti 2000, Porter 1995). Through analyzing the social and

ecological implications of beekeeping, it will be possible to assess beekeeping's status as

a management alternative that could favor the conservation of forested areas, and at the

same time increase the well being of its population.

In the tropical areas of Mexico, as in other parts of the world, the need to

implement alternative activities that may favor forest conservation is of outmost

importance given the fact that forest conversion is rapidly occurring in the face of modem

human activity. According to an FAO report from 1993, Mexico ranks first worldwide in

deforestation, with an estimated annual rate of 1.9% for the tropical deciduous forests

(Caims et al. 1995). While the causes of deforestation in Mexico are multiple, forest

conversion to agriculture, particularly cattle ranching, represents the leading factor.

Many tropical forests have been shaped to different extents by human societies,

creating a mosaic of forest patches that have "recovered" from different disturbances

(Browder 1996). Actually, in most "protected" tropical forests, archeological, historical

and ecological evidence exists showing a high density of human population in the past









and sites of continuous occupation over many centuries; and also intensively managed

and constantly changing environments (G6mez-Pompa and Kaus 1992). In that way,

forest patches making up the tropical landscape are dynamic entities, where wildlife and

vegetation are controlled and influenced by groups of interacting biotic and abiotic

factors, of which historic and present land use patterns are of great importance (Lyon and

Horwich 1996). Understanding the role of current human activity in this landscape is

needed to understand how forests are changing over time and across space (McIntosh

1981) and is critically important for the management and conservation of tropical forests.

This research is divided into 3 parts. First, the status of the current landscape is

analyzed, considering its history and its current trends. This information is presented in

Chapter 2, and it sets the background of what the main economic activities are today and

what their implications may be for the future landscape structure.

The second part of the research is presented in Chapters 3 and 4 and provides an

understanding of the current status of apiculture in the region as well as some of its socio-

economic implications. In Chapter 3, a literature review is presented in order to address

different issues related to the honey industry in Mexico and specifically in the Yucatan

Peninsula. The historical context and the changes that have resulted from the

Africanization of Apis mellifera bees are discussed. Findings by different authors that

have addressed the Africanization impacts are provided to show limitations and

opportunities for beekeeping in different contexts. Chapter 4 includes an analysis of the

socio-economic implications of beekeeping at La Montafia derived from information

from local beekeepers through interviews and close collaboration for a period of two

honey production seasons. Issues such as the benefits that the local families and









communities are obtaining from this activity, as well as the strengths and limitations of

honey production and marketing are discussed.

Chapters 5 and 6 present the third part of the research and include issues related to

the landscape ecology of apiculture, specifically flowering resource availability for bees

through space and time and its relation to beekeeping. Chapter 5 deals with the time

dimension and presents the results obtained from a community-level flowering phenology

study to assess resource availability for bees and its relation to the beekeeping annual

cycle. This chapter lists the most important melliferous and polliniferous species for this

area, according to beekeeper's information. Issues related to the implications for honey

production and marketing, and related to forest conservation are also addressed.

In Chapter 6 the space dimension is analyzed and an assessment of the

distribution and abundance of the melliferous and polliniferous species is presented. The

objective is to analyze how the landscape structure favors apiculture (particularly the role

of forested areas). The management given by beekeepers to vegetation for honey

production is also discussed in order to increase the understanding of the role of this

activity to the future structure and composition of forests.

Finally, in Chapter 7 some concluding remarks are made by integrating the

information derived from the previous chapters. An understanding of the ecological

implications of apiculture as well as the socio-economic characteristics involved in

beekeeping provides some insight into the role of beekeeping as a management

alternative that favors the conservation of this protected region.














CHAPTER 2
THE MAYA LANDSCAPE AT LA MONTANA: PAST AND FUTURE


Introduction

Forest systems in tropical areas are dynamic, and their structure and function are

molded by different processes acting at different scales (Waide and Lugo 1992).

According to Holling (1992) three different classes of processes operating in a

hierarchical manner determine structure and function in ecosystems. Factors at the

micro-scale of centimeters to tens of meters in space and days-to-decades in time, drive

the vegetative processes that determine plant growth, plant form, and soil structure. At

the other extreme, macro-scale factors drive the slow geomorphological processes that

dominate the formation of topographic and edaphic structures at scales of hundreds to

thousands of kilometers and centuries-to-millennia in time. In the middle, meso-scale

factors drive the formation of patterns over spatial scales of hundreds of meters to

hundreds of kilometers and on time scales of years to decades. Meso-scale factors

include natural disturbances such as fire, insect outbreaks, plant disease, and human

activities, that interact with topography and regional climate to form the ecosystem

specific patterns on the landscape, shaping the structure and composition of plant and

animal communities (Holling et al. 1996).

Meso-scale processes, therefore, lead to cycles of ecosystem development,

disruption, and renewal, configuring structure and pattern (Holling 1986). Forman

(1995) further differentiates between disturbance and stress. The latter refers to periodic









changes that become part of the ongoing background pattern of variation, while

disturbances are events that significantly alter the pattern of variation in the structure or

function of a system. Resilience is the ability of the system to maintain its structure and

patterns of behavior in the face of disturbance, which at the same time is molded by the

same pulses and variations of the system (Holling 1986).

There is increasing understanding of the role that disturbances (and stressors) play

in determining the structure and composition of tropical forests. Disturbances and

stressors differ in scales that vary in severity, spatial extent, frequency and duration

(Waide and Lugo 1992). While cycles of disturbance are integral to the maintenance and

generation of natural diversity (Asquith and Whittaker 1992, Goldammer 1992), tropical

forest ecosystems are also drastically being changed in the face of contemporary human

impacts and possibly climate change. Therefore, knowing the pattern, kind, frequency,

and magnitude of disturbances (or stressors) in the landscape is important for

understanding how forests are changing over time and across space (McIntosh 1981) and

is critically important for the management and conservation of tropical forests.

Most tropical forests have been inhabited by human societies for millennia and

have been submitted to numerous cycles of management and abandonment. Recent

palynological evidence suggests that such disturbances may have occurred for much

longer than previously believed (Binford et al. 1983, Flenley 1992). Many tropical

forests can be regarded as human social spaces that have been shaped to different extents

by human societies, creating a mosaic of forest patches that have "recovered" from

different disturbances (Browder 1996). In most "protected" tropical forests,

archeological, historical and ecological evidence exists showing a high density of human









population in the past and sites of continuous occupation over many centuries; and also

intensively managed and constantly changing environments (G6mez-Pompa and Kaus

1992). In that way, forest patches making up the tropical landscape are dynamic entities,

where wildlife and vegetation are controlled and influenced by groups of interacting

biotic and abiotic factors, of which historic and present land use patterns are of great

importance (Lyon and Horwich 1996).

The objective of this chapter is to assess the current status and trends of the

landscape at La Montafia, Campeche, Mexico, by interpreting remotely sensed digital

image data combined with household-level information. A change detection analysis of

two satellite images was used to assess forest clearing for agriculture from 1996 and

2000. The following questions were posed:

What are the activities affecting land use that are being carried out by the local
population?

What are the impacts in the landscape in terms of forest clearing?

And, how do current activities differ from traditional ones?

Through greater understanding of how the current landscape is being molded by

human activity, considering its history of land use, it should be more possible to

understand how current activities will affect future composition. Since La Montafia

represents a "buffer zone" for one of the largest reserve areas in the neotropics, the

Calakmul Biosphere Reserve, this has important implications for the search for

management alternatives compatible with larger regional conservation goals.

Study Area

The study area is north of the Calakmul Biosphere Reserve, in the center of the

Yucatan Peninsula, Mexico. The area is known as La Montafia and is part of the region









named Los Chenes in the state of Campeche (Figure 1). The study took place in the

eastern portion of this region and included 8 of the 11 ejidos1 that constitute La Montafia:

Xmejia, Ukum, Xmaben, Chan-chen, Chun-ek, Pachuitz, Nuevo Chan-yax-che, and

Xkanha; it also includes the ejido of Belha belonging to the Municipality of Calakmul.

The northeastern portion of the area is located at 19014'45" latitude and 89010'10"

longitude, and the southwestern portion at 18059'48" and 89025'07", respectively. The

total land area is approximately 200,000 ha, with a total population of less than 2,800

inhabitants, of which only 620 are ejidatarios (the persons holding the land title, which

in most cases include only the head of the family, most commonly adult males) (Galletti

2000).

La Montafia is considered a physiographic, economic and ethnic unit (Galletti

2000) with the population mostly Yucatec-Mayan. The land titles of these ejidos were

legally established between 1920 and 1940, when the Mexican public policy for land was

to grant forestlands to ejidal property (Galletti 2000). La Montafia is a forested area and

has been relatively isolated until recent times. The primary economic activities are

related to timber and nontimber forest products, together with subsistence agriculture.

North of La Montafia is the area known as La Milpa, a larger region constituting

different communities including the central Municipality, Hopelchen. This is the

economic and population center of the Los Chenes region. Topographic relief in this

northern part is less abrupt and there is greater population density, with more


1 An ejido is a tenure system in which the members forming the specific community have usufruct rights to
the land. Generally land is worked individually, although in some cases (e.g., pastures or forest lands)
rights are provided in common (Goldring, 1995). Ejidatarios (ejido members-not including their
descendents or other family members) constitute a corporate group, the ejido assembly, which is the highest
authority. Decisions on individual or collective use of land are made through this assembly. The State
grants permits and determines rules and regulations that must be followed. A 1993 amendment to Article
27 of the Constitution changed the ejido legislature and now ejido lands can be privatized.









agriculturally oriented communities. The population in this area is also mostly Yucatec-

Mayan, although there is a growing population of Mennonites.

South of La Montafia is the Calakmul Biosphere Reserve, created in 1989 and

registered in the UNESCO Man and the Biosphere Program in 1993. The reserve, the

largest tropical reserve in Mexico, totals 723,185 ha. It extends south into Guatemala,

abutting with the Maya Biosphere Reserve, making a 1.2 million hectare corridor2 of

permanent forest reserves, including the Sian Ka'an Biosphere Reserve at the

northeastern end in the state of Quintana Roo (Acopa and Boege 1998). Most of La

Montafia falls outside the Calakmul Biosphere Reserves boundaries, except for part of the

territory of Xkanha, and Belha, but is considered a "buffer zone" for the protected region.

The climate of the area has been determined, according to the Koppen

classification system as modified by Garcia (1973), to be a subhumid, warm tropical type,

having marked rainy and dry seasons (Flores and Espejel 1994). The mean annual

temperature is 26.2 C (1984-2000). The mean annual precipitation is 1223 mm, with

high variation from year to year, with a low of 648 mm (1986), and a high of 1807 mm

(1999) (CNA, unpublished data). The rainy season lasts from summer (May or June to

August) into fall (September to October).

Semi-evergreen and semi-deciduous upland forests cover most of the landscape in

the area, although substantial portions are also covered with lowland forests. Lowland

forests that are inundated for several months of the year, referred to as akalche', are

found in poorly drained sites and are sometimes large, particularly in the northeastern



2 This proposed corridor is now being formally established through the PNUMA (Programa de las Naciones
Unidas para el Medio Ambiente); La Montafia has been proposed to be the Focal Area #2 with high priority
for conservation (Galletti, personal communication).









portion of La Montafia. Deciduous lowland forests also occur in the area. The mean tree

height in the upland forests ranges between 25 and 35 m During the dry season, 25 to

50% (semi-evergreen) or 50 to 75% (semi-deciduous) of the tree species shed their

leaves. Deciduous lowland forests have a mean tree height that ranges from 6 to 15 m,

and most trees shed their leaves during the dry season. Akalche' forests have a mean

tree height of 7 m and about 50% of the trees shed their leaves during the dry season

(Flores and Espejel 1994). Patches of secondary vegetation in different successional

stages as well as patches of agricultural areas are found throughout the area. Hydrophilic

and savanna type vegetations are also found (Ucan et al. 1999). These vegetation types

comprise mostly Gramineae and Cyperaceae with differing associations of shrubby and

arboreal species. The distribution of these latter plant communities depends in part on

the levels of inundation and periods of desiccation, which vary highly (Flores and Espejel

1994). Fire is an important disturbance, particularly in savannas (Pennington and

Sarukhan 1998), as was reported to the author by people in the area who annually use fire

to enhance sprout growth for cattle grazing, to attract deer for hunting, and to favor

flowering of the melliferous tree Crecentia cujete.

Soils in the area were formed mostly by the dissolution of calcareous rocks and

over time have resulted in a karst topography (Flores and Espejel 1994). The topography

is typically undulating with few outstanding characteristics (Gates 1992). Drainage is

subterraneous with no superficial flows of water, except in the rainy season during storms

and periods of inundation, when rain can be very intense and water moves in temporary

surface flow channels. Lowlands are water-nourished by filtration of Karst during the

humid season and become dry to varying degrees depending on recharge through
















STATE OF CAM PECI IE


SCA[LAKMUI. BIOSPHERE RESERVE

.l iP.i''IPA.J ITY OF HCU I CHI-r,

O LA MONTANA REGION







,s i I
1 :. \ '



(^ '


1,'


21


C
C.


La Mortana


,-Ir .'. ;i i .


GUATEMALA


Figure 1. La Montafia area at Hopelchen, Campeche, Mexico


,Ir^;!nir l.









the dry season (Gates 1999). Some of the lowlands in the area have permanently humid

soils and there are permanent or semi -permanent water bodies at some sites.

Water is the most limiting factor for the people in the area. In the dry season,

people rely primarily on aguadas, small water bodies that fill up during the rainy season.

Aguadas are formed by depressions originating from the collapse of part of the terrestrial

crust, as a consequence of the existence of subterranean currents that dissolved

underlying calcareous rocks. These depressions are then filled by sediments with fine

textures that hold water and allow only slow drainage to occur (Morales 1992).

Flores and Espejel (1994) point out that the Maya, based on great knowledge,

have developed soil and vegetation classification systems that are still in use today. The

Mayan soil classification system is based on soil properties such as color, organic

content, drainage, ferrous oxides, and rock content (including the distinction of

anthropogenic rock remains), as well as associated vegetation (Table 1, Figure 2).

Landscape History

The structure and composition of the landscape at La Montafia has been molded

to a great extent by different factors, both human-induced and natural occurring, that

have played important roles in this forest ecosystem over space and time. According to

Konrad (1999), hurricanes in the area are recurrent and can be considered a constant

factor affecting the dynamics of succession as well as the stability of human settlement

since pre-Hispanic times. Between 1871 and 1990, nearly 14 major cyclonic events

affected the area (Konrad 1999). According to Flores and Espejel (1994), the return time

of hurricanes considered "dangerous" in the region is from 8 to 15 years. People in the

area commonly report the devastation of Hurricane Janet in 1955, which left large

portions of the forests resembling "land recently felled for agriculture." In their most









extreme cases, hurricanes create large natural clearings (Snook 1996), but these forceful

storms also result in the extensive inundation of lowlands and cultivated areas, as well as

defoliation and consequent forest fires, which can sometimes be very extensive (Konrad

1999).

Human intervention and management has shaped the landscape as well. Human

activity has a long history with periods of increasing and decreasing populations with

differing management goals. The four periods identified and described below are the

Maya period, the colonial and post-colonial period, the modern period, and the current

situation.

