<%BANNER%>

The Ecological Importance of Secondary Forests to Frugivorous Butterfly Communities in Mount Kanlaon National Park, Negr...

Permanent Link: http://ufdc.ufl.edu/UFE0045620/00001

Material Information

Title: The Ecological Importance of Secondary Forests to Frugivorous Butterfly Communities in Mount Kanlaon National Park, Negros, Philippines
Physical Description: 1 online resource (118 p.)
Language: english
Creator: Reeves, Lawrence E
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2013

Subjects

Subjects / Keywords: conservation -- disturbance -- forest -- lepidoptera -- philippines
Entomology and Nematology -- Dissertations, Academic -- UF
Genre: Entomology and Nematology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The Philippine Archipelago is a global conservation priority. Extensive deforestation, driven largely by logging and agricultural conversion, has resulted in near-complete loss of old growth forests. Concomitant with the rapid decline of old growth forests has been substantial increases in the land area occupied by secondary forests. It is possible that the formation of secondary forests could mitigate the effects of deforestation on forest species and play a role in augmenting existing conservation strategies. Few studies have investigated the ecological importance of Philippine secondary forests.  This study compares the diversity, community structure, vertical stratification and precinction levels of frugivorous butterfly communities occurring at secondary forest garden, post-extraction secondary forest and old growth forest sites within a contiguous forest tract on the island of Negros, Philippines. Sampling was conducted from August 2011-July 2012, using butterfly traps set in the understory and canopy, along transects established through each habitat.  Altogether, 1526 butterflies were collected, representing 21 species. There were no significant differences in species richness, abundance or diversity between the habitats and there was little distinction in the community structure of frugivorous butterflies at each habitat. The distribution of proportional species abundances, ranked by range-size, showed that levels of precinction were significantly different between all habitats except old growth forest and secondary forest garden. Although additional study is required to fully assess the conservation value of Philippine secondary forests, these results provide evidence suggesting that the integration of secondary forests into conservation plans may be beneficial, at least to frugivorous butterfly communities.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lawrence E Reeves.
Thesis: Thesis (M.S.)--University of Florida, 2013.
Local: Adviser: Daniels, Jaret C.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2013
System ID: UFE0045620:00001

Permanent Link: http://ufdc.ufl.edu/UFE0045620/00001

Material Information

Title: The Ecological Importance of Secondary Forests to Frugivorous Butterfly Communities in Mount Kanlaon National Park, Negros, Philippines
Physical Description: 1 online resource (118 p.)
Language: english
Creator: Reeves, Lawrence E
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2013

Subjects

Subjects / Keywords: conservation -- disturbance -- forest -- lepidoptera -- philippines
Entomology and Nematology -- Dissertations, Academic -- UF
Genre: Entomology and Nematology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The Philippine Archipelago is a global conservation priority. Extensive deforestation, driven largely by logging and agricultural conversion, has resulted in near-complete loss of old growth forests. Concomitant with the rapid decline of old growth forests has been substantial increases in the land area occupied by secondary forests. It is possible that the formation of secondary forests could mitigate the effects of deforestation on forest species and play a role in augmenting existing conservation strategies. Few studies have investigated the ecological importance of Philippine secondary forests.  This study compares the diversity, community structure, vertical stratification and precinction levels of frugivorous butterfly communities occurring at secondary forest garden, post-extraction secondary forest and old growth forest sites within a contiguous forest tract on the island of Negros, Philippines. Sampling was conducted from August 2011-July 2012, using butterfly traps set in the understory and canopy, along transects established through each habitat.  Altogether, 1526 butterflies were collected, representing 21 species. There were no significant differences in species richness, abundance or diversity between the habitats and there was little distinction in the community structure of frugivorous butterflies at each habitat. The distribution of proportional species abundances, ranked by range-size, showed that levels of precinction were significantly different between all habitats except old growth forest and secondary forest garden. Although additional study is required to fully assess the conservation value of Philippine secondary forests, these results provide evidence suggesting that the integration of secondary forests into conservation plans may be beneficial, at least to frugivorous butterfly communities.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lawrence E Reeves.
Thesis: Thesis (M.S.)--University of Florida, 2013.
Local: Adviser: Daniels, Jaret C.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2013
System ID: UFE0045620:00001


This item has the following downloads:


Full Text

PAGE 1

THE ECOLOGICAL IMPORTANCE OF SECONDARY FORESTS TO FRUGIVOROUS BUTTERFLY COMMUNITIES IN MOUNT KANLAON NATIONAL PARK, NEGROS, PHILIPPINES By LAWRENCE REEVES A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013 1

PAGE 2

2013 Lawrence Reeves 2

PAGE 3

To my grandmother, Henrietta Cavile, who made this project possible 3

PAGE 4

A CKNOWLEDGMENTS I am grateful for the guidance and support provided by my advisor, Dr. Jaret Daniels and my committee, Dr. Keith Willmott and Dr. Tom Emmel for their invaluable advice throughout the course of this project, from the initial planning of fieldwork through to preparation of the manuscript. In the Philippines, I thank the Department of Environment and Natural Resources for granting permission to collect specimens and conduct fieldwork. I am particularly grateful to the Region 6, Bacolod office a nd Angelo Bibar for advice and assistance in planning and executing fieldwork at Mount Kanlaon National Park. Fieldwork and data collection would not have been possible without the knowledge and expertise of Balmerie Villar. I also thank Bibing Romo, Boy E nguito, Ryan CaAgoy, Martin Monilla, Rey Estrelloso, Chris Johns and Jon Bremer for their assistance in the field. In addition, this project hinged on the support, logistical and emotional, of my family, the Cavile Clan, in Kabankalan City. Salamat gid sa tanan. I give particular thanks to, my grandmother, Henrietta Cavile, and Dayting Cavile, for all of the time they spent helping with logistics, organizing expeditions, and worrying about my safety. In the Philippines, Khorie Cavile was the foundation of all fieldwork, logistics and translation, navigating me through everything from permits to rainforests to ordering coffee. I am grateful for everything she has done to assist me in my work in the Philippines, and for her companionship along the way. I wou ld also like to thank the Dean of the College of Agriculture and Life Sciences at the University of Florida and the Natural Area Teaching Lab for providing funding via a teaching assistantship for the duration of this project. I would especially like to th ank Dr. Tom Walker and Dr. Jennifer Gillett Kaufman for their support and 4

PAGE 5

assistance. Fieldwork was supported by the Tropical Conservation and Development program at the University of Florida. This material is based upon work supported by the National Scie nce Foundation Graduate Research Fellowship. The McGuire Center for Lepidoptera and Biodiversity and the Florida Museum of Natural History also provided generous support at the outset and throughout the course of this project, and provided necessary space and equipment. I also thank Jonathan Colburn of the Florida Museum of Natural History for assistance with statistical analyses and Dr. Paul Choate for generous encouragement and long conversations. I thank my parents, Westley and Frances Reeves for their support and encouragement throughout the years and for launching me down a natural history trajectory. 5

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 8 LIST OF FIGUR ES .......................................................................................................... 9 LIST OF ABBREVIATIONS ........................................................................................... 10 ABSTRACT ................................................................................................................... 11 CHAPTER 1 GENERAL INTR ODUCTION .................................................................................. 13 Summary ................................................................................................................ 13 The Philippine Archipelago ..................................................................................... 16 Precinctive Species ................................................................................................ 20 Secondary Forests .................................................................................................. 23 Butterflies ................................................................................................................ 28 Synthesis ................................................................................................................ 32 2 COMMUNITY STRUCTURE, DIVERSITY AND VERTICAL STRATIFICATION .... 34 Introduction ............................................................................................................. 34 Methods .................................................................................................................. 40 Study Location .................................................................................................. 40 Study Group ..................................................................................................... 41 Transec t Sampling ........................................................................................... 42 Taxonomic Identification ................................................................................... 45 Statistical Methods ........................................................................................... 45 Results .................................................................................................................... 49 Community Diversity ........................................................................................ 49 Community Structure ........................................................................................ 51 Vertic al Stratification ......................................................................................... 52 Similarity Between Communities ...................................................................... 53 Discussion .............................................................................................................. 54 3 RANGE SIZE AND LEVELS OF PRECINCTION ................................................... 77 Introduction ............................................................................................................. 77 Methods .................................................................................................................. 82 Study Site and Sampling Methods ................................................................... 82 Taxonomic Identification and Geographic Distribution ..................................... 82 Statistical Methods ........................................................................................... 82 6