Maya Period

Vestiges of occupation in the Maya lowlands date to at least 1,000 BC, flourishing during

its Classic period at approximately A.D. 250 to 850 (Deevey et al. 1979). This was the

time when population densities were highest, reaching numbers greater than 100 per km2

(G6mez-Pompa and Kaus 1992, Turner et al. 2001). Although much of the population

was concentrated in urban centers such as Calakmul (located 190 km South of La

Montafia), known to have been inhabited since the pre-Classic period and becoming the

largest city of the Maya area during the late Classic times (Folan 1999), there is

archaeological evidence of occupation throughout the entire region. Although not much

archaeological exploration has been carried out at La Montafia, local people report the

existence of archaeological remains throughout the area. Folan (1999) has described the

northern part of the Rio Bec region and the Los Chenes area as one politically and

economically organized in a decentralized manner, with many different governing groups

distributed throughout the region.










Table 1. Major soil/vegetation types identified by local people at La Montafia, Campeche.
Their association to the FAO soil classification system is provided.
Mayan name Characteristics Vegetation type Importance to
agriculture


Yellowish soils, limy and deep,
related to chromic vertisols and
gleysols (1). Lowlands that
inundate at certain periods where
debris accumulates (2). People in
the area distinguish different types
of ak'alche' forests according to
characteristics such as the soil
type and the time it remains
inundated.)


Ak'alche'


These soils are not
good for agriculture
although there is a
recent trend to grow
one year of milpa and
after harvest establish
pasture for cattle*.


Low stature forests are present,
including perennial and
deciduous. Aquatic vegetation
and savanna type communities
are also present (2 Some of the
communities identified are:
Suut* forests. This name is
given by local people to
describe lowlands sites that
inundate for long periods of
time and that often are found
bordering savannas. Kat-luk*
are ak'alche' forests where soil
becomes very sticky when wet.
Chichi'tok* are ak'alche' forests
that do not inundate very often
(dry ak'alche'). People also
distinguish a different type of
savanna community, named
Xoch-chakan*, with a greater
abundance of Haematoxylon
campechianum. Chak'an*
vegetation communities. These
are savannas established in
ak'alche' terrain with a
distinctive vegetation
community mainly of grasses
from the Graminaceae and
Ciperacea, as well as dispersed
low stature trees (no greater
than 4 or 5 m in height) (2).
These communities have been
related to anthropogenic
activities and depend on
continuous disturbances such as
fire (3). Local people report that
these areas are intentionally
burned annually to encourage
grass sprout in order to feed
cattle, attract wild animals for
hunting, and favor the flowering
of Cresentia cujete, an
important melliferous tree.











Table 1 Continued
Mayan name Characteristics Vegetation type Importance to
I _agriculture


Ya'axhom


Dark-brown and dark reddish
brown soils, limy and deep,
related to chromic luvisols (2).
Flat lands with greenish dark soil,
with mud that absorbs water with
great quantities of vegetative
organic matter. They have
limestone that water currents
bring and that mixed with salt
obtain a greenish color (1).

Stony dark or reddish soils, less
than 30 cm deep. Moderate to
rapid drainage and easily eroded.
Lithosols and lithic rendzinas are
associated (1). Tsek'el s oils
contain big stones or sheets of
limestone with little organic soil
content.
Flat lands made up of red soil that
when dry brake up, becoming
hard and compacted. These red, or
brownish-yellowish soils (from
redzinas) have a silt texture, and
are usually deep (> 80 cm.),
having a rapid internal drainage,
and a superficial drainage from
moderately slow to slow (1)

These are places characterized by
having anthropogenic remains of
past times. They contain a deep
layer of vegetative soil found over
a limestone sheet usually
fertilized by part of the lime from
archeological sites, becoming
very fertile soils (2)


High, medium and low stature
forests (subperennial and
perennial) are found (1).








Medium and low-stature forests
are present (2).






High and medium stature forests
are present (1).








Medium and high stature
forests. Also called ramonal by
local people because of the
abundance of the ramon tree
(Brosimum alicastrum). Other
trees common to these areas are:
guava (Talisia olivaeformis),
beek (Ehretia ;1, ,I'lia 3 cocoyol
(Acromia mexicana) and ceiba
(Ceiba pentandra). These trees
are some of the most important
trees for the Maya culture, past
and present*.


(1)Arellano, Rodriguez, and Uuh 1992.
(2)Flores and Espejel 1994.
(3)Penningrong and Sarukhan 1998.
*Personal communication from local informants at La Montafia, Campech


This type of soil is
very fertile and often
cleared for milpa
establishment*.







Not good for
agriculture*.






This is the favored
soils type for
mechanized
agriculture*.






This is the favorite
soil type for milpa
establishment
although it is not
often cleared for this
purpose since these
are areas containing
many useful trees that
serve as fodder, fruit,
hunting areas, to
name a few*.


Tsek'el







K'ankab









Kakab or
Ramonal














Kakab Tsek'el
K'ankab




Seasonal water line


Figure 2. Soil/vegetation types identified by local people at La Montafia, Campeche and
their relation to topography. See Table 1.



Studies of ancient Maya production systems indicate that the Maya managed and

used a rich and diverse mosaic of ecosystems. A list of the different practices

hypothesized to have been used by ancient Maya, though not exhaustive, includes

terraces, raised fields and other hydraulic systems, sacred groves, door-yard gardens,

other silvicultural systems such as the tolche' and pet kot (see below), and the shifting

agricultural system called milpa (Barrrera et al. 1977, Caballero 1992, G6mez-Pompa

1991, G6mez-Pompa and Bainbridge 1995, Marcus 1982, Puleston 1982, Remmers and

Koeijer 1992, Turner and Harrison 1981).

The use by ancient Maya of terraces and intensive hydraulic systems such as

channels and raised fields has been observed in a large number of depressions in the

Maya lowlands (Adams et al. 1981, Matheny 1978). The high population densities

during Classic times are believed to have been sustained as a result of the implementation

of these types of intensive systems (Fedick and Ford 1990, Turner and Harrison 1981).

Although the abrupt collapse of the Maya civilization (at approximately A.D. 900) is not









well understood, it is believed to be due to a combination of different factors, including

environmental strain, economic and political crisis, and climate change (Deevey et al.

1979, Hodell et al. 1995). This period of high occupation and intensive land use left a

dramatic imprint on the landscape, as evidenced in pollen data that shows that forests

were greatly disturbed (Deevey et al. 1979). However, there is also evidence that Maya

production systems relied greatly on forest management systems, with a high degree of

biological diversity. This helps to explain both why the flora and fauna of the area are

still rich in species, and why the conservation of some natural ecosystems over their long

span of use was possible (G6mez-Pompa and Bainbridge 1995).

After the Maya collapse, lower population densities and the absence of intensive

agricultural systems gave way to forest recovery (Binford et al. 1987, Turner et al. 2001).

Turner (1990) estimated population changes during the two centuries between A.D. 800

to 1000, to have decreased from 2.6 or 3.4 million to around 1 million people,

representing a 53 to 65 % decrease.

Maya agrosilvicultural techniques persisted after the Maya collapse (Marcus

1982). Barrera et al. (1977) have argued that the present floristic composition of forests

in the Maya area is a result of ancient silvicultural practices, where human selection

processes have interacted with the regeneration of the natural system. Evidence for this

is based on current traditional management practices and by the dominance in the mature

vegetation of tree species that were and still are the same species protected, spared, or

planted in the land cleared for crops in the practice of shifting agriculture (G6mez-Pompa

and Kaus 1992).









The milpa is the traditional shifting agricultural method of a mixed agroforestry

system where corn, beans, and squash are the main products. In this practice,

uncontrollable fires are rarely caused but rather individual trees are saved from being

burned for the purpose of attracting wild animals, for fruit or wood, or simply for shade

(Barrera et al. 1977). The traditional milpa lasts for a period of 3 to 4 years, after which

the land is left fallow. The Maya had extensive knowledge about the regeneration

processes of vegetation. Different stages were identified according to the age of the

fallow and were used and managed in different ways (G6mez-Pompa 1987).

Dooryard gardens or homegardens were managed in many ways to provide

different goods and services: food, medicine, fodder, wood for construction, fuel, areas

for beekeeping, recreational spaces, etc. They are regarded as centers for plant and

animal experimentation. Dooryard gardens are managed through intensive silvicultural

practices and are structurally very diverse (Remmers and Koeijer 1992).

The term pet kot means "enclosed vegetation in a wall of stones" and it is a

reserved area in the forest, enclosed by stonewalls where numerous fruit trees and other

edible or useful plants are concentrated (Remmers and Koeijer 1992). These places are

chosen for the protection and cultivation of useful trees. They resemble forests in

composition and structure but have been enriched and manipulated, either by favoring

wild species coming from local forests, or by introducing different species from distant

and more humid forests of the Maya region (G6mez-Pompa 1987). The tolche' is a 10-

20 m wide strip of forest, usually created in the practice of making the milpa. These

areas serve as boundary markers, providing protected status to corridors between forest

patches. The management of the vegetation as a whole, not individually, characterizes









them, and their purpose is to provide multiple uses such as sheltered routes for moving

between fields, firebreaks, sources of different plant and animal species, and areas of bee

forage (Remmers and Koeijer 1992). Another practice that suggests that the Maya played

an important role in structuring the vegetation is that of protecting natural areas through

"sacred groves" (G6mez-Pompa 1987).

Colonial and post-Colonial period

La Montaia was still influenced by traditional land use systems. Nevertheless,

external factors and land use changes occurred at different points in time as a result of

several factors.

European contact and its resulting epidemics during the colonial period brought

about yet another decrease of the population throughout the Maya area. This

depopulation is estimated to have been greater than the depopulation after the Maya

collapse. In a period of 100 years (during the 16th century) population decrease is

estimated to have been as high as 75 to 95 % (Crosby 1972 as cited by Faust 1998). The

area of La Montafia, as well as much of the area that today makes up the Calakmul

Biosphere Reserve, remained somewhat in isolation from European contact until recently.

At the same time, though, it has still been an area where external interests have managed

to constantly exploit natural resources (Konrad 1999).

During the Colonial period, the area of La Montafia was characterized by an

indigenous, low density, nomadic population that appeared and disappeared, depending

on the milpa's cycle. These so-called independent Mayas, known as huites, had little

contact with the "white" population and maintained a traditional subsistence system

based on the cultivation of corn and beans and on trade. Trade occurred to the north with

population centers that had been established by Spanish friars --e.g., Hopelchen, which









was established in 1621 and by 1630 had grown to over 20,000 native people-- and to the

south with English settlers. Some of the products which were traded included honey,

wax, copal, cotton, tobacco, medicinal and aromatic plants, fibers, wood, live birds and

bird products, meat and skin from deer and other animals, and spices such as pepper and

annatto. In exchange, the Mayas obtained salt, rope, tools, and gunpowder (Ramayo-

Lanz 1996).

After Mexican independence (1821), the Maya population throughout the

Peninsula suffered the consequences of the new republic and its modernizing project.

With the objective of integrating the milpa areas into modem agriculture, post-colonial

agrarian policy formalized the dispossession of lands from native communities (Ramayo-

Lanz 1996). The forests of the Yucatan Peninsula suffered from the increase of farming

and ranching, where products such as rice and sugar cane were cultivated. The tribute

system, mainly in the form of payments of cotton and beeswax, as well as forced labor,

was maintained after independence from Spain. The imparting of justice, as well as

clerical services, reflected a new government that remained profoundly racist (Ramayo-

Lanz 1996). Ancient hierarchical forms of power, which the Mayas had been able to

sustain throughout the colonial period, also fell to the new government's social and

political reforms. According to Ramayo-Lanz (1996), these factors, together with false

promises made to the Mayas by different political groups of the then-forming government

("centralistas" and "federalistas") in order to recruit them into their militias, led to the

breakout of the so-called War of the Castes. This long and violent period lasted over 50

years (from 1847 to the beginning of the 20th century) and was decisive in the history of

Mexico in many aspects; it also strongly affected the future of La Montafia. This area









became the settlement lands of the "pacificos," a group of Maya willing to negotiate with

the government, as opposed to the more radical group, the "rebeldes." Manipulated by

the government, this division among the Mayas ultimately reduced their strength and led

to their eventual defeat (Ramayo-Lanz 1996).

The Maya population known as "pacificos del sur" (coming from other parts of

the Peninsula) congregated into settlements at La Montafia. The population in this area

fluctuated throughout the second half of the 19th century, according to different events

and circumstances. At that time, the valleys of the northern-most part of Los Chenes had

already become prosperous agricultural areas that produced export goods, such as sugar

cane, corn, rice, cotton and tobacco. This area was at the highest point of its economy,

and was generating pressure to the south. Many Mayas fleeing from slavery increased

population numbers in the southern hills of La Montafia, an area that maintained a certain

autonomy (Ramayo-Lanz 1996). By 1860, the area of La Montafia was considered a

municipality and had a population estimated at 14,400 (Dumond 1997).

By the second half of the 19th century, Campeche had become an international

economic enclave. With its "proyecto agroexportador," sugar-cane became an important

industry that competed with that of Cuba (Ramayo-Lanz 1996). Palo de tinte or

logwood (Haematoxylon campechianum L.) had already been an important export

product since the end of the previous century, particularly for the English market. An

important ingredient used for fixing dyes, logwood from Campeche was exported in vast

quantities to supply England's flourishing textile industry (Ferre-D'amare 1997). La

Montafia formed part of the commercial corridor that connected the southwest to the









southeastern English settlements through a network of paths where logwood, mahogany

and other forest goods found their way to the old world (Ramayo-Lanz 1996)1.

Nevertheless, the Caste War had its effect on a growing economy, which was

based to a large extent on slavery. Los Chenes was severely affected by the war, and its

"proyecto agroexportador" failed because of a reduction in the labor force (Ramayo-Lanz

1996). At La Montafia, different confrontations both with the government and with the

Maya "rebeldes" resulted in a decrease of the population as well. Whole towns were

abandoned. By the end of the 1860s there were only two population centers of

"pacificos" left: Xkanha and Ikaiche, and estimates are that the population had been

reduced to 12,000 people. By 1900 estimates are that the population had been further

reduced to around 8,000 individuals and the only remaining population center was

Xkanha (Dumond 1997).

The agricultural project of Los Chenes was further affected by a devastating

locust outbreak that occurred at the end of the 19th century. At La Montafia, hunger and

disease greatly diminished the remaining population, and by the beginning of the 20th

century, the government declared the forests of La Montafia to be "despoblados, or

uninhabited areas2 (Ramayo-Lanz 1996).

A new period began for the forests of Campeche, characterized by the

exploitation of resources through foreign hands and a few local businessmen (since many

had already gone north to invest in the henequen project of Yucatan). At the end of the

19th century, the government of Porfirio Diaz had already set the stage for forest



1 The importance of "palo de tinte" in the textile industry ended after synthetic dyes were invented by
William Henry Perking at the end of the 1850s (Ferre-D'amare 1997).
2 The population remaining at that time is not estimated, but it stopped being considered economically
important.









exploitation. Communication networks were built and forest concessions were granted to

foreign companies, particularly from North America3 (Ramayo-Llanz 1998).

La Montafia was left in isolation for most of the 20th century. During the first half

of the 1900s the area was characterized by the exploitation of the chicle tree (Manilkara

zapota L. Van Royen). The forests of Campeche and Quintana Roo produced

approximately 80% of the world's supply of the chicle resin used for making chewing

gum, an important export product, particularly during the period of the two World Wars4

(Acopa and Boege 1998). The product was brought from the forests along a network of

airplane landing strips, and chicleros (those who tapped the chicle trees) were imported

seasonally to work in the area. According to Konrad (1999) between 1901 and 1910, 10

million km of chicle were exported from Campeche, representing 50% of all national

exports. The benefits accrued during the early chicle period remained in the hands of the

foreign companies and contractors, with little profit left for local people in the area.

Nonetheless, the presence of chicle camps, together with water availability, guided the

re-establishment of communities at La Montafia (Ramayo-Lanz 1996).

During the 1920s, after the Mexican revolution (1910-1917), the population at La

Montafia was granted ejidal property rights. During the 1930s and 1940s, large forest

tracts were provided to the communities (Galletti 2000). These forestry ejidos responded

to chicle production needs and their large tracks were intended to guarantee good

production rates per person. At times population density was as low as 400 ha per person



3 In 1886 the Mexican government signed a contract in which several North American companies,
including the Pennsylvania Campeche Land and Lumber Company, the Laguna Corporation, the Mexican
Gulf Land and Lumber Company, the Campeche Timber and Fruit Company, and the Campeche
Development Company were granted concession to approximately 10,000 km2 of Campeche's territory
(Ramayo Lanz 1996).
SThe U.S. Army provided chewing gum to its soldiers as part of their rations (Acopa and Boege 1998).









(Acopa and Boege 1998). After the 1930s, chicle production in the area was

commercialized through the Cooperativa de Los Chenes (and later and until 1995,

through the Banco Mexicano de Comercio Exterior -IMPEXNAL), regulated by the

government. After the chicle boom ended (shortly after World War II, when synthetic

gums replaced latex for making chewing gum), chicle trees continued to be tapped with

up-and-down market demand, but to a much lesser degree.

According to Galletti (2000), timber exploitation at La Montafia began during the

1960s when the Zoh Laguna Mill extended its activities to the north. Over the following

decades, the forests were mined, and the most valuable species, particularly mahogany

(Swietenia macrophyla King) and spanish cedar (Cedrela odorata L.) were substantially

reduced. This period of timber extraction also left little or no benefit to the area (instead

reducing its economic potential), as lumbering was carried out with neither management

plans nor the participation of the local population more than as seasonal workers. The

result of this period of forest exploitation was the impoverishment of forests, and the

creation of a network of roads with its future consequences for resource exploitation

(Galletti 2000).

Modern period

During the 1970s and 1980s government initiatives were promoted with the

objective of integrating the area into the national economy. At that time, the rural sector

in Campeche was subject to a rash of agricultural development projects that promoted

colonization, mechanized agriculture and cattle ranching that resulted in the deforestation

of at least half a million hectares of forests in Campeche (Gates 1999a, 1999b).

Although the area of La Montafia remained somewhat inaccessible until the

construction of the Dzibalchen-Xpujil road in 1995, this period set the stage for the









current situation in the area. Several agricultural projects were promoted, including

beekeeping with Apis bees, which was adopted by traditional beekeepers in the early

1970s. In the early 1980s, a large-scale mechanized rice production project was

established, which resulted in the deforestation of 1000 ha of lowland forests

(information obtained from local people). Although this project failed and was

abandoned shortly after it was established as a result of technical and organizational

problems (as occurred with many other such projects in different areas of Campeche), it

opened land for the introduction of cattle ranching, and brought expectations for

mechanized agriculture to people in the area (Gates 1999b). These changes towards more

agriculturally oriented production systems occurred slowly in the area, but greatly

accelerated after the construction of the Dzibalchen-Xpujil road.

Pressure from areas surrounding La Montafia has occurred as a result of changes

brought about by the "modern" period of the rural sector in Mexico. The Escarcega-

Chetumal road that was built in 1967 brought thousands of settlers from all over the

country to the area that today is the Calakmul Biosphere Reserve (Gates 1999b).

Population in the region of Xpujil is estimated to have quadrupled from 1980 to 1990

(Ericson 1997). Human settlements within the Calakmul Biosphere Reserve and

surrounding areas (an area that remained until the 1950s in greater isolation than La

Montafia) have been estimated to total approximately 24,000 people settled in 72

establishments (Sanchez 1999). This population is very heterogeneous, coming from 23

different Mexican states, and including some Mayan ethnic groups such as Yucatec, Chol

and Tzeltal (Acopa & Boege 1998). These farmers mainly farm for subsistence, practice

forest extraction, grow commercial plantations of chile serrano (Capsicum sp L.), and









work as migrant laborers in nearby cities such as Campeche and Merida (ECOMAT

1995). Cattle raising is a strong factor in some of these communities. Beekeeping and

chicle extraction are also carried out, mainly by the Maya populations. Other

"experiments" in conservation are taking place related to the promotion of alternative

land use systems after the establishment of the Calakmul Biosphere Reserve in 1993,

including the production of other non-timber forests products, such as allspice (Pimenta

dioica (L.) Merrill), the establishment of agroforestry systems based on fruit tree

plantations, the use of native legume species as green fertilizers, and the encouragement

of ecotourism as an alternative industry (Acopa & Boege 1998).

North of La Montafia, a group of Mennonite settlements has been increasing in

population and area. Mennonites first occupied lands among the territory of the old-

established Maya population. During recent decades this population has increased,

arriving from Canada and northern parts of Mexico. By 1995 Mennonites occupied a

total land area of 45,000 ha and their population increased to 1,100 from 490 in 1990

(Ericson 1997). There are indications that this population is still increasing. In 2000, a

group of Mennonites bought 5,000 ha from one of the communities at La Montafia,

despite some disagreement within the community. This represents a major concern

because the Mennonites pursue intensive agriculture, felling a large fraction of the

forests, utilizing heavy machinery, and applying large loads of agrochemicals (Ericson

1997).

Current situation

The current population of La Montafia is about 2,800 (Galletti 1999), with some

slight growth. In 1996, the ejido of Chan-yax-che' was created by a relatively small and

heterogeneous group of migrants (about 12 families of people from Tabasco and









Veracruz). Sporadic immigration into old ejidos also has been occurring, in some cases

one or two migrant families asking the ejidos assembly for permission to occupy land,

but in other cases, by having many outside families as much as double the ejido's

population (as was the case of Chan-chen) promoted by government intervention.

Nevertheless, recent migration to the area as a whole has resulted in only limited total

population growth, with densities still only 116 to 783 ha per ejidatario (Galletti 1999).

The economy of the area depends on the use of a wide array of ecosystems from

which people obtain different goods and services. The traditional milpa is the basis for

subsistence. Other types of agriculture, like various agroforestry combinations including

homegardens, are also common and important for meeting family needs. People in the

area obtain meat from animal husbandry in their homegardens and from hunting in the

forest. Use of timber and non-timber products is important as well, as people obtain

wood for fuel and construction, fruits, medicines and other goods. People conduct

beekeeping and chicle extraction for cash income and occasionally sell timber to outside

companies. Cattle raising also has been gaining importance in recent years (Porter 1995,

Galletti 2000).


Methods

Remotely-sensed digital image data were used for describing the current

landscape at La Montafia by means of a change detection analysis of two scenes taken at

different dates. The objective was to monitor human activity as evidenced from forest

clearing. The scenes used for this study were a Landsat 5 TM scene from February 5,

1996, and a Landsat 7 ETM scene from March 27, 2000 (WRS 2, Path 20, Row 47).

According to Sader et al. (1994), who conducted an analysis of forest change estimates in









a similar area (northern Peten region of Guatemala), a 4 year time period between

satellite data acquisition is a good one for monitoring forest-clearing patterns, because a

fallow milpa site of two years can be clearly distinguished from active agricultural areas.

The scenes were analyzed by performing separate unsupervised multispectral

classification for each, and then conducting a post-classification detection of changes.

The software used for processing the data and conducting analyses were Envi and Arc

View. Multispectral classification is the procedure of grouping image pixels into spectral

clusters. This is carried out by assigning pixels with similar spectral levels into specific

classes of known reference signatures (Sohn et al. 1999). The post-classification

comparison change detection is used to compare the two classified images of the different

dates on a pixel-by-pixel basis using a change detection matrix (Jensen 1996). This

procedure creates "from-to" change class information that is used to create a change

image map.

A conceptual diagram depicts the procedures for analysis that were followed

(Figure 3). After resizing the two scenes to an area that included the extent of La

Montafia, the 1996 scene was rectified to the already georeferenced Landsat 7, 2000

image, obtaining an RMS error of less than 0.5 pixels. An ISODATA classifier, one for

each scene, was run using all but the thermal band. The classifier generated narrowly

defined spectral clusters (the procedure was set to obtain a maximum of 100 clusters), 85

for the 1996 scene and 86 for the 2000 scene.

Spectral pattern matching using information from a classification conducted for the same

2000 scene but at a smaller scale at the University of Quintana Roo, and information

derived from field work in the area during January 1999 to May 2000, were used to









merge classes and create a pre-classification of the study area. An image of this scene

was taken to the field during June 2001 for verification. Participatory image analysis was

conducted with 5 local people, and land/cover use types were assigned to the pre-

classified image according to visual interpretation and to the identification of specific

sites. This step was followed by transect walks conducted with the same people for

observation, and at different sites spatial data was collected using a hand-held Garmin 12

Global Positioning System (GPS). A re-classification followed, using a total of 78 GPS

points as reference points. Once each cluster was assigned to one of the reference

classes, spectral clusters that belonged to the same land cover/use type were merged

together to create a classified 2000 image. This image was re-evaluated by the group of

participants in the field. The 1996 image was classified accordingly, using the

information generated through the 2000 image classification. The two classified images

were combined to conduct a post-classification comparison change detection. This

generated a change detection matrix that was used to create a change image map.

Surface Analysis

A surface analysis was conducted to address different factors that may influence

forest conversion. The factors analyzed were distance to roads, distance to settlement

areas and soil type. For each of these factors several buffers were created and

subsequently related to the change image map. For assessing disturbance as a factor of

distance from roads buffers were created at 1, 3 and 5 km intervals. For assessing

disturbance as a factor of distance from settlements buffers were created at 3, 5 and 8 km.

For assessing the influence of soils, disturbed areas were quantified within a 0.5and 1.0

km buffer from lowlands.














PRE-PROCESSING


MULTI-SPECTRAL
CLASSIFICATION
TrRL' .LL IHi. Il l.
AFTER MERGIMG CLASSES


CLASSIFICATION
I JSI NO SPECTRAL
SIGNATURES


FIELD DAT
COLLECTION

/ pac Mipap+pii + >

U m Iumuamsuull umtlummm Ill


CHANGE
DETECTION
MATRIX


CHANGE
IMAGE MAP


Figure 3. Conceptual diagram of procedures for conducting a post-classification comparison change-detection










Results

The classified land cover/use maps of 1996 and 2000 are shown in Figure 4. The

2000 image was classified into 7 types, including: 1) upland forest, 2) lowland forest, 3)

advanced secondary succession/burned areas ("quemadal/acahual"), 4) secondary

succession/agriculture ("acahual/agriculture"), 5) agriculture/urban, 6) aquatic vegetation/

savanna, 7) and water. The classification of the 1996 scene included the same land

cover/use types, with an additional cloud mask class added. The latter was created after

performing a procedure to mask out clouds and their shadows; this was not done in 2000

since the scene was cloud-free.

Since the objective of the classification was to document forest-clearing, no

attempt was made to more finely classify different types of upland (e.g., sub-

perennifolius vs sub-deciduous), or lowland forests (e.g., akalche' vs deciduous

lowlands). Regarding the local people's classification system, upland forest includes

tzequel, kakab, kankab, and yash-hom; and lowland forest includes akalche' and suut

(Figure 2). Advanced secondary succession includes areas locally known as

quemadales, which are areas that have undergone fire (either natural or intentionally set)

and areas locally known as acahuales, which is a Spanish term used to refer to secondary

vegetation. Secondary vegetation/agriculture refers to early stages of recovery after an

intensive disturbance (e.g., agriculture). Since the spectral signatures of active

agricultural areas and urban settings were not differentiated, they were placed together as

a single class. Aquatic vegetation/ savannas, a class that includes areas with a distinctive

vegetation cover, can sometimes also be spectrally confused with quemadales or

acahuales.









The post-classification comparison yielded a change detection matrix (Table 2).

Since the accuracy of the change detection in this type of procedure is dependent on the

accuracy of the two separate classifications (Jensen 1996), several things should be taken

into consideration. For change detection using remotely sensed digital image data, the

date of data acquisition within the year is important, due to seasonal variations in the

spectral signatures (Brondizio et al. 1994).

Both images were obtained during the dry season. But the 1996 image was taken

at the beginning of February, when soil moisture was still high and trees had only begun

to shed their leaves, while the 2000 image was taken at the end of March, when the dry

season was advanced, soils dryer and the trees that shed their leaves had already done so.

Additionally, the quality of the 1996 image was not very good, presenting radiometric

error shown as striping. Although several filtering procedures were used to correct this

error, the condition did not markedly improve. Nevertheless, a principal components

analyses for the 1996 image showed that only 15% of the variation may have been the

result of radiometric error.












































Figure 4. Image classification of La Montafia portion of a) Landsat 5 TM image02/05/96; and b) Landsat 7 ETM 03/27/00









The 56 clusters generated in the post-classification comparison were merged to

produce a change image map (Figure 5). This map was merged into 11 classes using the

following criteria (the different colors in Table 1): 1) upland forest, 2) lowland forest, 3)

recovered vegetation (including vegetation that has recovered to either upland or lowland

forests) 4), new quemadal (including areas that have undergone a transition from lowland or

upland forests to quemadal, 5) same quemadal/acahual (including areas that have remained

as quemadal, and areas that are recovering from disturbance), 6) new agriculture/urban

(areas that have been cleared either from upland, lowland, or from quemadal, including new

urban areas), 7) Same agriculture/urban (areas that have remained under agriculture, acahual

or urban), 8) cloud mask; 9) aquatic vegetation/savanna (this includes the areas previously

identified as aquatic/savanna in the two images and areas that at either of the two dates were

covered with water, suggesting that they may be aquatic vegetation or savannas and whose

spectral signature when dry, is confused with acahual, quemadal, or agriculture (agriculture

is very unlikely to occur in areas inundated for long periods); 10) water; and 11) suut (these

areas were tentatively classified as suut -representing lowland forests that are inundated for

long periods of time, since they were classified at either date as forest and in the other one as

either water or aquatic vegetation. The three last classifications (aquatic vegetation/savanna,

suut, and water) are indicative of the complex hydrology of the area.