PAGE 7

Results .................................................................................................................... 85 Discussion .............................................................................................................. 87 4 GENERAL CONCLUSIONS ................................................................................. 102 LIST OF REFERENCES ............................................................................................. 105 BIOGRAPHICAL SKETCH .......................................................................................... 118 7

PAGE 8

LIST OF TABLES Table page 2 1 List and raw counts of species sampled by butterfly traps on each transect ...... 63 2 2 The five most abundant species on each transect, ranked by abundance ......... 63 2 3 Vertical stratification of butterflies captured by traps and results of exact binomial tests ...................................................................................................... 64 3 1 Species level precinc tion ranking scheme. ......................................................... 93 3 2 Subspecies level precinction ranking scheme. ................................................... 94 3 3 Species level results of MannWhitney U tests................................................... 94 3 4 Subspecies level results of MannWhitney U tests. ............................................ 94 3 5 Results of Fishers exact test comparing proportional abundance o f individual species between transects ................................................................................. 95 8

PAGE 9

LIST OF FIGURES Figure page 2 1 Study location, Negros, Visayas, Philippines ...................................................... 65 2 2 Remaining old growth forest tracts on Negros .................................................... 66 2 3 Sampling design. ................................................................................................ 67 2 4 Butterfl y trap. ...................................................................................................... 68 2 5 Example image of the method used to collect data ............................................ 69 2 6 Data collection .................................................................................................... 70 2 7 Coleman Species Rare faction Curve for each transect ...................................... 71 2 8 Proportional rank ab undance curve for each transect ........................................ 72 2 9 Log rank ab undance curve for each transect ..................................................... 73 2 10 Results of agglomerativ e cluster analysis by trap site ........................................ 74 2 11 Results of agglomerati ve cluster analysis by transect ....................................... 75 2 12 Results of nonmetric multid imensional scaling by trap site ............................... 76 3 1 Proportional abundance distribution of butterfly species, ranked by range size for Trans ect 1 (secondary forest garden) ........................................................... 96 3 2 Proportional abundance distribution of butterfly species, ranked by range size for Transect 2 (post extraction secondary forest) ............................................... 97 3 3 Proportional abundance distribution of butterfly species, ranked by range size for Transect 3 (old growth forest) ........................................................................ 98 3 4 Proportional abundance distribution of butterfly subspecies, ranked by range size for Trans ect 1 (secondary forest garden) .................................................... 99 3 5 Proportional abundance distribution of butterfly subspecies, ranked by range size for Transect 2 (post extraction seco ndary forest) ...................................... 100 3 6 Proportional abundance distribution of but terfly subspecies, ranked by range size for Transect 3 (old growth forest) .............................................................. 101 9

PAGE 10

LIST OF ABBREVIATIONS DENR Department of Environment and Natural Resources (Philippines) FAO Food and Agriculture Organization FMB Forest M anagement Bureau (Philippines) IUCN International Union for Conservation of Nature MKNP Mount Kanlaon National Park 10

PAGE 11

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Require ments for the Master of Science THE ECOLOGICAL IMPORTANCE OF SECONDARY FORESTS TO FRUGIVOROUS BUTTERFLY COMMUNITIES IN MOUNT KANLAON NATIONAL PARK, NEGROS, PHILIPPINES By Lawrence Reeves May 2013 Chair: Jaret Daniels Major: Entomology and Nemotology T he Philippine Archipelago is a global conservation priority. Extensive deforestation, driven largely by logging and agricultural conversion, has resulted in near complete loss of old growth forests. Concomitant with the rapid decline of old growth forests has been substantial increases in the land area occupied by secondary forests. It is possible that the formation of secondary forests could mitigate the effects of deforestation on forest species and play a role in augmenting existing conservation strategi es. Few studies have investigated the ecological importance of Philippine secondary forests. This study compares the diversity, community structure, vertical stratification and precinction levels of frugivorous butterfly communities occurring at secondary forest garden, post extraction secondary forest and old growth forest sites within a contiguous forest tract on the island of Negros, Philippines. Sampling was conducted from August 2011July 2012, using butterfly traps set in the understory and canopy, along transects established through each habitat. 11

PAGE 12

Altogether, 1526 butterflies were collected, representing 21 species. There were no significant differences in species richness, abundance or diversity between the habitats and there was little distinction in the community structure of frugivorous butterflies at each habitat. The distribution of proportional species abundances, ranked by rangesize, showed that levels of precinction were significantly different between all habitats except old growth forest and secondary forest garden. Although additional study is required to fully assess the conservation value of Philippine secondary forests, these results provide evidence suggesting that the integration of secondary forests into conservation plans may be ben eficial, at least to frugivorous butterfly communities. 12

PAGE 13

CHAPTER 1 GENERAL INTRODUCTION Summary Globally, the resources available to conservation organizations are sorely insufficient to respond to all threatened species and ecosystems. As a result, it is necessary to identify conservation priorities (Wilson et al. 2006). On a global scale, precinction (endemism) level and degree of habitat loss are widely used as criteria in prioritizing geographic locations for conservation efforts (Myers et al. 2000). B y these criteria, the Philippine Archipelago is a top global conservation priority (Brooks et al. 2002, Brummitt and Lughadha 2003, Ovadia 2003, Fonseca 2009). The archipelago supports a biota that is exceptionally species rich and which represents one of the greatest concentrations of precinctive species on the planet. Simultaneously, the Philippines has been subject to severe and extensive deforestation, with most estimates placing remaining original forest cover at less than 5% of the Philippines land area (Galang 2004), thus putting forest taxa in a precarious situation. Throughout tropical Asia, and much of the tropical belt, the loss of original forests to logging and agricultural conversion has been accompanied by an increase in area occupied by sec ondary forests (FAO 2010). Philippine forest cover has undergone a similar progression. Over the past century, original forest cover has dramatically decreased in area, primarily as a direct and indirect result of the commercial timber industry (Heaney and Regaldo 1998). As the Philippines forests are harvested of their timber, they are transformed to secondary forests or settled for agriculture. The product of this dynamic is the predominance of secondary forests in the makeup of the Philippines remaining forestlands. With secondary forests now accounting for 83% of 13

PAGE 14

the Philippines forestlands, human populations, in particular lower socioeconomic classes, are increasingly reliant on these forests (Lasco et al. 2001). Considering the extent of original f orest loss, these same forests may also be of considerable ecological importance to forest taxa and hold potential in mitigating the effects deforestation. The Philippines biota consists of a significant proportion of species unique to the archipelago. I nherently, these precinctive species, through occupying a limited geographic area are more vulnerable to global extinction than wideranging species. For this reason, precinctive species are often targets of conservation efforts. As natural habitats and ec osystems become increasingly degraded to support the growing human population, species extinctions in the Philippines are most likely to occur among species unable to adapt to a humanmodified landscape. Previous studies across a range of taxa and geographic locations find wideranging species better able to adapt to modified landscapes than limitedrange precinctives, which remain largely associated with pristine habitats (Thomas 1991, Fjeldsa 1999, Cleary and Genner 2004, Wijesinghe and Brook 2005, Charrette et al. 2006). The extent of original forest loss and the need to develop and implement effective conservation measures in the Philippines has made it necessary to foster an understanding of the habitats important to precinctive forest taxa. The abilit y of secondary forests to act as surrogate habitats for forest taxa is uncertain. As primary forests are lost, the resulting secondary forest often contains biotic elements that are relicts of the former old growth forest (Dent and Wright 2009). The preser vation of precinctive forest species is a conservation goal in the Philippines. Thus, an understanding of the species composition of secondary forest communities and the 14

PAGE 15

degree to which forest precinctives are represented and able to persist in these habit ats is required. This study uses the frugivorous guild of nymphalid butterflies to evaluate a secondary forest in Mount Kanlaon National Park for its ability to provide habitat for the precinctive component of the guild. Butterflies and other invertebrates are functionally important to tropical forest ecosystems (Wilson 1987). In these systems, invertebrates perform many of the roles critical to the maintenance of the ecosystem (e.g. pollination, decomposition, soil nutrient cycling). Despite their importance, the response of invertebrate communities to habitat disturbance and their importance to forest conservation has received little attention in comparison to vertebrates (Bossart and Carlton 2002). This is particularly true in the Philippines, where biodiversity conservation is primarily focused on mammals and birds. Of the Philippines invertebrate taxa, butterflies are the best known. As herbivores and pollinators, they are important components of Philippine forest ecosystems. In addition, butterflies are diverse and their communities are sensitive to changes in habitat quality (Posa and Sodhi 2006). Butterflies are also often used as biodiversity surrogates (Cleary 2004). As a result, butterflies have been used as ecological indicators in tropical fores t systems in the Neotropics, Africa and Asia (Kremen 1994, Hill and Hamer 1998, Howard et al. 1998, Brown and Freitas 2000). In the Philippines, frugivorous butterflies are abundant, relatively diverse taxonomically tractable, sensitive to the effects of disturbance and utilize a diversity of life history strategies, thus, ideally suited for assessments of habitat quality or the ecological importance in habitats. In this study, the guild is used as a study taxon to evaluate the ecological importance of a secondary forest in Mount Kanlaon National 15