Approximately 35.5% of the area classified either as upland or lowland forest was

misclassified, as shown in Table 2 in the matrix boxes belonging to upland/lowland and

lowland/upland. This probably resulted from environmental variations given by differences

in the scene dates and from the bw quality of the 1996 scene. Also, a vegetation

sampling carried out in the area (Chapter 6) reviled that 90 species were common for both

forest types, out of a total of 115 for lowlands and 112 for uplands, which could result in










Table 2. Change detection matrix obtained by combining classifications made of the La
Montafa portion of Landsat 5 TM, 02/05/96 and Landsat 7 ETM 03/27/00 scenes.
Numbers are in hectares
20
Upland Lnd LowuandviaV Acahuall Agnciuirir.' Aquaicveg./
1996 Forlsi Foresi ac.ahjal agriculture irban savanna Water TOTAL

Upland Forest 61800.8 14849,9 2396.5 73 0 1364.9 96.2 336.8 81598.1
Upland Lowland rew qurind. l New Acahual New aagn Suut Suut
Loawand Forest 43029.1 43091.2 4344.1 1)61 6 3893.4 7' 4 1il 4 95654.1
pl___an LoMraln Ne' urra, dil NeA Aarhual Ne* Agri SIulu Suul _
UucITtilad'
S!F 1 ;015 2 660I
,,a qj.I l .,'2 I..'. t ,6- ;06B? 1.2 ; it' ;....
o_______Rcce'eid R.4:o-er.j SjrrF qucrriad Nei Acahuai Sarre aqrn ArjIiac Aquai ____
AyhulI'
agrilture 319.5 496.8 661 5 552.7 1272.5 5.3 2.9 3311.1
__Recmred Recodaeed Sainie qurriad Sare aaill Same an AJu3\1C Aqujaic ____

'.rtwri :,036 95 7 IC 10673 27 4 17 A G 050 4
Rec.er",d Recered 1i6. Sajn acal Sarre an Aui31Aqu ____A__
Aqualic veg/
savanna 168.5 7251 67-7 24.0 94,2 355.5 -P" 1564.6
Suu Suutr Aln A',ric A9uaiI, qual,: ,C A. _i

a,14 7; 15376 11 21 54C 277 3 6
Suul k alSuu A c Aqualtie Aquatic Aquaix Waler
Clouds 2076.4 557.3 1'2_ 41 55.6 129.6 4.56 .2 2934.1
Clouds Clouds Clouds C ouds ClOLIS Clouds Couds


TOTAL 109906.5 61471 1 12352.5 4240.0 11052.9 796.7 966.1 1955898


difficulties for differentiating spectral signatures. Other areas such as those classified as

quemadal or acahual, may also contain some errors. However, these errors do not affect our

interpretation of land use change since the areas classified as agriculture and urban are

unlikely to be in error, as their shape (e.g., squares, rectangles) and position in the landscape

(e.g., near roads, settlement) are other good indications of their land use/cover types. For

Figure 6 upland and lowland forests were classified according to the 2000 image.

















































*ee
S
0 20
J>' ,' .', !. f*f 'v .,' 'T- '>^


A Settlements
Sy' Rniad


SI Upland Forest
M Lowland Forest
SRecovered WVeg


New QuemadailAcahualI Water
-jrnc Qucmsrhi .[A|Ci.u.T M Clouds
SNew AgricultureiUrban
SI SBI AgTiculIt&Urtabn N
I Aquatic veg.ISavannas
SSuuts A


Figure 5. Change map created by combining classification data of the La Montafia
portion of Landsat 5 TM, 02/05/96, and Landsat 7 ETM 03/27/00 scenes.


--2'-










Four percent of the area was cleared for agriculture or developed as urban areas

during the four year period (Table 3). This includes areas that were in the early stages of

recovery (acahual). Additionally, 2.6% of the area remained as agriculture, urban, or

acahual. Together, areas in 2000 that were represented as active agriculture, urban

development, or early stages of recovery (acahual), added up to 6.6% of the total area.

Areas in advanced regeneration stages or that had undergone a fire (quemadal) represented

4.5% of the area, of which 3.5% were recently disturbed. Areas that had recovered to either

upland or lowland forests represented 2.5% of the area. Not considered in the classification

was 1.5% of the area masked out as clouds or shadows.


Table 3. Changes in pixel count, area, and percentage of total area of land use/ cover
types obtained from a post-classification analysis of Landsat TM, 02/05/96 and
Landsat 7 ETM, 03/27/01.

Percentage of total
Land uselcover type Pixel count Area (ha) area
Upland Forest 1290611 104829.9 53.6
Lowland Forest 713340 57941.04 29.6
Recovered vegetation 61173 496878 2.5
New quemadal 62987 6740.6 3-5
Same quemadal 252S2 2053.53 1
Same agriculture/early acahual 61628 343 2.6
New agnculture/early acahual 95778 7779.6 4
Aqualic vegelatonl savanna 16458 1336.8 0.7
Suut 20585 1672.02 0.8
Water 4035 327.7 0.2
Cloud mask 36123 2934.1 1.5

TOTAL 2408000 195589.8 100

Results from the surface analyses show that although proximity to roads and

settlements may be a factor in forest clearing, most of it is occurring as a function of soil

type. This corresponds with information obtained from local people who stated to choose

their agricultural sites mostly on the basis of environmental conditions (soil type and

vegetation cover), although accessibility to town was also considered important.









Buffers made at 1, 3 and 5 km from roads show that disturbance is not very related to

proximity to roads (Figure 6 A). Approximately 7.7%of total area falls inside a 1 km buffer

from major roads and only 1.7% of this area is disturbed, which corresponds to 16.7% of the

total disturbed area in the image. A 3 km buffer accounts for about 26% of total area of

which 3.9% is disturbed (38% of total disturbed area). A 5 km buffer from roads accounts

for about

40% of the total area of which only 5.4% corresponds to disturbed areas, accounting for only

52% of total disturbed area in the image. It should be taken into consideration that the roads

are relatively new and that people have been established in the area for at least a century.

Buffers from settlements were made at 3, 5 and 8 km (Figure 6 B). A 3 km buffer

from settlements accounts for almost 13% of total area of which only 2..5% is disturbed,

representing 24.3% of total disturbed area. A 5 km buffer from roads accounts for 28.5% of

total area of which 4% is disturbed, corresponding to almost 40% of total disturbed area. An

8 km buffer accounts for almost 50% of total area of which little over 6% is disturbed,

representing 60% of total disturbed area in the image. While proximity to settlements may be

a factor in establishing agricultural areas, people do not limit themselves to areas close to

their towns and almost 40% of total disturbed areas falls outside the 8 km limit from

settlements. People in the area report that very often they establish their agricultural sites in

areas where they have to camp, sometimes for over two weeks.























4.?+ P~:
*j

~ 17 v-t


A. Road Buffers at 1, 3 and 5 km
SDistubed areas
Lowland Forest A


Matrix


B. Settlement B uffers at 3, 5 and 8 km


C. Lowland Buffers at 0.5 and 1.0 km


Roads
Settle ments


Figure 6. La Montafia area, Campeche, Mexico. Buffers made at: A. 1, 3 and 5 km from roads; B. At 3, 5 and 8 km from roads; and
C. At 0.5 and 1.0 km from lowlands


u









Buffers from lowlands indicate that people largely choose their agricultural sites based on

soil type since up to 53% of the total disturbed area falls within a 0.5 km buffer from

lowlands, and up to 67% within a 1 km buffer (Figure 6 C). This corresponds with

information provided by local people that indicated the yash-hom and kakab soil/vegetation

types were preferred for milpa establishment (Figure 2, Table 1).


Discussion

Clearing Rate

If forest clearing at La Montafia is assumed to have occurred equally over the

four-year period, an estimated annual clearing rate of 1% and an annual regeneration rate

(also divided between the four year period) of 0.635% provides a net deforestation figure

of 0.365% per year. This figure considers only the recently disturbed areas.

Additionally, 2.6% of the area previously disturbed has not been allowed to recover,

probably remaining under continuous use. If the latter figure is partitioned into the four

year period, an additional 0.65% provides a clearing rate of a little over 1% per year.

Another 4.5% of the area has been disturbed as quemadal, of which most, 3.5%, has

occurred over the four-year period. Areas under quemadal either resulted from

wildfires, or were burned or felled by local people, indicating that they may be in a

transition to a more intensive use (when preparing agricultural areas from mature forest

people in the area usually fell trees at the beginning of the year, during February or so).

Quemadales may also be areas in an advanced stage of regeneration. They may also be

misclassified and could be areas of very dry forest that shed most of their leaves

(considering that the 2000 image was taken during the peak of the dry season).









Turner et al. (2001) reported annual deforestation rates caused by anthropogenic

disturbance in the region of the Calakmul Biosphere Reserve to be between 0.32 and 0.39

% for the years 1969 to 1997. Sader et al. (1994) reported annual rates of forest clearing

at the northern Peten region of Guatemala to have been 0.4 % at the end of the 1980s,

with a regeneration rate of only 0.03%. They consider the clearing rate as low, which

they attribute to the inaccessibility of the interior of the Maya Biosphere Reserve.

Nevertheless, in a second study, Sader et al. (1997) reported annual clearing rates at the

buffer zone of the Maya Biosphere Reserve (in the western and southern borders) to have

been as high as 3 % at some sites during the early 1990s. They attribute the higher forest

clearing rates to agricultural expansion and accessibility (e.g., roads, pipe lines, and

rivers).

According to an FAO report from 1993, Mexico ranks first worldwide in

deforestation, with an estimated annual rate of 1.3 % (Cairns et al. 1995). Considering

the different forest types, tropical deciduous forests are second in their rate of forest

clearing, after tropical evergreen forests, with an estimated figure of 1.9 % per year. The

causes of deforestation in Mexico are multiple, with forest conversion to agricultural

uses, particularly cattle ranching, the leading factor. Forest fires, hurricanes and

commercial logging are also considered important factors (Cairns et al. 1995).

Although the estimated 1% annual forest clearing rate at La Montafia is lower

than the average estimated figure of 1.9 % for deciduous tropical forest in other areas of

Mexico and Guatemala, it is still higher than the rate estimated for the interior of the

Calakmul and Maya Biosphere Reserves ( 0.4%) (Sader et al. 1994, Turner et al. 2001).

This suggests that forest clearing may be intensifying, although further research would be









needed to understand more about the regeneration process of these forest ecosystems, and

consequently more about their dynamics (e.g., through conducting a similar analysis

utilizing images taken at a previous date -particularly before the construction of the

Dzibalchen-Xpujil road).

Land Use Change at La Montafia

There are factors that indicate that land use may be intensifying and that

management objectives for the local populations are changing. The milpa is traditionally

established in upland forest preferably of the yash-hom, kakab, and kankab types

(Figure 2) (note that uch of the agricultural area in the images are shown bordering

lowlands). None of it is established in akalche' forests, since the probability of losing

the crop by inundation is high. Nevertheless, local people report that the recent trend is

to establish the milpa in akalche' areas and, after the first year's harvest, to plant grass.

This new trend changes the traditional slash-and-burn system, where forests are left to

regenerate, to a system where fallowed areas are replaced with grass for rearing cattle. It

can be observed in the images that a large portion of the newly cleared areas is found

within the lowland forest areas. Also the 2.6% of the area that showed continued use

over the four-year period suggests that forests are not being left to recover (this could be

further tested through analyzing images of more dates).

Cattle raising at La Montafia is an activity that has been gaining in importance in

recent years. Government programs have aided in its promotion, and even though most

people have only a few head of cattle, some individuals are investing much of their

capital and labor into increasing their herds and pasturelands. Cattle raising is currently

perceived as an activity important for accumulating capital and gaining social status, as

well as a good source for quick cash. In recent times, the government has provided









considerable credit to some individuals for establishing cattle ranches (Porter 1995).

There are also people who invest in pastures even though in some cases they still do not

own cattle (Galletti 2000).

Cattle raising at La Montafia has strong implications for forests in the region and

may be the reason for the forest clearing rate observed here. According to Masera et al.

(1997 citing FAO 1990) the deforestation process in Mexico is weakly correlated with

population growth and, although the key driving forces of deforestation are varied, direct

or indirect subsidies to cattle ranching have been one of the outstanding factors. Changes

in the traditional agricultural systems may be posing more rapid threats to the current

forest ecosystems at La Montafia than population growth.

However, people at La Montafia carry out other forestry related activities, which

are economically important and which favor (or could favor) the conservation of a

forested landscape. These activities include chicle production, timber extraction,

beekeeping, traditional agroforestry systems, and the traditional milpa.

Chicle production continues to be an important economic activity in the area.

This activity is favored and at the same time favors the conservation of forests, where

large chicle trees are found. Although restrained by market fluctuations and by

government permits, it is potentially important for the economy of the region.

Timber harvesting is limited by impoverished forests, where the most

commercially valuable species (mahogany and Spanish cedar) have been extensively

extracted. Nevertheless, timber harvesting still represents an important potential

economic activity in the region, which given the necessary conditions could favor forest

conservation. Species with commercial potential, such as granadillo (Platymiscum









yucatanum Standley), siricote (Cordia dodecandra A.DC.), guayacan (Tabebuia

guayacan), and chaka (Bursera simaruba (L.) Sarg.), are currently being sold from

forests at La Montafia. However, timber extraction is an activity limited by government

permits and markets. Currently, there is no organization or structure in which the local

population (or the forests) can benefit, and timber extraction is being carried out with no

formal management plans. Usually, wood is sold as standing volume to companies that

bring their trucks and equipment to extract the timber, many times illegally. Local people

mainly receive money for the labor of felling the tree, but with not much other gain, and

most of the time they are unaware of the volumes that are being extracted (Porter 1995).

Timber extraction carried out in this way is clearly detrimental to forest conservation.

Cairns et al. (1995) point out that the deforestation process in much of the Mexican

forests typically begins with timber extraction in this unorganized way. The

consequences are that the economic potential of forests is reduced after valuable timber

species are depleted and the associated road infrastructure encourages colonization and

agricultural expansion. However, timber extraction could become an incentive for forest

conservation, provided the necessary conditions are created.

Honey production is currently an important income generating activity in the

region. Although beekeeping is a traditional activity, beekeeping with Apis mellifera L.

is relatively new and has been developing into an important economic activity in recent

years, establishing market ties with national and international industries. In the region,

beekeeping with Apis bees has been favored by a mosaic-like landscape formed by

patches of different vegetation ages and types, rich in melliferous flora. The extent to

which this activity favors (or could favor) forest conservation is an important question









that should be addressed in the search for sustainable management alternatives for the

development of the region (see following chapters).

The traditional milpa as well as different agroforestry systems including home-

gardens and the parcela agroforestal', are important subsistence activities for people at

La Montafia. These activities are inherited traditional practices based on solid

environmental knowledge that result in a high degree of biological diversity. The slash-

and-bur milpa system as practiced traditionally can be regarded as an agroforestry

system where fallow and regeneration of the vegetation are part of the management

practices (Gomez-Pompa 1987, Sohn et al. 1999) and where multiple products and

services are obtained. While the milpa system itself does not guarantee forest

conservation or a sustainable use of natural resources, as populations grow, forests are

converted to other land uses and new technologies are adopted (e.g., use of chainsaws,

use of agrochemicals, seed sources, etc.). This must be balanced by the wealth of

knowledge that the Maya people have regarding vegetation, soil, hydrology, and climate

that should not be disregarded when management alternatives are sought for forest

conservation at La Montafia and similar areas in the tropics.