PAGE 16

Park. Precinctive species are a greater conservation concern than widespread species. The objective of this study is to determine the ecological importance of this secondary forest to precinctive members of the guil d and to investigate the effects of forest disturbance on the diversity and species richness of frugivorous butterflies. The Philippine Archipelago The Republic of the Philippines consists of an archipelago of more than 7,000 islands situated in the western Pacific Ocean. Combined, the Philippine Archipelago occupies a land area of approximately 300,000 km2 (similar in size to the US state of Arizona). The ten largest islands account for close to 93% of the archipelagos total land area. The archipelago has undergone a complex geologic history, and is made up of both continental and oceanic islands, some of which have been sporadically connected to each other or other landmasses during periods of lowered sea levels (Vallejo 2011). The geologic history of t he Philippines has been the driving force in the formation of the countrys high species diversity and faunal patterns observed today. The distribution patterns of Philippine butterflies and mammals suggest the archipelago consists of six faunal regions: G reater Luzon, Greater Mindanao, Mindoro, Greater Palawan, Sulu and Greater Negros Panay (VaneWright 1990). Each of these biogeographic units corresponds to an island formed during a period of lowered sea level in the Pleistocene. Biotically, the Philippine Archipelago has been influenced by, and contains elements of Sundaland, Wallacea, New Guinea and Australia (Vallejo 2011). Over time, the Philippines faunal regions have had various interactions with these neighboring landmasses. The result of the Phili ppines geologic history and archipelagic nature has been the formation of a biota rich in species diversity and precinction. The archipelago supports over 900 terrestrial vertebrates and between 16

PAGE 17

10,000 and 14,000 plant species (Galang 2004), qualifying the Philippines as a country of megadiversity, one of seventeen countries recognized by Conservation International for their high levels of species richness. The Philippines is also recognized for the number of species within its biota that are found nowhere else in the world. Of the countrys 900 terrestrial vertebrates and 10,00014,000 plant species, approximately 50% are precinctive to the Philippines (Heaney and Regaldo 1998) making it one of the 25 identified global biodiversity hotspots (Myers 2000). Biodiversity hotspots are defined not only by exceptional concentrations of precinctive species, but also by the presence of an exceptional degree of habitat loss. This combination identifies regions of the world that are both biologically irreplaceabl e and vulnerable to habitat loss. The Philippine Archipelago repeatedly ranks high in terms of precinctive species and degree of habitat loss (Myers 2000, Brooks et al. 2000, Brummitt and Lughadha 2003, Ovadia 2003, Fonseca 2009). As a result, the archipel ago represents one of the highest conservation priorities in the world. Rapid and widespread deforestation has placed the Philippine Archipelago at the forefront of the current biodiversity crisis (Myers 1993). Prior to Spanish colonialism in the 1500s, f orests were the predominant land cover on the archipelago, occupying more than 90% of the Philippines land area (Bankoff 2007). In addition to supporting a diverse and unique biota, these forests represented a valuable natural resource to the Philippine e conomy as well as local human populations. Over the past century, unsustainable exploitation of forest resources has resulted in a dramatic reduction in the land area occupied by original forests. At the end of Spanish colonialism, approximately 70% of the Philippines remained forested (Bankoff 2007). During the American and 17

PAGE 18

Japanese occupations of the Philippines, forest cover was further reduced to approximately 50% (Kummer 1992). Following World War II, the percentage of the Philippines land area occupi ed by forests continuously decreased, with the most rapid declines observed during the presidency of Ferdinand Marcos (Vitug 1998). Current estimates place forest cover at 20% of the Philippines land area. Of this, the land area covered by closed canopy forests is estimated at 2.5 million hectares or 8% of the Philippines total land area (FMB 2010). Furthermore, the extent of unlogged, original forest cover is estimated to be as low as <1 million hectares or <5% of the total land area (Lasco et al. 2001). Deforestation in the Philippines is largely driven by logging and agricultural conversion, working in concert (Suarez and Sajise 2010). Tree species in the family Dipterocarpaceae are a major component of most Philippine forests. Until recently, the Philippines 50 dipterocarp species, often referred to as Philippine mahogany, formed the basis of the Philippines timber industry (Heaney and Regaldo 1998) and were of exceptionally high monetary value in the global market (Kummer and Turner 1994). Logging activities in forested areas targeting these species creates post extraction secondary forests that facilitate further human intrusion (Lasco et al. 2001). Logging roads built to extract the valuable tree species from a forest make convenient inroads for h uman migration, while post extraction secondary forests, lacking large trees, are easily cleared for agriculture. Rapid population growth, coupled with high levels of poverty in the Philippines rural areas draw large numbers of subsistence farmers, or kai ngeros to postextraction secondary forests. The Tagalog word kaingin refers to upland agriculture, often shifting or slashand burn agriculture (Lawrence 1997). Aside 18

PAGE 19

from the practice of kaingin by indigenous groups, kaingeros are often landless, migr ating to loggedover areas out of socioeconomic necessity, where they set up shifting agricultural plots, or in areas of high population density, permanent agricultural plots. Socioeconomic inequality and political factors have been the underlying forces contributing to deforestation in the Philippines. In addition to the promotion of kaingin by the prevalence of poverty in rural areas, an unequal distribution of the economic rewards generated by logging furthers these issues. During the post war period, l ogging rights were routinely granted to a small number of wealthy or powerful individuals belonging to an elite socioeconomic class or used as political spoils. As a result, capital generated by logging directly benefited these individuals and did not proc eed to the greater community (Galang 2004). Corruption within the government, particularly during the Marcos Regime, has allowed control of forest resources to remain concentrated among the elite while enabling logging activities to progress virtually unregulated. During the Marcos presidency (19691985), deforestation proceeded at the highest rates observed (Vitug 1998). Following his overthrow, government management of forest resources underwent a significant restructuring, a moratorium on logging in prim ary forests was put in place and an environmental consciousness emerged in sections of Filipino society (Posa et al. 2008). While these changes help to mitigate habitat loss, illegal logging and the resource requirements of a rapidly increasing human popul ation place continued pressure on remaining forestlands. Considering the extent of deforestation and the ongoing threats to Philippine forests, it is crucial that the Philippines receive increased attention from the global 19

PAGE 20

conservation community. There rem ain significant gaps in our understanding of the Philippines terrestrial biota and its response to widespread habitat modification. Even basic species inventories are incomplete, with new taxa, including large vertebrate species, being described with rema rkable frequency (e.g. Allen et al. 2004, Welton et al. 2010). In addition, previous studies on Philippine taxa and the effects of habitat disturbance have largely focused on vertebrates, particularly mammals and birds. In tropical forests, invertebrates are responsible for many of the key processes and services that maintain the ecosystem (Wilson 1987). Despite this, few studies have investigated how invertebrate communities respond to habitat changes in the Philippines (Alcala 2004). The coupling of an exceptional biota with extensive habitat loss makes the Philippines one of the greatest conservation priorities in the world. The causes of habitat loss in the Philippines are firmly entrenched by socioeconomic and political factors. Despite this, the count ry is undergoing significant changes at the governmental and societal levels which promote conservation of biodiversity and habitats and the country is receiving increased attention from the global conservation community (Posa et al. 2008). In order to mit igate the effects of largescale habitat modification, it is critical that these trends continue. Precinctive Species In biogeography, the term endemic species is often applied to describe species that are naturally confined to a particular geographic loc ation. However, endemic has contrasting meanings in various fields, in particular, ecology and epidemiology (Anderson 1994). Frank and McCoy (1990) list four conflicting definitions of endemic and promote the use of the alternate, more valid term, precinct ive In order to 20