Conclusions

The forest landscape of La Montafia has been shaped by a range of factors, that

include natural forces such as hurricanes and fires and human-induced disturbances,

through the long span of history of occupation in the area. Human activity in the area of





1 A traditional agroforestry system similar to that of homegardens but which may or may not be established
next to the house, and that currently has been including the establishment of different native and non-native
fruit and timber trees encouraged by NGO.









La Montafia has been constant, with episodes of higher and lower occupation. Since

recovery of the forests from the high populations and intensive use during the Classic

Maya period, forest cover has been steady. Human activities in the area until recently

have been related to the use of timber and non-timber products, guided by internal and

external interests that have molded the current status of the forested landscape. The

traditional milpa agriculture, as well as other traditional agroforestry systems of the

Maya, developed into management practices where regeneration of forests was part of the

systems, becoming endogenous "stresses" to the forest ecosystems and very likely

affecting the structure and composition of the vegetation and animal communities

through time. Chicle extraction provided an incentive for the legal establishment of

"forestry" ejidos in the area. Timber extraction, on the other hand, left an impoverished

forest, depleted of the most valuable timber species.

Today, land use activities are changing towards more agricultural production,

with potentially large implications for the forested landscape of La Montafia. This

analysis of the four-year period of remotely-sensed imagery indicates that forest

conversion may be intensifying, as cattle raising becomes more common; communication

at the regional, national and international levels increases; and the population grows.

Although the economy in the region is still greatly dependent on the management

of forest resources, changes are occurring which are visible at the landscape level. The

question now is how will current decision making regarding land use affect the future

structure and composition of forests at La Montafia. The following chapters will analyze

beekeeping as an activity that could favor the conservation of a forested landscape in this

area.














CHAPTER 3
BEEKEEPING IN THE MAYA AREA: HISTORY AND CURRENT STATUS


Introduction

Beekeeping in the Yucatan Peninsula has been a traditional activity since ancient

times. Maya people are known to have practiced intensive management of stingless bee

species, including Melipona beecheii Benneth. In Mayan philosophy, humans and bees

were very related; the Zayawinnicoob, or first humans, were considered to be very

industrious people and were represented as having bee-like eyes (Chemas and Rico-Gray

1991).

Although meliponiculture (beekeeping with native stingless bees of the subfamily

Meliponinae) is now seldom practiced in the Maya area, beekeeping with Apis bees has

become an important activity in the region. Beekeeping is now carried out with modem

techniques (movable frames) and with bees of the species Apis mellifeara L. Over 40

percent of Mexico's honey production comes from this region, contributing to Mexico's

importance as one of the world's leading producers and exporters of honey (Angulo-

Carrera 1992, Crane 1990, Echazarreta et al. 1997, Guzman-Novoa 1996).

A literature review is presented to address issues related to the honey industry in

Mexico and specifically in the Yucatan Peninsula, considering its historical context and

the changes that have resulted from the Africanization of Apis mellifera bees. Findings

by different authors that have addressed the Africanization phenomena are provided in

order to illustrate limitations and opportunities for beekeeping in different contexts.









Beekeeping Industry in Context: Mexico's Place in World Honey Production and
Relevance of the Yucatan Peninsula.

The beekeeping industry provides society with many important products and

services. The traditional products, honey and beeswax, have been used since ancient

times in all parts of the world (Crane 1990). Honey, the final product of nectar collected

and modified by bees, is used as a sweetener, either served at the table, or in the food and

drink industries. It also has many uses in medicine and cosmetic manufacturing.

Beeswax, which is secreted by worker honey bees from wax glands, was more widely

used in the past, before synthetic products were substituted (e.g., paraffin). Nevertheless,

beeswax is still used as a component of cosmetics and pharmaceuticals, as well as for

other minor uses. According to Crane (1990), most of the beeswax that is produced

today is wasted, some of it is used in traditional ways, and a small part is used in the

industry. A big portion of beeswax is re-used in the beekeeping industry for producing

comb foundations.

Other products derived from beekeeping began to be produced after World War

II, when low prices for honey, replacement of beeswax by other substances for many

industrial purposes, new technologies, and a widespread new interest in health foods and

dietary supplements led beekeepers to diversification. These products include pollen,

propoli, royal jelly, bee venom, and bee brood (Crane 1990). The beekeeping industry is

also important for providing pollination services for agriculture, and has become an

integral part of production technology worth millions of dollars in industrialized

countries (Basualdo 1996). Currently, the pollination of fruit and seed crops by bees

(including bees other than honey bees) provides a greater economic benefit to agriculture









world-wide than the beekeeper's income from honey and other hive products (Crane

1990).

The basis of today's beekeeping industry is the honey bee Apis mellifera L. (order

Hymenoptera; family Apidae; subfamily Apinae), even in areas of the world where this

species is not native (such as the American continent). Honey is also produced in

different parts of the world by other Apis species and by different species of the

subfamily Meliponinae (the stingless bees), but mostly for local consumption or regional

trade (Crane 1990). Honey production from Apis mellifera is the most widespread and

has an established world trade market. Worldwide honey production from this species

reaches numbers that approximate one million metric tons per year (Angulo-Carrera

1992).

Mexico is one of the principal countries in the world-wide beekeeping industry.

In 1991, Mexico was the fourth honey producer, after the former USSR, China and the

United States (Sanchez-Vazquez and Colli-Ucan 1992). Data obtained from the Centro

de Estadistica Agropecuaria1 report an average annual honey production in Mexico of

60,000 metric tons of honey from 1980 to 1999, and an average annual production of

4,217 metric tons of beeswax for that same period. According to Angulo-Carrera (1992),

Mexico was the leading exporter for the period of 1974 to 1984. In 1995, Mexico was

the ninth honey producer, after China, USA, Argentina, the former USSR, and four other

Asian countries, which together accounted for 46% of that year's world production, and

the fourth largest exporter, after China, Canada and Argentina (Troncoso 2000). The use

of bees for pollination has also gained importance in some states of Mexico, mainly in the









northern states, and has been estimated to have a greater value than honey production and

beeswax for that area (Guzman-Novoa 1996).

The southeastern part of Mexico is one of the most important honey producing

regions. This honey producing area is comprised of the Yucatan Peninsula (Campeche,

Quintana Roo, and Yucatan). Representing only 8% of the national territory, this region

accounts for 30 to 40% of annual national honey production (Echazarreta et al. 1997).

The relevance of this area as a honey-producing region has been explained by Guzman-

Novoa (1996) to be the result of its "exuberant" native melliferous/polliniferous flora,

and because of the beekeeping tradition of the Maya people, who have been keeping bees

since ancient times.

In Mexico, honey production is very important for the rural population as an

alternative income activity and 80% of the beehives are owned by small farmers

(Guzman-Novoa 1996). Therefore, the honey industry represents an important source of

income and jobs for low-income families in the rural sector. Beekeeping has also been

considered an alternative land use in areas where forests are to be conserved and as a way

to provide added value to secondary vegetation and mature forests (Acopa and Boege

1998). Being a profitable non-timber forest product with an established nIrket,

beekeeping is considered to have great potential in protected regions such as the

Calakmul Biosphere Reserve and its surrounding areas (Acopa and Boege 1998, Galletti

2000).






1 The Centro de Estadistica Agropecuaria (Agricultural Statistical Center) is part of the Secretaria de
Agriculture, Ganaderia y Desarrollo Rural (SAGAR), the Mexican Secretariat of Agriculture and Rural
Development.









History of Beekeeping in the Maya Area.

Although there are no native honey bees (Apis) in the American continent,

Mesoamerican cultures managed stingless bees of the genera Melipona and Trigona

(order Hymenoptera; family Apidae; subfamily Meliponinae) for over 3,000 years

(Guzman-Novoa 1996). Of particular importance is the management given to the species

Melipona beecheii Bennett by the Mayas. Management of the xuna'an kab (xuna'an=

queen or princess, and kab=bee or honey), the Mayan name given to Melipona beecheii,

developed into an extensive bee culture, where honey and wax were used for trade,

religious ceremonies, and medicinal purposes (Sanford 1973). Bishop Diego de Landa

(who arrived in Yucatan in 1549) described the life and customs of Maya Indians,

including beekeeping and the different festivities and offerings dedicated to bee gods

(Weaver and Weaver 1981). The Troano Codex also refers to different aspects related to

beekeeping in the Yucatan Peninsula (Labougle and Zozaya 1986). Weaver and Weaver

(1981) describe some of the bee-related festivities dedicated to gods that were still

practiced until recently by current Mayas. Honey produced by the xuna'an kab is still

(in few cases) used for making balche', an alcoholic drink used in religious ceremonies.

Balche' is prepared using the bark of the balche' tree (Lonchocarpus longistylus Pittier)

and honey made by Melipona beecheii. Honey was paid as tax since pre-Hispanic times

and was the subject of an intense commerce that took place from Tabasco to different

areas of the Peninsula, by sea to Honduras and Nicaragua, and to the west with the

Mexica Empire. Mayas traded beeswax and honey probably for cacao seeds and precious

stones (Labougle and Zozaya 1986).

After the Spanish conquest, honey and wax produced from xuna'an kab were

paid as taxes to the Spaniards. During this time, beeswax was particularly important in









religious activities carried out by the Spaniards. Honey as a sweetener was gradually

replaced by sugarcane. Nevertheless, the beeswax produced from stingless bees supplied

a vast area of the Spanish Empire practically for the entire duration of the Colonal period

(Labougle and Zozaya 1986). Labougle and Zozaya (1986) reported taxes paid to

Spaniards by the Maya population in the yearl547; of 173 settlements in the state of

Yucatan, only 5.8% did not pay with honey or beeswax. Total production in this area

during that year approximated 29,300 kg of wax and 3,300 kg of honey; every 20 persons

contributed 12 kg of wax and every 295 persons, 12 kg of honey.

European honey bees (Apis mellifera) were not introduced into the Yucatan

Peninsula until the end of the 19th and beginning of the 20th centuries. Although there are

reports of their presence in the center of Mexico at the end of the 18th century, this did

not prove to be the introduction to the Yucatan Peninsula. Modem apiculture (carried out

with Apis mellifera bees and using movable frames) began to be practiced after the 1920s

(Labougle and Zozaya 1986). In the 1930s and 1940s, large amounts of Apis honey

started to be produced in the Yucatan Peninsula (Chemas and Rico-Gray 1991). During

the 1960s and 1970s beekeeping with Apis mellifera by the Maya population gained

importance, becoming one of the principal economic activities for the rural sector

(Echazarreta et al. 1997).

Beekeeping in the area is now carried out primarily with Apis bees using modem

techniques (Sanford 1973), although management of Meliponinae bees is still practiced

by a decreasing number of Maya people. The arrival of the Africanized bees (see below)

into Mexico in the last half of the 1980s has brought different challenges and

opportunities for beekeepers in the area.









Beekeeping with the Africanized Bee

The Africanization of the European races of Apis mellifera in the American

continent in the last fifty years has resulted in profound changes in the beekeeping

activity in a very vast area of tropical and subtropical America. These changes bring

different challenges and opportunities for beekeepers that will be briefly discussed in the

next section. These changes also raise important questions regarding the role of exotic

honey bees as pollinators in natural areas and the possible effects that may occur on the

native flora and fauna populations.

Africanized Bee

Africanized bee is the name given to races of European Apis mellifera bees that

have hybridized with an African race of Apis mellifera (Apis mellifera scutellata Ruttner)

in the American continent (Crane 1990, Daly 1991). The Africanization process began in

1956, when 26 absconding swarms of Apis mellifera scutellata escaped from an apiary in

Camaquan, Sdo Paulo, Brazil. These bees had been brought from Africa in order to

conduct breeding experiments. By 1991 Africanized bees were found in many different

habitat types in an area about 20 million km2 that extended from northern Argentina to

Mexico (Roubik 1991). In the last decade these bees also reached parts of the southern

United States and in all likelihood will continue to expand in North America.

The Africanized bee has been able to quickly occupy most of tropical and

subtropical America. This astonishing success is in part due to a highly evolved capacity

to exploit unpredictable and varied resource conditions and in part due to its unique

reproductive biology (Ratnieks 1991). It has been calculated that the "colonizing front"

of Africanized bees has moved at an average rate of 300 to 500 km/year (Crane 1990,

Otis 1991).









The African bee that was brought to Brazil was adapted to arid and semi -arid

environments ranging from tropical East Africa to subtropical South Africa. The

environmental conditions in this area are varied, but the major selection pressure for the

bees was likely the unpredictability of rainfall which influences flowering. On the other

hand, the European bee evolved in a much more predictable environment where annual

snow melt provided a secure source of moisture for flowering to occur (Rinderer and

Hellmich 1991). This adaptation to unpredictable conditions has provided a great

advantage to the Africanized bee when colonizing new areas.

A summary of the characteristics of the reproductive biology that have helped the

colonizing success of the Africanized bee (and that affects traditional beekeeping

schemes) is as follows (Otis 1991, Ratnieks 1991, Rinderer and Hellmich 1991):

* Swarming. This is the process by which bees reproduce (through colony fission).
European colonies usually swarm once a year, while Africanized bees can swarm
up to four times annually.

* Absconding. Africanized bees tend to abandon their nests or abscond.
Absconding occurs either as a result of disturbance to the colony, or as a result of
shortages in resources. The latter occurs in periods of extended dearth (when
there are no flowering resources or water available), or when other resource rich
areas have been identified. European bees rarely abscond; it is more common that
the colony perishes when faced with resource shortages or external threats.

* Differential reproductive rates. Africanized bees develop large populations in
short periods in response to the existence of harvestable resources (as a result of
their adaptation to unpredictable resource availability). They have higher
fecundity and shorter development times than European bees.

* Social reproductive parasitism This is a term that includes different behavioral
characteristics of African races of Apis mellifera. A form of social reproductive
parasitism is the capacity of reproductive females of this subspecies to take the
place of the queen of a different colony, accelerating Africanization. Another
form of social reproductive parasitism is that drone populations (drones are male
bees whose basic function in the colony is to mate queens) are larger in African
colonies. Therefore, during mating flights there is a greater probability that the
queen will mate with drones of the Africanized population.











As a result of these collective characteristics, Africanized bees have now spread

to most of the American continent, replacing European honey bees. Africanized honey

bees are now found not only in areas where traditional beekeeping existed, but also in

natural areas forming numerous feral populations (wild colonies), where before they were

relatively scarce (Michener 1973, 1975). One of the characteristics that makes the

Africanization process even more striking is the fact that African traits of the Apis

mellifera subspecies have persisted almost entirely. Genetic studies indicate that

Africanized bees resist changes in gene frequency toward the European type through

partial reproductive isolation between European and Africanized bees. This partial

reproductive isolation operates in ways that perpetuate the African ancestry and give the

Africanized bee further advantage over European bees (Daly 1991, Fletcher 1991,

Ratnieks 1991, Spivak 1991).

New Challenges and Opportunities

Changes in honey bee populations have brought about many challenges to

beekeepers. Behavioral characteristics of the Africanized bee, such as their defensive

nature and their tendency to swarm and abscond, have destabilized beekeeping and honey

production. At the same time, it has become common that after an initial period of

adjustment, the Africanization actually favors beekeeping, at least in some tropical and

semitropical settings.