PAGE 21

disambiguate these meanings, precinctive is used here, in place of endemic to describe species that are confined to the area under discussion (Frank and McCoy 1990). Furthermore, there is conflict in the methods used to recognize a species as precinctive. Species have been classified as precinctive based on their occurrence only within a particular geographic area or the small area of their geographic range (Peterson and Watson 1998). Here, precinctive is used to describe species that naturally occur only in a particular geographic region and rangerestricted is used to describe species that occupy a small geographic area. Previously widespread species that have experienced a range contraction are not considered here as precinctive or rangerestricted. Richness of precinctive species can be used to quantify biotic irreplaceability in a geographic area. The presence of high concentrations of precinctive or rangerestricted species is one of the criteria most often considered when assigning g eographic conservation priorities (Mittermeier et al. 1998, Myers et al. 2000). Precinctives are often conservation targets because they are intrinsically more vulnerable to extinction than cosmopolitan species. Precinctives and rangerestricted species occupy small geographic areas and, as a result, local extirpation is more likely to lead to global extinction. Numerous studies investigating ecological correlates of extinction risk across a variety of taxa find small rangesize to be significantly associat ed with increased vulnerability to extinction (Purvis et al. 2000, Jones et al. 2001, Koh et al. 2004). As a result, rangesize is often used to aid in the determination of a species conservation status (Millsap et al. 1990, IUCN 2010). In addition, evidence from the fossil record suggests increased extinction proneness among rangerestricted species and 21

PAGE 22

contemporary anthropogenic disturbances to habitats are likely to adversely affect precinctives (Jablonski 1991). Worldwide, anthropogenic disturbance of habitats has profound effects on the distribution of species. McKinney and Lockwood (1999) suggest a global trend towards biotic homogenization, in which cosmopolitan generalists replace precinctive specialists in human altered habitats. A general feature of precinctive and rangerestricted species is their tendency to be ecological specialists (Kitahara and Fujii 1994, Harcourt et al. 2002), while their widespread, cosmopolitan congeners tend to be ecological generalists (Brown 1984). The degree of ecolog ical specialization within a species can affect its response to habitat disturbances (Devictor et al. 2008). Ecological generalists tend to be more resilient to habitat disturbances and are more likely to benefit from disturbances than specialist species. Conversely, specialists are often more sensitive to disturbance and are more likely to be adversely affected (e.g. Kotze and OHara 2003, Munday 2004, Juliard et al. 2006, Kadlec et al. 2009). It follows that precinctive and rangerestricted species are at greater risk of extirpation following disturbance events. Studies investigating a range of taxa and locations have observed precinctive and rangerestricted species to be largely absent or rare in disturbed areas, where they are replaced by cosmopolitan s pecies (Thomas 1991, Stephenson 1993, Lewis et al. 1998, Fjeldsa 1999, Wijesinghe and Brook 2005). In these studies, precinctive and rangerestricted species were largely associated with pristine habitats and absent from disturbed habitats. Further, evidence from Malaysian butterflies suggest that precinctive and rangerestricted species are less likely to recover than widespread species following major disturbances to habitat (Charrette et al. 2006). Thus, human modified 22

PAGE 23

and disturbed habitats and ecosystems tend to be dominated by widespread cosmopolitan species, thereby promoting biotic homogenization. Precinctive and rangerestricted species are often the targets of conservation initiatives because their small rangesize makes them inherently more vulnerable to global extinction. Compared to cosmopolitan species, precinctives are particularly sensitive to the effects of habitat modification. Globally, the impacts of human activities on natural habitats are increasing, resulting in range contractions in p recinctive species (McKinney and Lockwood 1999, Olden et al. 2004). In terms of precinctive species and degree of habitat loss, the Philippine Archipelago is consistently recognized as a top biodiversity hotspot. There, exceptional levels of precinction oc cur alongside extensive habitat loss. The Philippines forest cover has decreased dramatically over the past century. Currently, only 20% of the archipelagos land area is classified as forest cover (FMB 2010), with original forest occupying less than 5%. The majority of land classified as forest cover represents brushlands, grasslands and disturbed and secondary forests of unknown value to the Philippines precinctive species. Secondary Forests The decline of old growth forests throughout the tropical bel t coincides with an increase in restructured forest types that are often described as secondary forest. The term secondary forest has been used with much ambiguity. Here, secondary forest will be defined after Chokkalingam and de Jong (2001) as forests that are regenerating largely through natural processes after significant human and/or natural disturbance of the original forest vegetation at a single point in time or over an extended period, and displaying a major difference in forest structure and/or canopy species composition with respect to nearby primary forests on similar sites. This definition can be further refined 23

PAGE 24

to describe types of secondary forests that have formed as the result of a particular type of disturbance. The types of secondary forest most relevant to this work, as defined by Chokkalingam and de Jong (2001), are post extraction secondary forests (secondary forests that have formed following tree extraction e.g. logging activities), swidden fallow secondary forests (cultivated land allowed to naturally regenerate for the purpose of restoring land for cultivation, following complete, or near complete loss of all forest communities ), secondary forest gardens (secondary forest that receives low levels of management and contains a planted component), post abandonment secondary forest (secondary forest that has formed following the abandonment of an alternative land use e.g. agriculture or pasture) and rehabilitated secondary forest (secondary forests occurring on degraded lands, often s upported by rehabilitation efforts). Anthropogenic deforestation and forest conversion in Earths tropical regions have pronounced effects that manifest at the global and landscape level (Fearnside and Laurance 2004). At a landscape scale, large areas of old growth forest have rapidly been deforested and converted to agriculture, pasture or secondary forests. Current estimates suggest that 3550% of the planets tropical old growth forests have been lost (Wright and Muller Landau 2006), with global annual deforestation rates in old growth forests remaining high at .4% (FAO 2010). Many tropical countries have experienced significant decreases in the land area occupied by mature forest cover, coinciding with increases in secondary vegetation (Bhat et al. 2001, Lasco et al. 2001, Perz and Skole 2003, Wright 2005, FAO 2010) and it is estimated that more than 50% of all tropical forests are secondary forests (FAO 2010, Wright 2010). In developed countries, a process of forest transition has been observed as a country undergoes economic 24

PAGE 25

expansion (Rudel et al. 2005). Under this process, a country experiences a net loss in forest cover through deforestation, followed by a period of reforestation and a net increase in secondary forest cover As economies develop, the migration of people from rural to urban areas coincides with a decrease in agriculture, and thus, formation of post abandonment secondary forests as neglected agricultural lands undergo succession (Mather 1992). This trend has been observed in many developed European countries, as well as China, India and Vietnam (Mather 2001, Mather 2007) and is expected to continue as the economies of developing countries expand (Guariguata and Ostertag 2001). Currently, secondary forests account for 57% of the worlds forests and are the principal component of forest cover in the majority of tropical countries (FAO 2010). As the ratio of primary to secondary forest cover shifts, local human populations become increasingly reliant on secondary forests as a source of forest products. Because secondary forests hold value to humans (Brown and Lugo 1990), they often remain dynamic and receive sustained human disturbance (e.g. swidden fallow secondary forests and secondary forest gardens). Similarly, secondary forests may have significant ecological importance and conservation value. Habitat loss in the form of tropical deforestation is among the greatest current threats to biodiversity (e.g. Castelletta et al. 2000). As the area covered by old growth forest declines and second ary forest expands, it is possible that the formation of secondary forests could mitigate the effects of tropical deforestation on biodiversity by providing habitat for displaced forest taxa (Wright and Muller Landau 2006, Bowen et al. 2007, Tabarelli et al. 2010, Pinotti et al. 2012). A review of 65 studies investigating the similarity of faunal communities between 25

PAGE 26

secondary and old growth forests reports that numerous forest species are able to persist in secondary forests (Dent and Wright 2009). Following a significant disturbance, a secondary forest undergoes succession as it naturally regenerates toward a mature state. Throughout this process, the forest undergoes structural, functional and compositional transitions. Initially, a disturbed forest receiv es an influx of generalist species while forest specialists remain absent (Pardini et al. 2009). The colonization of a disturbed forest by these generalist species can allow the species richness of a site to rapidly return to old growth levels. In the Neot ropics, species richness of trees and some animal taxa has been observed to recover in as little as 2040 years ( Guariguata and Ostertag 2001). However, recovery of precinction levels and species composition can take significantly longer, up to hundreds of years (Dunn 2004). Animal species have been observed to recolonize disturbed sites at species specific stages of succession. Over time, the habitat structure of a disturbed site regenerates towards maturity and habitat specialists and precinctive species recover (Pinotti et al. 2012). This process can be hastened if the disturbed site is in close proximity to a mature site, which acts as a source of recolonization (Chazdon 2003). Because the factors that drive succession and recolonization of forests sp ecies (disturbance history, proximity to source habitats, etc.), particular forests are idiosyncratic in their rate of recovery towards a natural state. Although secondary forests do not hold the conservation value of old growth forests, they may represent a significant opportunity to enhance conservation efforts in countries that have experienced substantial forest loss, particularly if managed for this purpose (Horner Devine et al. 2003, Gardner et al. 2009, Berry et al. 2010, Edwards et al. 2011). 26