The behavioral trait of Africanized bees that has made them widely known among

the common public, and one of the biggest challenges for beekeepers, is their defensive

behavior (Breed 1991, Rinderer and Collins 1991). Africanized bees show greater

sensitivity to colony disturbance, have a greater ability to communicate alarm within and









between colonies, and have a greater capacity to respond quickly by mounting massive

attacks on intruders (Michener 1973). This defensive nature of the Africanized bee

makes their management even more difficult. Beekeepers have had to understand

different aspects of the Africanized bee biology in order to modify management practices

and adapt to these new conditions.

To deal with the defensive nature of the Africanized bee, beekeepers have learned

to relocate apiaries away from homes, barns, animals and roads (Kent 1991). They also

now need to use more appropriate clothing. Beekeepers have learned when to work in

the apiary, since the temperament of a colony varies depending on weather and foraging

conditions, food stores, prior treatment, colony size, etc. (Michener 1973). The

tendencies of the Africanized bee to frequently swarm and abscond have made

beekeeping a more intensive activity, requiring more visits to apiaries than before.

Beekeepers have learned to capture swarms in bait hives as a procedure to increase

colony numbers and reduce wild populations (Helmich and Rinderer 1991). Beekeepers

have also learned to select toward mildness by destroying and requeening the most

defensive colonies (Michener 1975). Africanized populations do not represent a well-

defined phenotype, but rather display a wide range of behavioral characteristics, ranging

from relatively gentle and manageable to extremely defensive and difficult to manage

(Spivak 1991). Spivak (1991) provides examples from Costa Rica where manageable

and productive Africanized colonies with intermediate characteristics have been

successfully maintained by beekeepers who have modified their management practices

and requeen or kill undesirable colonies.









According to Michener (1973), most of the beekeepers in Brazil have learned how

to handle the new honey bee and actually prefer it in spite of its drawbacks, because of its

high productivity. He found that Africanized bees in Brazil are reported to produce more

honey per colony than European bees. According to Goncalves et al. (1991), beekeepers

in Brazil, with the aid of government technicians and university scientists, have been able

to adapt. Brazil began producing large quantities of honey and became one of the largest

honey producers in the world only after the introduction of the Africanized bee. Most

beekeepers in Brazil are happy working with Africanized bees because it is easy to obtain

bees and because of their higher productivity (Goncalves et al. 1991).

Kent (1991) reviewed the impact of Africanized bees in Peru, comparing areas

with and without Africanized bees. He concluded that average colony production was

consistently greater in Africanized regions. He found that it is common after an initial

period of adjustment when honey yields may decline, that yields then tend to increase

notably. Small beekeepers tended to abandon beekeeping, while commercial beekeepers

with greater financial resources and technical knowledge were more likely to continue.

To adapt, beekeepers became better equipped and more progressive in their management

practices as well as more involved in information networks to share experience and

knowledge.

Spivak (1991) concluded that in different regions of Costa Rica, the Africanized

bee caused beekeepers who were not willing to modify their practices and adapt, to

abandon beekeeping. Nevertheless, other beekeepers were able to reduce the impact of

the Africanized bee by employing good management and selection practices. She states

that if beekeepers are willing to follow advice from other beekeepers and researchers who









have had positive experiences with Africanized bees, the Africanization process can have

a positive effect overall.

Hellmich and Rinderer (1991) show that the introduction of the Africanized bee to

Venezuela greatly decreased beekeeping initially. However, after a decade, beekeepers

began to learn to manage the Africanized bee. These authors conclude that a new

generation of beekeepers has emerged in Venezuela, better informed and prepared.

However, they report that honey yields were still greater before the Africanized bee,

attributing lower present productivity to the increased competition for nectar sources by

the numerous feral colonies.

It has been the general case that beekeeping in many areas of South and Central

America has been favored by Africanization. However, it is important to note that

beekeepers that have been able to adjust and take advantage of the new situation have

been those with the capacity of acquiring equipment and becoming better informed,

either by technical training or through different media channels (by personal means or

through government programs).

In Mexico, the effects of Africanization on total national honey production have

been minimal, basically because beekeepers have had time to adapt to the new

management conditions (Echazarreta et al. 1997). Governmental programs such as the

Program Nacional para el control de la Abeja Africana (PNPCAA) formed as an

agreement between the Agricultural Secretary (at that time SARH2) and the United States

Department of Agriculture (USDA), has aided in the mitigation of the impact of

Africanization (Sanchez-Vazquez and Colli-Ucan 1992). Data from the Centro

Estadistico Agropecuario show that national annual honey production decreased slightly









in the 1990s, averaging approximately 58,000 metric tons from 1990 to 1999, as opposed

to an average of 60,000 for the previous decade. Part of this decrease may be the result

of Africanization, but other factors such as deforestation, intensive use of agrochemicals,

national and international low prices, market competition, and the spread of Varroa

jacobsoni (an exotic mite that affects honey bees and that has been spreading in recent

years negatively affecting honey production), have also been used to explain such

decreasing tendency (Angulo-Carrera 1992, Sanchez-Vazquez and Colli-Ucan 1992).

Nevertheless, certain areas, such as La Montafia (see chapter 4), have actually increased

honey production in recent years, favored by Africanization and other factors.

Conservation Concerns

There is increasing concern about possible effects that introduced bee species may

have on native fauna and flora. Exotic honey bees may compete with other species of

flower-feeding animals, such as nectar-feeding birds and other bees (Sugden 1996,

Sugden et al. 1996). One concern is the effect that the increasing number of Africanized

bees, particularly as feral populations, may have on the natural assemblage of pollinators

(Buchmann 1996, Corbet 1996). Not only is there evidence that the Africanized bee

displaces native pollinators (Roubik 1978, 1996a, 1996b; Roubik et al. 1986), but also

that they may be ineffective at pollinating native flowers and may interact in complex

ways with native pollinators, reducing the amount and efficiency of pollen transfer

(Kearns and Inouye 1997). Although the direct effect that this may have on the faunal

and floral compositions in the mid- and longer terms is hard to measure (Buchmann

1996, Roubik 1996a, Sugden 1996), it is unlikely that it will have no effect. It is


2 Secretaria de Agricultura y Recursos Hidraulicos









important to keep this considerations in mind when promoting beekeeping as an activity

compatible with forest conservation.

Research regarding the biology, reproduction and ecological relation with other

organisms of stingless bees in the Yucatan Peninsula and in Mexico is only beginning

(Quezada-Euan 2001). Not only does beekeeping with stingless bees represent part of a

cultural wealth of indigenous populations in Mexico, including the Maya area, but it is

also considered to have great economic potential for honey production and for

pollinization of commercial products. Appropriate technology could make this activity a

viable and profitable one (Quzada-Euan and Gonzalez-Acereto 1994) and should be

promoted in the context of current honey production.


Conclusions

Apiculture is an important economic activity in Mexico and the Yucatan

Peninsula is one of the most important areas within the country (Echazarreta et al. 1997,

Sanchez-Vazquez and Colli-Ucan 1992). In part, this results from the rich natural

melliferous and polliniferous flora found in the area (Chapter 5) and from the beekeeping

tradition of the Maya population. Beekeeping with native bees of the subfamily

Melinponinae began with the Mayas over 3000 years (Guzman-Novoa 1996). Intensive

management of certain species, particularly Mellipona beechei, resulted in a bee culture

important in religion, medicine, food, and trade (Labougle and Zozaya 1986, Sanford

1973). During the 20th century, beekeeping with Apis mellifera started in the Yucatan

Peninsula, increasing in importance throughout the century. During the 1960s and 1970s

beekeeping with Apis mellifera by the Maya population also gained importance,

becoming one of the principal economic activities for the rural sector (Echazarreta et al.









1997). During the last half of the 1980s, the arrival of the Africanized honey bee resulted

in new challenges and opportunities for beekeepers, as it did throughout most of the

American continent (Spivak et al. 1991). In the Yucatan Peninsula and the rest of

Mexico, beekeepers have had time to adapt to the new management conditions and the

effects of Africanization have been minimal.

While beekeeping in the Yucatan Peninsula with Apis mellifera results in a viable

and profitable economic activity, more efforts should be made to promote beekeeping

with stingless bees. Future studies should address the impacts that beekeeping with

Africanized Apis mellifera may have on native pollinators, and ultimately to the function

and structure of flora and fauna.














CHAPTER 4
SOCIO-ECONOMIC IMPORTANCE OF BEEKEEPING IN THE MAYA FORESTS:
THE EXAMPLE OF THE LA MONTANA AREA


Introduction

Beekeeping in the Yucatan Peninsula has been a traditional activity since ancient

times. Maya people are known to have practiced intensive management of stingless bee

species such as Melipona beecheii Benneth. Although meliponiculture (beekeeping with

native stingless bees of the subfamily Meliponinae) is now seldom practiced, beekeeping

with Apis bees is an important activity in the region. Beekeeping is now carried out with

modem techniques (movable frames) and with bees of the species Apis mellifer L. Over

40% of the national honey production comes from this region, contributing to a large

honey production industry with world-wide importance, since Mexico is one of the

leading producers and exporters of honey (Angulo-Carrera 1992, Crane 1990,

Echazarreta et al. 1997, Guzman-Novoa 1996).

In the Yucatan Peninsula, about 80% of the beehives are owned by small farmers,

making beekeeping an important source of income and jobs for the rural sector (Guzman-

Novoa 1996). In protected regions such as the Calakmul Bioshpere Reserve and

surrounding areas, honey production is considered to have great potential for

development (Acopa and Boege 1998, Galletti 2000). The Maya landscape hosts a very

rich melliferous and polliniferous flora found in the different forests types, yielding

abundant nectar and pollen for honey production, and representing an important

economic resource (Rico-Gray et al. 1991).









The objective of this research it to analyze the socio-economic implications of

beekeeping at La Montafia, an area found north of the Calakmul Biosphere Reserve.

Beekeeping in this area is the number one income generating activity (Galletti 2000,

Porter 1995). Through conducting research with beekeepers from the area for a period of

two honey productive season, information was sought to understand the socioeconomic

importance of beekeeping, the benefits that the local families and communities are

obtaining from this activity, as well as the strengths and limitations of honey production

and marketing.


Methods

Study Site

La Montafia is an area north of the Calakmul Biosphere Reserve in the state of

Campeche, Mexico. It is formed by a group of eight ejidos occupying an area of

approximately 185,000 ha, and having a population of about 2700 inhabitants mostly of

Yucatec-Mayan origin (Galletti 1999). The type of climate for the area has been

determined, according to the Koppen climate classification system as modified by Garcia

(1973), to be subhumid, warm tropical (Flores and Espejel 1994). The mean annual

temperature is above 220C and the mean annual precipitation is 1,223 mm1. The most

dominant vegetation type is medium-stature subperennifoleous forests, where the mean

height for trees is 25m and where from 25 to 50% of the tree species shed their leaves

during the dry season. This type of vegetation is mixed with high-stature

subperennifoleous forests, low-stature forests that inundate periodically, aquatic

vegetation, and secondary vegetation in different successional stages (Ucan et al. 1999).









Soils in the area have been formed mostly by the dissolution of calcareous rocks that over

time has resulted in a karst topography (Flores and Espejel 1994). Drainage is

subterraneous with no superficial flows of water, except in the rainy season during storms

and inundations when rain can be very intense and water is transported in temporary

surface flow channels (Gates 1999).

Research Conducted

The field study lasted from November 1998 to June 2000 and included two honey

production seasons. The research was facilitated by Oxfam/Comadep2, a non-

governmental organization working in the area, which was assisting a group of

beekeepers to develop a legally established cooperative to commercialize and increase

honey production. Formal and informal interviews with 94 beekeepers, many of them

members of the cooperative formed with the assistance of Oxfam/Comadep (Sociedad de

Solidaridad Social, Productores Unidos Lol k'ax -from now on I will refer to them as Lol

k'ax), provided information about honey production and marketing, as well as about

other relevant issues related to beekeeping in the area. Information was also derived by

assisting the cooperative in their financial and organizational tasks. The members of Lol

k'ax are representative of beekeepers from seven different localities (Xmaben, Ukum,

Chunek, Pachuitz, Xmejia, Belha, and the ranch of Xnoha) that have been organized

since 1998. Other beekeepers in the area are members of different cooperatives or

market their honey through the informal honey market. Informal interviews were also

conducted with external key people, such as the veterinarian working in the area, Victor


1 Data for temperature and precipitation were obtained from the Comision Nacional del Agua for the period
from 1984 to 2000.









Camara, who has worked in the Yucatan for more than 30 years, and with members of

Miel Maya (the company that has bought honey from Lol-K'ax). Regular assistance to

the training sessions carried out with subgroups of beekeepers in the different localities,

especially a group of women beekeepers from the Belha community, provided

information about different issues of beekeeping management practices in relation to

honey marketing.


Results

Beekeeping at La Montafia

The inhabitants of La Montafia, of Mayan descent, have had a tradition of

beekeeping for many generations. People in the area remember their parents, or

themselves when young, keeping large numbers of colonies of xuna'an kab. These bees

are traditionally kept in specially made hollow logs (called in Mayan, hobones) in

thatched structures within the solar (the Maya homegarden), or near their agricultural

plots (the milpa). Wax and honey from these bees are greatly valued for different

religious and medicinal attributes. Meliponiculture (the termed used for beekeeping with

bees from the subfamily Meliponinae) is now seldom practiced. Even though

occasionally some still keep a few hobones of these bees, beekeeping activity has

changed to be a more commercial one, directed towards regional, national, and

international markets, and carried out with Africanized strains of Apis mellifera.

Currently, beekeeping is the number one income generating activity at La Montafia,

bringing considerable amounts of capital into the area (see below).



2 Oxfam/Comision Mesoamericana de Asistencia Para el Desarrollo Popular, Asociaci6n Civil, at that time
Oxfam-Belgium who began a project in the area in 1993 funded by the European Community and by the
Oxfam office in Belgium, that has extended until recently.









Modern beekeeping was promoted in the area during the early 1970s by

government initiatives (specifically by the PIDER -the Programa de Infraestructura para

el Desarrollo Rural) through an effort to integrate the area into the national economy

(Camara, personal communication). During that time beekeeping was carried out with

the "Italian" bee (Apis mellifera ligustica), one of the European subspecies. The arrival

of Africanized strains of Apis mellifera (Apis mellifera scutellata) into this area occurred

during the last half of the 1980s (Crane 1990; Camara, personal communication). With

this, many people discontinued beekeeping. Numerous stinging incidents, fatal in some

cases, are reported to have occurred. Nevertheless, beekeeping continued, especially

among the younger population.

During the 1990s, beekeeping actually gained importance. Two main factors

made this possible. First is the construction of the Hopelchen-Xpujil road, which was

finished in the early 1990s. This road opened access to the area and led to many changes

particularly in communications and market accessibility, and with these, the involvement

of the population in the regional economy. Changes have been astonishingly visible from

year to year in many different social, economical and political aspects. For beekeeping in

particular, alternatives to the traditional commercialization scheme have emerged. For

example, instead of one or a few intermediaries (the coyote) with absolute control over

markets (buying the honey from individual producers and re-selling it to a buyer), many

people are now organized into beekeeping cooperatives, although the coyote scheme also

remains in operation. The road has also meant access to information and awareness of

the different limitations and possibilities for commercializing bee products (aided by non-

governmental organizations that have established in the area). Second, the arrival of the









Africanized bee, with high numbers of feral colonies, has meant greater bee availability.