PAGE 27

In th e Philippines, extensive logging and agricultural expansion over the past century has transformed the archipelagos landscape. Historically, old growth rainforests were the dominant land cover and occupied more than 90% of the Philippines land area (Liu et al. 1993). Currently, the Philippines Forest Management Bureau classifies only 6 million ha (20%) of the countrys land area as forest (FMB 2010) Of the land classified as forest, less than 1 million ha (17%) represent old growth forest, while 5 million ha (83%) consist of various types of secondary forest, primarily post extraction and swiddenfallow secondary forests (Lasco et al. 2001). Thus, secondary forests are now the dominant forest type in the Philippines. In 1992 the Philippine government banned all logging in primary forests (Posa et al. 2008). As a result, secondary forests have become the Philippines primary source of wood and other forest products, making them vulnerable to additional disturbance (Lasco et al. 2001). Currently, most conser vation strategies in the Philippines focus on remnant tracts of old growth forest. Protection of these sites is critical, but if secondary forests hold conservation value to the Philippines precinctive species, the development of conservation strategies t hat integrate secondary forests may also provide a worthwhile opportunity for the conservation of threatened forest taxa. Unfortunately, the role these forests may play in providing habitat for the Philippines precinctive species is poorly understood. Few studies have investigated whether these forests are able to mitigate the effects of deforestation on precinctive species. Considering the extent of deforestation that the Philippines has experienced, the use of secondary forests as a conservation tool could bolster efforts to protect the Philippines threatened biodiversity. 27

PAGE 28

Butterflies In terms of diversity and abundance, insects and other invertebrates are the dominant animal taxon in most biomes (Wilson 1987). Insects hold functional roles in most terr estrial habitats and directly interact with many ecosystem components (Losey and Vaughan 2006). Despite this, they are rarely considered as conservation targets. Over the past several decades, increased attention has been given to the conservation of insec ts and other arthropods (Pyle et al. 1981, New 1993, New 1999, Black et al. 2001), but implementation of conservation measures considering insect groups remains limited (Bossart and Carlton 2002). Increased attention from the conservation community is required in order to protect imperiled insect species and the ecological functions they perform. In the past 600 years, 70 insect extinctions have been documented, with a further 40,000 estimated to have occurred (Dunn 2005). Because of the close ecological as sociations between insects and their ecosystems, conservation measures involving these communities can be beneficial because they often promote an ecosystem or habitat level approach to conservation and work to preserve ecological functions (Miller 1993). Further, the close associations of insects to other elements within an ecosystem often make them sensitive to changes in habitat quality. As a result, insect communities can be useful in assessments of habitat quality or ecosystem health (Brown 1997, Hughes et al. 2000, Pearce and Venier 2006). Further, studies of insect assemblages have been used to evaluate a wide range of anthropogenic impacts including global warming, habitat degradation and fragmentation, pollution and success of habitat restoration (e.g. Rosenberg et al. 1986, Warren et al. 2001, Gibb and Hochuli 2002, Colwell et al. 2008, Miller et al. 2010, Edwards et al. 2012). 28

PAGE 29

Butterflies have played a prominent role in invertebrate conservation. Of the insects, butterflies are an ideal flagship t axon, or a taxon that serves to catalyze conservation actions, and have been used around the world to promote conservation programs (New 2011). They are one of the most charismatic insect groups and possess significant appeal among the general public. As a result, species specific conservation initiatives involving butterflies often receive public support and generate further interest in conservation (New et al. 1995). Owing to their aesthetic appeal, butterflies, and Lepidoptera in general, have received attention from the scientific community for hundreds of years and are one of the most well studied insect orders. Compared to other insect taxa, the ecology, taxonomy and biogeography of butterflies, particularly in temperate regions, are fairly well known (Thomas 2005). Lepidoptera also have significant ecological importance as pollinators and herbivores and are often sensitive to changes in habitat quality. For these reasons, butterflies are a practical study group for investigations on the effects of anthropogenic habitat modifications on insect communities and may be a useful surrogate group, indicative of the effects of habitat loss to other invertebrate or vertebrate taxa. In the tropics, butterfly assemblag es are readily sampled by fruit or carrion baited traps or insect nets. Because butterflies are diverse and easily studied, they are frequently used to investigate the effects of disturbance on biodiversity (Shahabuddin et al. 2000, Bonebrake 2010) and have been used with varying success as biological indicators of ecosystem health (Kremen 1994, Brown and Freitas 2000). Tropical butterfly assemblages respond rapidly to habitat disturbance and changes in environmental quality. Habitat disturbance has a variety of effects on butterfly 29

PAGE 30

communities that c an manifest as changes in species composition, diversity or abundance, shifts in the vertical stratification of butterflies and changes in precinction levels within a community. Throughout the tropical belt, there has been a proliferation of studies invest igating the changes in species richness and community abundance of tropical butterflies in response to anthropogenic forest disturbance (e.g. Brown and Freitas 2000, Shahabuddin and Ponte 2005, Molleman et al. 2006). Despite this, there is little consensus on the positive and negative effects of disturbance on richness and abundance within butterfly communities, with studies reporting increasing and decreasing richness and abundance with disturbance at similar frequencies. These discrepancies are likely the result of the various spatial scales used in individual investigations. Hamer and Hill (2000) report that measurements of the effects of disturbance on butterfly communities are dependent on spatial scale. Further, they found that studies over smaller spatial scales were more likely to find greater species richness and abundance within disturbed sites than undisturbed sites and studies over larger spatial scales were more likely to observe the opposite. Koh (2007) emphasizes the importance of explicitly reporting the spatial scale utilized in studies on the response of butterfly communities to disturbance. In tropical forests, habitat disturbance alters the availability of certain microhabitats. In particular, removal of large trees disrupts the vertical st ratification of butterfly communities (Fermon et al. 2005). Logging negatively affects understory butterflies by eliminating their preferred microhabitats and brings canopy specialists closer to ground level. In addition, logged forests often contain a hig her abundance of species that are gap specialists. Although disturbed species may contain high levels of diversity, the dominant species often tend to be widespread 30

PAGE 31

generalists. Thus, forest disturbances can decrease precinction levels within butterfly com munities. Numerous studies (e.g. Thomas 1991, Hamer et al. 1997, Spitzer et al. 1997, Ghazoul 2002, Lien and Yuan 2003, Dumbrell and Hill 2005, Fermon et al. 2005) have found that butterflies with small geographic distributions are largely intolerant of habitat modification and widespread, generalist species tend to dominate disturbed habitats, indicating that precinctive species may be particularly sensitive to the effects of habitat disturbance. The Philippine Archipelago contains a diverse biota that exhibits exceptional levels of precinction. Among butterflies, 927 species and 939 subspecies have been described, with 41% of all species precinctive to the archipelago (Treadaway 2012). In addition, 133 taxa are recognized as conservation dependent (Daniels en and Treadaway 2004). Over the past century, the country has undergone dramatic deforestation and landscape modification. Few studies have investigated the effects of this widespread habitat loss on butterflies and other insects. One exception is Posa and Sodhi (2006), who observed that butterfly communities at a site on the island of Luzon exhibit decreased diversity in humandominated landscapes and that precinctive species are less tolerant of disturbed habitats than widespread species. They also found the butterfly assemblages of open and closed canopy forests to be similar, supporting the idea that secondary forests hold potential conservation value to precinctive species. Koh (2007) highlights the need for increased attention to Southeast Asian butterflies and how they respond to landuse changes. This is particularly true in the Philippines, where the landscape has been transformed by human activities and high levels of precinctive species are vulnerable to extirpation and extinction. 31