A prospective beekeeper needs only to acquire the basic equipment (mainly wooden

boxes which are produced in the area) to establish a productive unit. Many of the local

population (particularly younger people) have taken advantage of the opportunities for

producing and commercializing honey.

Socioeconomic Importance of Beekeeping

Honey production has traditionally been part of a larger natural resources

management system in which activities complement each other in space and time.

Segregation in space is accomplished by the use of different ecosystems for different

productive activities, and in time by the distribution of different activities throughout the

year according to season (Porter 1995). In this way, people follow the strategy indicated

by Toledo (1991) to be typical of indigenous management systems, maximizing the

utilization of all available ecosystems and getting a number of essential products

throughout the year. The strategy buffers households and communities against market

fluctuations and environmental hazards.

People at La Montafia practice a type of slash-and-burn agriculture, called the

milpa, in which corn, beans and squash (among other perennial and annual crops) are

planted. This is an important traditional activity carried out by all men in the area. It is a

subsistence activity and most food consumed by families, or used for feed for domestic

animals, is produced in this way. Other types of agriculture, like agroforestry plots and

homegardens, are also common and important for meeting family needs. Agricultural

activities are mainly for subsistence and play only a small role in bringing direct cash to

the families, since only a few of the products are sold.









Cattle raising is an activity that has recently gained in importance. Government

programs have aided in its promotion and even though most people have only a few head

of cattle, some individuals are investing much of their capital and labor into increasing

their herds and pasture lands. A socioeconomic analysis conducted in the community of

Xkanha showed that about 68% of the families owned cattle, although, of these, 75% had

<10 head, 22% had >10 but <16, and only 4% had >40 animals (Porter 1995).

Nevertheless, cattle raising is an increasing trend and an activity perceived as important

for accumulating capital and gaining social status. Also, cattle are a source for quick

cash and used in emergencies. In recent times, the government has given considerable

credit to some individuals for establishing cattle ranches. Currently, cattle is at an

investment stage with many people investing their surplus income in preparing pasture

lands; there are people who invest in pastures even though in some cases they still do not

own cattle. According to Galletti (2000), the last 5 years at La Montafia have shown

potrerizaci6n (pasture growth) and cattle raising continues to gain importance.

Use of timber and non-timber resources is important for people in the area to

different extents. Timber extraction is not a very important economic activity for these

communities in general. Most of the commercial species, mainly cedro (Cedrela odorata

L.) and mahogany (Swietenia macrophylla King), have already been eliminated from the

forests by logging. Other native species with commercial potential such as granadillo

(Platymiscum yucatanum Standley), siricote (Cordia dodecandra A.DC.) and chaka

(Bursera simaruba (L.) Sarg.), among others, are being currently extracted. Although

timber extraction is limited by government permits and markets, there is no official

organization or structure in which the local populations can benefit. Usually, wood is









sold as standing volume to companies that bring in their own trucks and equipment to

extract the timber. Local people often receive money for the labor of felling trees, but not

for much else, and most of the time they are unaware of the volumes that are being

extracted. Usually people obtain very little revenue from this activity, which favors

companies or only a few of the local people involved in woodshops for making furniture

and beekeeping equipment. The only community that is currently benefiting in a

significant economic sense from timber harvesting is Xkanha, where most of the timber

extraction is being done in an illegal manner with no management plan. Even in this

case, extraction has again favored only a few specific families (Porter 1995).

Chicle extraction has historically been an important economic activity in the area.

Chicle is the gum tapped from the Manilkara zapota tree, which is used for making

chewing gum. Historically, this activity has been very important in the area and until

recently it was the activity that generated the most income. Chicle has always been an

export product. At different points during the last century (specifically during the periods

of the World Wars), chicle camps guided the establishment of communities.

Nevertheless, most of the income benefited outside companies and contractors and little

benefit was felt by the local people (Ramayo-Lanz 1998). After the chicle boom ended

(shortly after World War II), basically when synthetic gums replaced latex for making

chewing gum, chicle trees continued to be tapped with an up-and-down market demand,

but still constituting an important source of income. Now, chicle production is the

second income generating activity after honey production. In an interview conducted by

Galletti (2000), 43% of total ejidatarios at La Montafia were reported to be practicing

chicleros (many chicleros, although skilled, stopped conducting the activity because of









age or other reasons, since it is a highly dangerous job), and in some of the communities

practically all ejidatarios are chicleros (e.g., Xkanha, and Pachuitz). Annual chicle

production in the region ranges from 17 to 26 metric tons of dried gum (Galletti 2000,

Porter 1995). As a result of the Asian economic crisis, starting in 1997, chicle has not

been bought from the area because the Japanese company that bought most of the product

went bankrupt. Chicleros are now waiting for a new contract to continue with this

activity.

Other economic activities contribute in different degrees to the economies of La

Montafia families and communities. Some individuals have woodshops or other

commercial establishments, or own a car or a chainsaw. People may also work for wages

either within or outside the communities. Women make embroidery and other crafts,

which although limited by market availability and low prices, sometimes represent an

important income to the family (mainly because it goes directly to family use). Other

activities such as hunting may also be important for increasing the well being of families,

mainly because meat is an important protein source. Government subsidies can also

contribute greatly to total family income, but this depends on the specific government

programs at the moment (Galletti 2000, Porter 1995).

Beekeeping is currently the number one income-producing activity in the area.

As explained before, it is a relatively recent activity as practiced today with Africanized

Apis mellifera and given the current marketing possibilities through better accessibility to

the area. In a survey conducted in the region, 76% of the men interviewed reported

practicing beekeeping (Galletti 2000). In a study conducted in Xkanha in 1995, over

60% of ejidatarios were involved in beekeeping (Porter 1995). An average of about 200









metric tons of honey are produced annually in the region, depending on weather

conditions (Galletti 2000). Prices vary greatly (from 1994 to 2000 it varied from 1 to 12

pesos/km), but considering a common current price in the region of $0.7 USD/km, yields

an annual income of $140,000USD/year. Only about 40% is re-invested into the activity,

but considering that the minimum wage can be as low as $3.00 USD/day, it is a

considerable amount of money.

Most of the population has only small productive units, with few colonies and

involving a minimum of investment. In interviews carried out with 63 beekeepers of four

different communities conducted as part of this research in 1999, 32% of the producers

had <10 colonies, 46% had 11 to 20, 17% had 21 to 30 and only 5% had >30 colonies. In

a study conducted in Xkanha, 60% of total income made from beekeeping in 1995 was

generated by only three producers (Porter 1995). Thus, beekeeping revenues still largely

benefit a relatively small sector of the population. As practiced today, beekeeping is

limited by technical knowledge and the investment capacity (time and money) to increase

colony numbers. Nevertheless, current programs established in the area (e.g.,

Oxfam/Comadep, and EDUCE,A.C.3) aim at increasing the potential in the region, by

assisting in the organization of beekeepers' cooperatives, establishing markets and

providing technical advice.

Honey Production and Marketing

Honey harvest season at La Montafia runs from mid- to late February through late

May or early June. This timing varies from year to year depending on the winter rains

and their timing. There are several factors that determine production success, besides the









necessity of having the skills to work with Africanized bees. The length and regularity of

precipitation controls the flowering period (Chapter 3). Also, beekeepers need to know

how to manage their colonies to prevent absconding and swarming. People report that it

was as easy to expand the number of colonies by catching moving swarms, as it was to

lose them all because of lack of assistance or an external factor that led colonies to

abandon their nests. Of the 63 beekeepers that I surveyed during 1999, only 14%

reported no change in colony numbers from 1998 to 1999 (Figure 7). Actually one

producer reported to have gained 100% of his colonies during 1999 and another one

reported to have lost 100%, so that colony numbers can clearly change greatly from year

to year.

The main cause of colony loss was reported to be absconding during the dearth

period. Other factors that affect colony numbers included pests and predators that either

lead colonies to abscond or to perish. The most threatening predators were reported to be

ants, especially army ants of the Ecitoninae subfamily, locally known in Mayan as xulab,

and the oso colmenero or hive eater (Eira barbara senex) a mammal of the Mustelidae,

known by its Mayan name as sanjol. These predators are also reported in the literature to

disturb beekeeping in other parts of the Peninsula (Echazarreta et al. 1997, Sanchez-

Vazquez and Colli-Ucan 1992). The varroa mite (Varroajackobsoni) is a recent pest that

can greatly affect colonies. According to Echezarrieta et al. (1997) the varroa mite has

become an increasing problem in the Yucatan Peninsula since 1995. At La Montafia, all

beekeepers reported to have problems with varroa to different degrees and needed to treat




3 Educacion, Cultura y Ecologia, A. C., a non governmental organization based in Hopelchen assisting the
organization of the beekeepers cooperative named Kabi-tah, of which some of the beekeepers of the La
Montafia region are members.











colonies either with chemical products (e.g., fluvalinate or flumethrin) or organic ones

(e.g., formic acid). Other diseases are reported to affect the well-being of apiaries to

lesser degrees.

The absconding of colonies can be avoided to a certain extent by applying

appropriate management practices (e.g., killing ants before they attack the colony;

feeding; removing or providing frames; Camara, personal communication). The effects

of diseases can also be diminished through management. Factors other than colony

number, such as the age of the queen and colony size and strength, are also important in

determining productivity. These factors can also be controlled with management

(Camara, personal communication).


14-* I

12--/


C) AC C C) C) AO
o o o A c- O O A
oco L
(C0 '- CO
% of colonies gained of colonies lost


Figure 7. Number of beekeepers that gained, lost, or had no change in their colony
numbers from 1988 to 1999, according to a survey of 64 beekeepers of the communities
of Chunek, Pachuitz, Ukum and Xmaben conducted during 1999.




In summary, production is determined by both natural factors and management

practices which depend on the beekeepers' skills and the resources invested in the

activity (both in time and supplies). The average kg of honey produced per colony was


a 1



0
o

g
.

2
z


0-.f

8-

6-


4 -....









reported by a group of 64 beekeepers of four different communities to be 60 kg/colony,

although some reported to have produced 10kg/colony and some up to 150 kg/colony.

As mentioned above, honey is commonly sold through a coyote (the intermediary), who

buys honey directly from a producer and re-sells it to a buyer. The intermediary, thus

determines price. The coyote also has the role of providing credit to producers for

maintaining and managing apiaries throughout the year. Credit provided during the off-

season benefits the coyote, who then sets the prices, terms and conditions. Within this

commercializing scheme, there is little room for capitalization or re-investment.

Most recently, some producers have organized themselves into cooperatives, with

assistance of non-governmental organizations (NGOs). The purpose is to create a system

where members are provided with social and administrative mechanisms in order to (1)

increase their technical knowledge and productivity (through technical assistance

facilitated by the non governmental organizations), (2) obtain government credit, (3)

negotiate directly with markets, and (4) create mechanisms of capitalization that increase

the cooperative's capacity (technical, organizational and administrative) and ultimately

create structures that benefit members in the mid- and long terms.

Cooperatives have also been formed for chicle production, with assistance from

the Plan Piloto Chiclero. The objective of the Plan Piloto Chiclero is to organize

chicleros to create a "commercialization front", in which producers are empowered to

negotiate directly with the buying companies, establishing terms and conditions in their

favor. The cooperative also has the role of creating services to benefit members in the

mid- and long terms, such as employing new technology, and offering loans, life

insurance, and retirement funds (Aldrete, personal communication).









The Lol-K'ax Cooperative Example

The cooperative of Lol-k'aax was formed at the beginning of 1999 with the

assistance of Oxfam/Comadep, which started a program of technical assistance for

beekeepers in 1998. The initiative for forming the cooperative came both from

Oxfam/Comadep and from a group of young beekeepers who had already participated in

the Kabi-tah cooperative, but had renounced their membership. Kabi-tah is a cooperative

whose headquarters are in Hopelchen (formed through the assistance of EDUCE, A.C.)

which has been in operation for some years. The Kabi-tah cooperative incorporates a

much larger region, particularly in the north. Some of the people at La Montafia found

that they were in at a disadvantage because their harvest season begins later than in the

north. At the end of the harvest season, people at La Montafia are affected the most if

resource shortages occur at the cooperative (cooperatives function in a way that they need

capital to buy honey from individual producers before receiving payment from the buying

company). Also, after a Kabi-tah experience in which a shipment of honey exported to

Germany was found to be contaminated with chemical residues and rejected (something

that at that time resulted in a general crisis), beekeepers realized the importance of having

adequate technical training and a more regulated supervision of quality control.

Oxfam/Comadep assisted Lol-K'ax to legally establish the cooperative, to obtain

credit, and to locate markets. It has also carried out a formal technical assistance

program. Nevertheless, Lol-k'ax has been in charge of much of its own management.

Currently, Lol-K'ax has 72 members from 6 communities (Xmaben, Ukum, Pachuitz,

Chunek, Xmejia, and Belha) and one settlement (Xnoha -a group of people that have

settled in National Territory for over 50 years). During the last two years, they have sold









part of their product to Miel Maya Honing, part of the Fair Trade Market4, with

headquarters in Belgium. The experience of Lol-K'ax in their export enterprise has not

been an easy one. For example, Lol-k'ax has had to establish mechanisms of self-

financing for product commercialization. They also have had to cope with strict quality

control norms established by the international market. Miel Maya buys honey of the best

quality: no more than 18% water content and free of chemical residues. This has resulted

in a very strict program of quality control and has necessitated increased technical

training for cooperative members.

Contamination of honey usually results from use of chemicals and antibiotics for

pest control, so that producers have had to use alternative pest control measures. Water

content in honey can vary greatly, based on natural causes such as nectar type and

weather conditions, as well as management practices. High water content results in

fermentation. Although the international market usually only buys honey with a water

content 18%, 20% water content is considered to be good quality honey within the

national market.

In 1999, Lol-k'aax established a contract for 20 tons of honey with Miel Maya,

although the cooperative was able to only gather <12 tons, mainly because Lol-k'ax did

not find mechanisms to finance the commercialization until the end of the season, when

the honey had a naturally high water content. The honey that arrived in Belgium had a

water content >18 % and failed to meet the terms of the agreement. Nevertheless, Miel


4Fair Trade markets belong to the International Federation of Alternative Trade (IFAT) whose "code of
practice" establishes: "To trade with concern for the social, economic and environment well-being of
marginalized producers in developing countries. This means equitable commercial terms, fair wages and
fair prices. Unfair trade structures, mechanisms, practices, and attitudes will be identified and avoided. To
cooperate and not compete. To promote fair trade and social justice in the interest of the producer, and not
to maximize profit at the producer's expense" (taken from internet web page
httIp w\\\\.goedwerk.nl/engels/en page2.html).









Maya gave Lol-k'aax a second opportunity and established a new contract for another 10

tons of honey for the year 2000. After one of Miel Maya's representatives came to the

area to supervise and test honey quality, Lol-k'aax shipped to Belgium almost 12 metric

tons of honey of excellent quality.

Lol-k'ax produced approximately 44.5 metric tons of honey in the year 2000

(Figure 8), of which 26% was honey with 18% water content (that sold to Miel Maya).