PAGE 32

Furthermore, butterflies make an ideal study taxon for investigations on how habitat loss affects the Philippines insect fauna. In the Philippines, butterflies are diverse, and of the countrys insects, butterflies are most well known and thus, taxonomically tractable. Synthesis The Philippine Archipelago supports a diverse biota that exhibits exceptional levels of precinction. Over the past century, the islands have experienced massive and widespread deforestation. Originally almost entirely covered by forests, logging and agricultural expansion has reduced old growth forest cover by more than 95%. Concomitantly, the land area occupied by various types of secondary forest has expanded dramatically. The formation of secondary forests may provide an opportunity to mitigate the effects of widespread landuse change on displaced forest taxa. Following a disturbance, a forest is able to retain components of old growth communities. As the forest ages, biotic communities recover and eventually resemble those of old growth forest s. Thus, the conservation value of secondary forests increases with time. The Philippines precinctive species are of particular concern in the development of conservation strategies in the country. Precinctive species are known to be particularly sensitiv e to changes in habitat quality and are often absent from modified habitats. In order to hold conservation value, secondary forests in the Philippines must provide habitat for the precinctive component of the Philippine biota. The conservation potential of secondary forests in the Philippines remain poorly known and further study is required to assess the value of these habitats to conservation efforts. In the Philippines, few studies have investigated the response of insect communities to landuse changes Butterflies are an ideal study taxon because they are they are diverse, taxonomically tractable, readily sampled, ecologically important and 32

PAGE 33

sensitive to changes in habitat quality. This study investigates the frugivorous butterfly assemblages at three f orested sites in Mount Kanlaon National Park on the island of Negros, Philippines. The study sites differ by level of disturbance and are located in a secondary forest garden, a post extraction forest and an old growth forest. The objectives of this study are to compare differences in frugivorous butterfly community structure between the study sites and to determine if precinction levels within each community differ. 33

PAGE 34

CHAPTER 2 COMMUNITY STRUCTURE, DIVERSITY AND VERTICAL STRATIFICATION Introduction Acros s the tropical belt, forest cover is decreasing at a rapid and alarming rate and the loss of tropical forests around the world, along with its compounding effects on climate change, is one of the greatest threats facing biodiversity today (Laurance 1999). The ecosystems supported by these forests contain the overwhelming majority of the planets biological diversity (Myers 1988). The loss and degradation of tropical forests have significant implications for these ecosystems and is believed to be driving a contemporary mass extinction event (Laurance 2006). The Philippine Archipelago has experienced massive and widespread deforestation leading t o the near complete loss of old growth forests on most major islands (Heaney and Regaldo 1998). Current government s ponsored conservation strategies concentrate primarily on pres erving remnant fragments of old growth forest. Considering the extent of deforestation in the Philippines, it may be practical to include secondary forests in future conservation planning. Further, as rural human populations increasingly depend on secondary forests as a source of wood, charcoal and other forest products, these forests remain dynamic and susceptible to further degradation (Lasco et al. 2001). Elsewhere in the tropics, secondary forests have been observed to hold considerable conservation value (Horner Devine et al. 2003, Bowen et al. 2007, Berry et al. 2010, Edwards et al. 2010). In the Philippines, the conservation value of secondary forests is unknown, but has potential in mitiga ting the effects of deforestation on forest species. Conservation biology, faced with a dearth of resources, is responsible for developing optimal conservation strategies to preserve imperiled species, ecosystems 34

PAGE 35

and habitats. In order to quickly and affordably assess the conservation value or health of a habitat or ecosystem, the use of environmental indicators (a specific taxon or limited set of taxa) has been proposed. Communities of certain taxa are particularly sensitive to changes in habitat health a nd can provide a reflection of the impact of a disturbance on an ecosystem (McGeoch 1998). Data generated by studies on these taxa can provide insight into the general health or quality of the habitat or the broader response of a community to a disturbance at a local scale. This approach is especially useful in complex ecosystems, such as those of tropical forests, where it is not feasible to rapidly cultivate an extensive understanding of the locations ecology, and landscape and population dynamics. Howev er, identification of ideal indicator taxa can be problematic. In particular, there is often limited congruence between the response of an indicator taxon and other taxa to changes in habitat quality (Prendergast and Eversham 1997). Despite this and consi dering the magnitude of the challenges facing tropical forest conservation, studies examining or furthering the effective use of potential indicator taxa, especially in cases where the indicator is also a conservation target, are practical and worthwhile ( Devries et al. 1997). A number of insect taxa, along with several vertebrate groups are often proposed as environmental indicators in tropical forests (e.g. Andersen et al. 2002, Rainio and Niemela 2003, Fernandez et al. 2005). Several insect groups make effective environmental indicators because they are diverse and abundant, reproduce rapidly and thus, their populations often respond quickly to changes in environmental quality. In tropical forests, butterfly assemblages are ideal environmental indicators because they 35

PAGE 36

are often the most well known insect group, represent a variety of lifehistory strategies and can be conveniently and reliably sampled (Brown and Freitas 2000). The Philippine Archipelago contains a diverse butterfly fauna. Currently, there are 927 described species, of which approximately 40% are precinctive to the Philippines (Treadaway 2012). Butterflies are the most well known insect group in the Philippines. Despite this, few studies have investigated the group beyond basic taxonomic an d biogeographical descriptions (VaneWright 1990). There exist significant gaps in our understanding of how butterfly assemblages have responded to the Philippines widespread landuse changes. This is particularly unfortunate because of the potential util ity of butterflies and other insects as environmental indicators of habitat quality. Further, it is expected that as the Philippines remnant old growth forests decrease in area, a large proportion of the countrys butterfly fauna will be at risk of extinc tion (de Jong and Treadaway 1993). Danielsen and Treadaway (2004) recognize 133 species as conservation dependent, of which, 70% occur on lands that are not established conservation sites. Thus, a large number of the Philippines butterfly fauna are conser vation targets and studies investigating their response, and the response of butterfly communities in general, to disturbances in habitat quality are sorely needed in order to develop and implement sound conservation strategies for threatened species or understand the potential of butterflies as environmental indicators in Philippine forests. Forest disturbances have significant effects on the dynamics of butterfly communities (Hamer et al. 1997, Hill 2001, Lewis 2001, Benedick et al. 2006, UeharaPrado et al. 2007). Though few studies have investigated these effects in Philippine forests, Posa and Sodhi (2006) observed butterfly communities in closed and open 36

PAGE 37

canopy forests to be similar, but dramatically different from those of modified habitats. Similar studies conducted in the forests of nearby Southeast Asian countries suggest that butterfly communities in this region experience notable detrimental effects following disturbances to forest structure (Bowman et al 1990, Hill et al. 1995, Dumbrell and Hill 2005). These effects range from shifts in diversity and abundance, changes in species composition, extirpation to disruption of vertical stratification. Forest disturbance is often followed by distinct changes in the diversity and abundance of species in butterfly communities (Hill and Hamer 1998). Within a Southeast Asian forest, butterfly communities consist of species with strong preferences for particular microhabitats within the forests structure (e.g. understory, canopy, forest gaps). Disturbance al ters the availability of these microhabitats, thereby influencing shifts in butterfly communities (Lien and Yuan 2003). Potentially due to sampling biases, little congruence has been observed in the effect of disturbance on butterfly diversity and abundanc e, with studies reporting both positive and negative effects, as well as no effect (Hamer and Hill 2000, Koh 2007). Butterfly communities of tropical rainforests are vertically stratified (Molleman et al. 2006). Species composition of forest canopies is m ore similar to those found in disturbed areas than to those of the understory and following a disturbance, species that typically occupy the canopy of mature forests may move towards the understory (Fermon et al. 2005). Many of the studies that have found disturbance to have a positive effect on diversity and abundance have failed to sample forest canopies (Koh 2007). Thus, the movement of canopy species to the understory may artificially increase diversity and abundance within disturbed sites in studies that do not account for the 37

PAGE 38

vertical stratification of tropical forest butterflies. Despite the lack of congruence in studies on the effects of disturbance on butterflies, studies on the structure of butterfly communities in tropical forests can provide insi ghts into the overall quality of a habitat (DeVries 1997, Uehara Prado 2009). This study investigates the frugivorous butterfly communities of two types of secondary forest and an old growth forest in Mount Kanlaon National Park on the island of Negros, Philippines. In this study, attributes of the frugivorous butterfly communities of each forest type are compared against each other. Thirty years prior to the study period, two of the three forest types sampled experienced a significant disturbance to forest structure (i.e. timber extraction). Significant disturb ance to the structure of an old growth forest, in the absence of further modification (e.g. conversion to agricultural land) is followed by the formation of secondary forest. Secondary forests are d efined here after Chokkalingam and de Jong (2001) as a forest that is regenerating largely through natural processes after significant human and/or natural disturbance of the original forest vegetation at a single point in time or over an extended period, and displaying a major difference in forest structure and/or canopy species composition with respect to nearby primary forests on similar sites. In terms of diversity and abundance, forest butterfly communities in Southeast Asia that occur in secondary f orests have been obs erved to differ from nearby old growth sites (Willott 2000, Ghazoul 2002, Schulze et al. 2004, Veddeler 2005). In butterflies and other taxa, secondary forests can support at least a subset of the species found in old growth forests (Dent and Wright 2009). Thus, secondary forests may hold conservation value and their integration into protected areas could augment 38