The rest was honey of <19% water content (47% of total honey produced), or <20%

water content (27% of total honey produced). Honey of 19 and 20% water content was

marketed nationally. Quality control in this case was much lower and Lol-K'ax was able

to sell honey of < 20% water content at the same price and were also exempted from

conducting lab tests for chemical residues. The price of honey sold to Miel Maya (the

honey of best quality) was set at $10.50 pesos per kilogram (approximately $1.05 USD),

with Miel Maya paying for shipment and containers. The price of honey sold to the

national company was also $10.50, but Lol-K'ax had to pay for shipping and containers,

which resulted in a net price of $9.00 pesos per kilogram of honey sold (approximately

$0.9 USD).

The amount of honey produced during 2000 is not representative of the

production capacity of Lol-K'ax. That year was drought, with only 22.3 mm of

precipitation for the months of January to May compared to a 17 year mean of 56 mm

(SD 22.1). People complained about the lack of flowering especially at some locations.










50 -
45 -
40
0 35
30 -
0
5 25
0 20 -
'- 15
10



<18% <19% <20% Total
Maximum water content (%)


Figure 8. Honey produced by Lol-K'ax during 2000. Values are given for 18, 19 and 20
percent water content and for total production. Data obtained from Lol-k'ax financial
summary of 2000.



Total production for Lol-K'ax in 1999 was not recorded, since most of the people

sold their honey in that year to the coyote. In my survey conducted after the harvest

season of 1999, 65 beekeepers reported that they had produced 69 tons in 1999, and 58

tons of honey in 1998. These figures are only estimates and more years of production

records are needed before production capacity for Lol-K'ax can be reliably predicted.

An assessment of the advantages of selling honey internationally through

alternative markets such as Miel Maya is critical. Although prices have been better when

honey is sold to Miel Maya, this has been achieved only through a large effort by the

cooperative and through subsidies from Oxfam/Comadep. Although the "Fair Market"

provides better prices, the cooperative needs to find mechanisms for self-financing.

Producing the high-quality honey demanded by the international market means

increased investment, not only by producers who need to use alternative products for pest

control, more intensive management, etc., but also from the cooperative, which lacks









mechanisms for self-financing and does not have adequate infrastructure. Much of the

Lol-K'ax failure to produce 18% honey in 1999 was the result of poor handling practices

due to a lack of infrastructure (containers were not well sealed and were stored in the

open). Competing in the highest quality honey market with no infrastructure places the

cooperative at a disadvantage and in a very fragile position. The Miel Maya experience

resulted in relationships of trust-and-distrust among the different parties. The company

has not been sure of the cooperative's capacity to produce good quality honey, but has

nevertheless given it the opportunity to do so. The cooperative is wary of the company's

fairness, because it has not shown sensitivity to the cooperative's current situation and

has not paid in advance, even while knowing that there is a lack of self-financing capacity

and the infrastructure needed to meet the strict quality control terms. There is also some

mistrust from Lol-k'ax members towards the cooperative itself, given that the Lol-K'ax

enterprise of selling honey to alternative markets has implied eliminating the coyote-

producer relationship without sufficient alternative financing. Lol-k'ax membership has

changed from year to year; about 36% of the members resigned in 1999, while 50% of

the members were new in 2000.

It is commonly agreed that the cooperative-Miel Maya business relationship has

potential to bring great benefits in the future. However, for this to happen Lol-K'ax

needs to strengthen its self-financing mechanisms, develop a trustworthy relationship

with all of its members, and understand and predict production capacity. For Miel Maya,

Lol-K'ax as a supplier is still an "experiment," and it will not consider the cooperative as

part of the enterprise until further demonstrations of production capacity are completed.

Lol-K'ax 2000's commercialization was largely a failure, because although it was able to









purchase honey produced from all of its members (through state credit5) and fulfilled its

agreement with Miel Maya successfully, this represented only 25 % of the total

production. The rest of the honey was sent to a national market under a previous "word

agreement". This honey was not paid for as of 2000, and Lol-k'ax has also delayed

paying the state government. Thus, Lol-K'ax is in a very fragile position. However, the

cooperatives' experience and knowledge about honey commercialization is increasing

and perhaps moving it closer to achieving production and commercialization goals.

Prospects for Lol-K'aax

Regardless of the problems that have come about with establishing markets, Lol-

k'ax has begun to create mechanisms for establishing their enterprise by investing with

the only capital they have: bees and equipment. With the contribution of one brood

chamber of bees per member, the cooperative established several apiaries at the

beginning of 2000. The idea was to create self-sustaining production systems to support

the cooperative. Apiary production for 2000 was very low (4.2 % of total production)

because the apiaries were still in the establishment phase. Nevertheless, they were

established with the supervision of an apiculture specialist, using advanced techniques for

obtaining high production. Some of the production practices carried out in the

cooperative's apiaries and presented in technical training programs to Lol-K'ax members

are provided in Table 4.









5 The credit was provided by the Fondo Nacional de Apoyo a la Empresa Social (FONAES), which
translates into National Fund for Aiding Social Enterprise.










Table 4. Some production practices taught at the Technical Training Program for Lol-
K'ax members and used in the cooperative's apiaries at La Montafia.

HARVEST MANAGEMENT METHODS SPECIFICALLY TO CONTROL WATER CONTENT
CONTROL OF CONTAMINATION OF HONEY FROM THE USE OF CHEMICALS AND ANTIBIOTICS USED IN
CONVENTIONAL PEST AND PARASITE CONTROL
USE OF ORGANIC ALTERNATIVES FOR CONTROLLING PESTS AND PARASITES.
QUEEN REARING.
QUEEN SYSTEMATIC REPLACEMENT.
USE OF QUEEN EXCLUDERS
ADEQUATE SUPERVISION OF BROOD AND HONEY SUPERS.
USE OF ORGANIC WAX FOUNDATIONS (MADE FROM BEESWAX RATHER THAN SYNTHETIC ALTERNATIVES).
USE OF ADEQUATE CLOTHING AND EQUIPMENT.





The cooperative's apiaries are intended to provide a source of cash for the

cooperative itself, as well as to serve as models and learning centers. As explained

before, beekeeping skills in the area are low and technical support for increasing

productive capacity in the region is very important. Only a small group of beekeepers are

substantially contributing to production. 2000's production is very illustrative of the

latter. Only 15% of the members of Lol-K'ax produced over 44.4% of the total

production for the year. The rest, about 85% of total producers, accounted for only

55.6% of total production (Figure 9).

Another objective of the cooperative's apiaries is to establish productive units

where organic honey can be produced. This will potentially bring added value or market

share to production. At the same time, it is intended that in the mid- to long-term, the

apiaries can serve as models for producing other bee products for sale in alternative

markets.











25

20 --
U
2 15--
0.
0
S10- ----

z 5


0-025 0251-05 0501-075 0751-10 101-125 1251-15 1501-175 1751-20 >2 0
Metric tons of honey produced per beekeeper in 2000

Figure 9. Metric tons of honey produced in 2000 by each of the 74 producers belonging
to the Lol-K'ax cooperative. Data obtained from Lol-k'ax financial summary of 2000.



Benefits for Lol-k'ax so far, in addition to the new technology and

commercialization, have been the acknowledgment from members that beekeeping can

provide a viable alternative income for beekeepers and their families, provided there is an

investment. Another issue is the inclusion of women in honey production, especially if

they want to be part of the Fair Market6. Generally, Maya culture does not encourage the

participation of women in these types of businesses. Currently, the beekeepers of the

community of Belha are women, who were only given the opportunity of belonging to

the cooperative after Lol-K'ax learned about the advantages of having women as

members. They now intend to include more women in beekeeping in the region.

Lol-K'ax goals are to be able to strengthen the cooperative and to reinforce its ties

with Miel Maya, as well as with other markets. Ties with domestic markets are also

important, since a large part of their production is not export quality honey. Other




6 The IFAT's "code of conduct" also includes a section under Equal Employment Opportunities, which
states "to oppose discrimination and ensure equality of employment opportunities for both men and women
who suffer from the exploitation of their labor and effects of poverty and racial, cultural or gender bias"
(taken from internet web page http://www.goedwerk.nl/engels/en page2.html).









alternative markets, such as the organic market and the market for other bee products,

could expand possibilities for Lol-K'ax and for the region in general.


Conclusions

At La Montafia, honey production is currently the number one income generating

activity. Although there are traces of traditional beekeeping, as some individuals still

keep colonies of native bees, this is becoming less and less common. A large proportion

of the economically active population manage Apis mellifera bees, with the objective of

commercializing their products through the regional, national and international markets.

The arrival of the Africanized bee (with its high numbers of feral populations), together

with better communication and transportation systems, has brought new prospects for

beekeeping in the region.

At La Montafia, the use of timber and non-timber forest products has been the

major land use type. Today, timber resources do not bring much profit to the

communities, as most of the valuable species (e.g., Cedrela odorata and Swietenia

macrophylla) have already been logged out and the use of other species has not

developed. Non-timber forest products, such as chicle extraction and honey production,

represent the current best alternatives for making a profit from forest resources. As

agricultural activities gain importance (e.g., ranching and mechanization), it is important

to consider their impacts on apiculture, as well as their viability of non-timber forest

products as economic options for development in the region, if the forest cover is to be

conserved.

Chicle has been a major export commodity in the region throughout the 20th

century, but current market trends have not been favorable. On the other hand,









beekeeping has recently developed, and much of the honey produced is being sold

nationally and internationally. Unfortunately, profits are benefiting only a small sector of

the population that has better management skills and investment capacity.

Beekeeping could provide a better economic benefit to a larger sector of the

population. An example of this is provided by Lol-k'ax, a cooperative formed recently

by a small number of beekeepers from different communities. Lol-k'ax was formed with

the assistance of Oxfam/Comadep with the objective of providing technical training to

increase productivity and also to facilitate commercialization. Lol-k'ax is beginning to

establish market ties with the international "Fair" market which provides better prices.

Nevertheless, the experience has not been an easy one as Lol-k'ax needs to find

mechanisms of self-financing and comply with high quality standards at a time when

skills and resources are still extremely limited. Better technology and skills, as well as

the diversification of the activity (e.g., to include the production of organic honey, pollen,

royal jelly, propoli, etc.) could provide more benefits to the region. Nevertheless, for this

to happen there needs to be external investment such as from NGOs or other agencies,

particularly to increase the organizational, administrative and technical skills of

beekeepers.

In spite of the limitations, there are strong prospects for beekeeping in the area.

The next chapters will deal with the implications of beekeeping at the landscape level and

its status as a non-timber forest product that favors the conservation of the forest cover.

It is important to keep in mi nd that beekeeping as practiced today favors the increase of

an exotic bee species that has proven to displace natural pollinators with unknown

consequences for floral and faunal composition (Buchmann 1996, Roubik 1996a, Sugden






85


1996). Nevertheless, beekeeping at La Montafia and other similar areas in the Yucatan

Peninsula presents an economic alternative to other land uses that will surely produce a

great impact on forest composition, such as ranching.














CHAPTER 5
FLOWERING PHENOLOGY OF MELLIFEROUS SPECIES AND ITS RELATION
TO THE BEEKEEPING CYCLE


Introduction

Throughout the Maya area, honey production has been part of management

systems since ancient times. Maya people are known to have carried out intensive

management of native bees of the subfamily Meliponinae, especially the stingless bee

species Melipona beecheii Bennett (Chemas and Rico-Gray 1991). The current

landscape is now populated by introduced honey bees (Apis mellifera L.) that share many

of the same resources with native bee species, particularly of the genus Melipona (Roubik

1991, Roubik and Buchmann 1984, Roubik et al. 1986). Today, beekeeping with Apis

mellifera represents an important economic activity for the Yucatan Peninsula. Although

comprising only 8% of the Mexican territory, it contributes nearly 40% of national honey

production. This figure is more significant in light of Mexico as one of the most

important global honey producers and exporters (Echazarreta et al. 1997).

The landscape in the Yucatan Peninsula is known to be rich in nectar

(melliferous) and pollen polliniferouss) yielding species (Rico-Gray et al. 1991). This

characteristic favors beekeeping as nectar and pollen represent the major food sources for

bees. Floral sources are distributed in the landscape both in space and time. In space,

according to the distribution and abundance of melliferous and polliniferous species, and

in time, according to the flowering periods of these species.









Flowering phenology among species varies in timing, duration and intensity, as

well as in between-year variability in response to shifts in exogenous cues (Bronstein

1995). From a "plant-pollinator landscape" perspective, flowering phenology can be

viewed as the regional flowering sequence of the entire set of acceptable plant species for

a particular pollinator. According to Bronstein (1995), the plant-pollinator landscape is

determined by an interacting set of plant and pollinator attributes, including flowering

phenology and the pollinators' search behavior and dietary specificities. For generalist

pollinators, such as the honey bee, inter-annual variation in flowering patterns of a single

species makes little difference, but rather it is the entire set of acceptable plant species

that is important. Honey bee colonies adjust their foraging patterns in time and space,

tracking the changing mosaic of nectar sources (Sneyd and Camazine 1991). Therefore,

foraging decisions are made at a spatial scale other than the one at which phenology

varies within a single species and foragers are likely to switch to a different species

instead of searching out phenologically retarded patches or individuals of the same

species (Bronstein 1995).

Flowers provide bees with the food sources they require in the form of nectar and

pollen, although extrafloral nectar, honeydew produced by insects from plant sap, and

exposed sap may contribute to some extent in certain settings (Crane 1990). Nectar is

produced by plant species as a reward to visitors and therefore not all flowering species

produce nectar and not all nectar produced by flowers may be accessible to honey bees

(Bastiaan 1984). According to Crane (1990), only 16% of the world's flowering plant

species contribute to honey bees as a food source and only 1.6% are the source of most of

the world's honey. Nevertheless, the latter figure represents as many as 4000 different









flowering species that produce nectar. Nectarless flowers that produce large loads of

pollen and that are known to be visited by bees may also be important bee resources

(Roubik 1991; Villanueva 1994).

Lists of important melliferous/polliniferous species in the Yucatan Peninsula and

other tropical areas have been compiled by a number of researchers (Crane 1990, Espina

and Ordetx 1983, Lopez-Palacios 1986, Roubik 1991, SAGAR 1998, Sarah Workman,

unpublished, Souza-Novelo 1981, Villanueva 1994). These species may be identified by

conducting melissopalynological studies (pollen analysis in honey samples in order to

determine the nectar sources of honey bees) (Roubik 1991, Villanueva 1994), or by using

"scale-hive records", measuring changes in the weight of a colony of bees and relating it

to flowering sources (Crane 1990). However, the opinion of experienced beekeepers can

serve just as well, if not better, in forming an idea of which nectar sources are of

importance to honey bees (Chemas and Rico-Gray 1991, Roubik 1991).

The contribution of a species to honey production is not only a product of its

flowering phenology, but also varies according to nectar quality and load. The energy

value of nectar and the nutritional value of pollen vary markedly among plant taxa.

Nectar sugar concentrations in flowers range from 10 to 80 percent, with concentrations

varying not only according to plant species, but also at the flower level throughout the

day (Roubik 1991). The composition of pollen also varies greatly according to the plant

source; for example, protein concentrations range from 7 to 30 % (Crane 1990). Relative

abundance may also be important in determining a species' contributions to honey

production. Nevertheless, many flower visitors such as honey bees have ranges that

include many patches in which they forage very selectively. For example, Roubik (1991)