PAGE 39

current conservation practices (Horner Devine et al. 2003, Gardner et al. 2009, Berry et al. 2010, Edwards et al. 2010). This study uses exploratory methods to compare the community structure of frugivorous nymphalid butterflies in three neighboring forested sites of Mount Kanlaon National Park on the island of Negros, Philippines. The sites occur within three forest types: a se condary forest garden, post extraction secondary forest and a near old growth forest, all part of a larger contiguous forest tract. The objective of this study is to determine if there are significant differences in the community structure, vertical strati fication, diversity, richness and abundance of the frugivorous butterfly communities occurring at each site. Concomitant with a sharp reduction in oldgrowth forest cover, has been the formation of a large area of secondary forests (Lasco et al 2001). As the Philippines forest cover continues to transition from old growth to secondary, the relevance of secondary forests to the conservation of the Philippines forest communities will increase. Currently, alarmingly little is known about the response of Phi lippine forest taxa to anthropogenic habitat disturbances. This study seeks to contribute information on the ability of secondary forests to mitigate the effects of deforestation on the Philippines forest taxa. Information derived from studies on the comm unity level responses of forest taxa to disturbance is important to the evaluation of the conservation potential of secondary forests in the Philippines. Furthermore, butterflies are well recognized as indicators of environmental quality. Although beyond t he scope of this project, changes in the community structure of butterflies may be indicative of general habitat health and offer insight into the response of other taxa to forest disturbance. 39

PAGE 40

Methods Study Location The Philippine Archipelago is situated in the western Pacific Ocean, north of Sulawesi and south of Taiwan. The archipelago consists of more than 7000 islands, many of which are small and uninhabited. Together, the Philippine Islands occupy a land area of approximately 300,000 km2 (similar in s ize to the US state of Arizona). The ten largest Philippine islands account for 93% of the countrys total land area. The archipelago has experienced a complex geologic history and consists of islands of both continental and oceanic origin (Vallejo 2011). Biogeographically, six faunal regions are recognized that correspond with land connections during periods of lowered sea levels in the Pleistocene (VaneWright 1990). The Philippines faunal regions are Greater Luzon, Greater Mindanao, Mindoro, Greater Pal awan, Sulu and Greater Negros Panay. Each of these biogeographic units corresponds with an island formed during the Pleistocene. The study took place on the island of Negros in the Visayan Region of the central Philippines (Fig. 2 1). Negros occupies a lan d area of 13,328 km2 and is the 4th largest island in the archipelago. The island is bisected by a mountain range that stretches from the northeast corner to the southern tip and divides the island into two provinces, Negros Occidental and Negros Oriental. Negros belongs to the Greater Negros Panay faunal region, which also includes the islands of Cebu, Panay and Masbate (VaneWright 1990). The majority of the Philippines sugarcane production takes place in the Visayas, with Negros producing more sugar than any other Philippine island. As a result, Negros has lost the vast majority of its forest cover to sugarcane plantations and other forms of agriculture. Negros has followed deforestation patterns that are typical of 40

PAGE 41

the Philippines (Heaney and Regaldo 1998). Historically, Negros was entirely covered by old growth forests. Extensive logging of the islands valuable dipterocarps and other forest trees precipitated conversion of forests to sugarcane plantations and agricultural lands until logging was banned in the 1970s (Hamann et al. 1999). As a result of human population pressures, forest loss continues today as kaingin, or slashandburn agriculture, encroaches on remaining forests (Heaney and Regaldo 1998). Currently, forest cover on Negros occupies only 14% of the islands land area (FMB 2010), of which only 4% is old growth forest (Brooks et al. 1992). Although a few small remnant fragments of lowland forest occur in southeastern Negros, all remaining original vegetation is montane or submontane rainfor est (Hamann et al. 1999). Furthermore, the Visayas remaining montane and submontane forests are recognized as centers of precinction and biodiversity and have received a classification of highest conservation priority by the IUCN (Dinerstein 1995). On Neg ros, only three substantial tracts of montane and submontane forest remain (Fig. 22). These tracts are protected as Mount Kanlaon National Park (24,288 ha), Northern Negros Forest Preserve (80,454 ha) in the north and Balinsasayao Twin Lakes Natural Park (8,016 ha) in the south Study Group The frugivorous guild of the butterfly family Nymphalidae served as the study group. The species of this guild observed at the site included members the tribes Nymphaliini, Limenitidni, Charaxini, Amathusiini and members of the subfamily Satyrinae. The frugivorous guild was selected as the study group because they are taxonomically tractable, relatively diverse and abundant, easily sampled and often responsive to disturbance or changes in habitat quality. In the Philippines, butterflies are the most well known insect group and their taxonomy and distribution are fairly well 41

PAGE 42

studied (Treadaway 2012). Further, butterflies have been widely used as environmental indicators because of their sensitivity to changes in habitat q uality. For these reasons, frugivorous butterflies are an ideal study group for investigations into the community structure of Philippine forests and ecological value of secondary forests. Transect Sampling The study site consists of a large, contiguous t ract of submontane rainforest. From its periphery, the forest tract transitions from a highly disturbed forest to an essentially intact old growth forest. Three 360 m transects were established along preexisting trails within the tract (Fig. 23), one wit hin each of the three following forest categories: Transect 1 (secondary forest garden) Secondary forest gardens are post extraction secondary forests that receive low levels of management and contain a planted component (Chokkalingam and de Jong 2001). The site surrounding Transect 1 consists of a secondary forest approximately 30 years in age. Throughout the site are poorly tended plots and untended patches of plantation crops, primarily bananas, coffee, cassava and Eucalyptus Macaranga, a species char acteristic of secondary growth, is common throughout the site and in some areas makes up the dominant canopy species. Dipterocarps are largely absent and the few that occur are saplings. Of the three transects, this site receives the highest level of small scale disturbance, primarily in the form of gathering dry wood for charcoal, harvesting of forest products and small scale agriculture. In this area, trails are widened to accommodate carabao or water buffalo that are used to move fuel wood out of the for est. The area is generally more open, and carabao are left to browse in gaps in the forest. The mean elevation of this transect was 625 m. Transect 2 (post extraction secondary forest) Post extraction secondary forests are secondary forests that have formed following the extraction of trees (Chokkalingam and de Jong 2001). Transect 2 was situated within a post extraction secondary forest approximately 30 years in age. Botanically, this site is similar to Transect 1, except that it contains fewer plantati on crops. Like Transect 1, the area surround Transect 2 lacked a canopy of mature dipterocarps. The area receives significantly less human disturbance, with much of the fuel wood gathering, collection of forest products and small scale agriculture restrict ed to the outer margin of the forest tract where Transect 1 is situated. Approximately 30 years ago, the area was logged of its dipterocarps. Subsequently, small scale 42

PAGE 43

removal of trees by locals has occurred. Mahogany and Gmelina have also been planted sporadically in the area. The mean elevation of this transect was 727 m. Transect 3 (old growth forest) Tran sect 3 is located within an old growth forest. This forest has not experienced significant disturbance in known history. However, there is evidence of low levels of tree removal, likely by illegal loggers. Evidence of hunting and gathering of forest products was also observed. There are numerous mature dipterocarps, forming the dominant canopy species. Macaranga is present but not as common. The under story contains sparser undergrowth than both Transect 1 and 2. Mean elevation of this transect was 792 m. Transects were established a minimum of 60 m from the edge of each habitat. Assistance in defining habitats was provided by Philippine Department of Natural Resources personnel and field guides familiar with each area and its history. Along each transect, 5 trapping sites were placed at 90 m intervals (Fig. 23). Trapping sites were established approximately 10 m off of preexisting trails. In many areas of the sample site, particularly the secondary forest transects, undergrowth restricted movement. For this reason, trapping sites were placed near trails. Trapping sites consisted of two butterfly traps. One trap was placed in the understory and another in the canopy. Nearby, o n Borneo, vertical stratification has been observed in forest butterfly communities (Tangah et al. 2004). Thus, ground based surveys are likely to exclude the canopy component of a sites butterfly fauna. To ensure the inclusion of this component in the sample, both canopy and understory traps were used. The canopy traps were consistently placed as close to the top of the canopy as possible. A Big Shot arborists slingshot was used to set the canopy traps and enabled placement of t he traps within the upper reaches of the canopy. The understory traps were set by hand and placed approximately 1 m from the forest floor. All traps were hung from nylon strings looped over the branches of trees, or in the case of some understory traps, vi nes. Each trap was assigned an identifying number. 43

PAGE 44

Butterfly traps (Fig. 2 4) were designed after UeharaPrado (2007), with a modification to the base and without the use of an interior funnel. Traps consisted of a gauze cylinder 100 cm in height and approximately 30 cm in diameter. The gauze used to make the cylinders was white sheer voile, made of polyester. The cylinder was knotted at the top to close it off. Two pieces of wire weaved through the circumference of the cylinder at the top and bottom of the trap provided a ring frame for support. A square platform of corrugated plastic was hung 3 cm beneath the open bottom of the cylinder. The corrugated plastic platforms were modified from the design of UeharaPrado and had a 10 cm hole drilled through them The hole was sized to accommodate a 236 ml plastic dish such that the top of the dish was level with the upper surface of the platform. This modification was included to prevent the fermenting fruit bait from sliding off the platform in wind and to prevent desiccation of bait exposed to sun or heat. A standard bait of fermenting fruit consisting of rotten and mashed bananas, mangos, jackfruit and tuba, a native coconut wine obtained locally, was used to attract frugivorous butterflies to the traps. The bait was allowed to ferment inside an 11.35 l plastic bucket for at least 48 hours prior to use. Fermenting fruit bait was added once a week, or as necessary. Traps were checked twice weekly for a period of close to one year. Transects were established and butterfly traps were set in August 2011 and remained in place until July 2012. However, typhoons and theft of traps caused two interruptions in data collection during the months of January and February 2011 and June 2012. Data were collected and traps were checked and maintained by local field assistants. Twice a week, field assistants checked every trap and all butterflies were 44

PAGE 45

recorded, marked on the ventral surface of the wing with the trap number of the trap it was observed in, using a black Sharpie mar ker and subsequently released unharmed at the exact point of capture. All traps per transect were checked on the same day. Traps were checked in the same order on every occasion, beginning with Transect 1 and followed by Transect 2 and 3. One side of the v entral surface of each butterfly caught in a trap was photographed using a Canon PowerShot A3300 IS point andshoot digital camera against a background indicative of the trap and transect that the butterfly was collected in (Fig. 2 5 and 26 ). Exif data associated with the resulting jpeg file recorded the date of capture. Once a month, the images were uploaded to a filesharing website and accessed remotely. For identification purposes, up to five voucher specimens of each species were collected and deposited at the University of Floridas McGuire Center for Lepidoptera and Biodiversity at the Florida Museum of Natural History. Taxonomic Identification Using vouchers and images of butterflies collected in the traps, species level identifications were made after Okano et al. (1989), Tsukada (1982) for the subfamily Satyrinae, Schroeder and Treadaway (2005) for the morphine tribe Amathusiini and Treadaway (2012). Subspecies level identification relied on distribution records published in Treadaway (2012). S tatistical Methods Koh (2007) highlights the exclusion of taxon sampling curves from comparisons of community diversity as a major pitfall in studies investigating the response of butterfly assemblages to disturbance. Sampling effects (e.g. differences in number of individuals sampled, differences in sampling effort) can make the comparison of diversity between communities problematic (Gotelli and Colwell 2001). Taxon sampling curves such as 45

PAGE 46

this are used to determine if the observed species diversity suffi ciently measures the true species diversity, implying that sampling saturation has been achieved. These curves plot the accumulation of species in a sample as a function of sampling effort. The curve rapidly rises as novel species are encountered but stabi lize at an asymptote as sampling saturation is approached. To assess the validity of measures of community diversity at each transect, a samplebased Coleman Rarefaction Curve, using the accumulated number of individual butterflies was plotted. Curves were plotted over 50 randomizations using the software EstimateS (Colwell 2005). After plotting Coleman Rarefaction Curves for each transect, Simpsons Reciprocal Index was used to calculate frugivorous butterfly diversity for each of the fifteen trap sites. The formula used to calculate Simpsons Index was: (Equation 21) Where n = number of individuals of a particular species and N = total number of individuals of all species Simpsons Reciprocal Index was obtained by calculating 1 / D. Simpsons Reciprocal Index was selected to measure diversity over other indices because it takes into account both richness and abundance. Using this formula, a community dominated by a few species is considered to be less diverse than a community of equal richness, but increased evenness in abundance (Nagendra 2002). Simpsons Reciprocal Index was calculated for each trap site. There were five trap sites per transect. Simpsons Reciprocal Index for each trap site was averaged to obtain a mean diversi ty measure for each transect. A oneway analysis of variance (ANOVA) 46

PAGE 47

was used to compare this measure of diversity between transects. This analysis was conducted using the statistical program R. It is important to emphasize that this measure of diversity reflects the diversity of butterflies collected in the traps and may not be an accurate measure of the true diversity of members of this guild within the sampled habitats. Between transects, there is a high degree of species overlap. Thus, these diversity measures should accurately represent the relative diversity between transects. Proportional rank abundance curves for each transect were plotted to model the species abundance of each transect. These curves yield a model of the species abundance at a site by plotting the relative abundance of a species against their rank in abundance. These models allow for comparisons of species richness and abundance between sites that do not necessarily share the same species composition (Foster and Dunstan 2010). In th ese curves, a sharp slope indicates uneven species abundances while lower slopes indicate increased evenness in species abundance within a community (Magurran 2004). These curves are based on data collected using baited traps. Such sampling methods can mak e rank abundance curves problematic to interpret. However, because this study is focused on comparisons between habitats that hold an almost homogenous species composition, they are included here. Hill and Hamer (1998) suggest that rank abundance curves can be used to identify disturbance in habitats. In butterflies, they report that butterfly communities occurring in disturbed habitats fit a log series distribution, while those of undisturbed habitats fit a log normal distribution. Abundance data generated by traps may not be a valid reflection of true species abundances in nature because there is variability in the attraction to traps 47

PAGE 48

among species. Despite this, rank abundance curves are presented here to visualize differences in the abundance and richnes s between transects Data from all transects was pooled and exact binomial tests were used to test if a butterfly species exhibits significant vertical stratification across all transects Exact binomial tests were also used to test each species within individual transects for vertical stratification. These tests were conducted under the null hypothesis that a species abundance in the understory is equal to its abundance in the canopy. For each species, exact binomial tests were also used to compare the proportional abundance of individuals collected in the canopy to the proportional abundance collected in the understory on each transect. Results of these tests were considered significant at the 95% confidence level. Species that were identified as being s ignificantly vertically stratified in the pooled sample or on individual transects were investigated individually using Fishers exact test. Fishers exact test performs well in the presence of small samples. This test was selected over a chi square analys is because of low sample numbers observed in some species. This test was used to compare the proportion of understory to canopy captures between Transect 1, 2 and 3. Results were considered significant at the 95% confidence level. Agglomerative cluster analysis was performed on the data to visualize similarities between trap sites and between transects. This analysis arranges sites into groups based on a selected measure of similarity in species composition. The most similar groups are clustered together and the most similar clusters are further clustered together until all groups have been joined (Hammer 2001). The Bray Curtis distance was selected as the measure of ecological similarity. The software Past was used to 48

PAGE 49

create a matrix of the Bray Curtis distances between each pair of trap sites and each pair of transects. Cluster analysis yielded dendrograms showing the similarity between each trap site and each transect. Ordination using nonmetric multidimensional scaling (NMDS) was also used to visualize the Bray Curtis distance between trap sites. NMDS is an ordination technique that calculates an index of similarity between each pair of samples. In NMDS, the data are fitted to a selected number of axes chosen before the analysis is conducted. The stress value is reported to give an indication of the accuracy of the ordination. The Bray Curtis distance equation that was used in Past was: (Equation 2 2) Results Altogether, a total of 1526 nymphalid butterflies, representing 21 species were collected throughout all transects (Table 21). A total of 563 individuals of 20 species, 473 individuals of 19 species and 490 individuals of 16 species were collected on Transect 1 (secondary forest garden), Transect 2 (post extraction secondary f orest) and Transect 3 (old growth forest), respectively. Community Diversity Coleman Rarefaction curves plotted for each transect stabilized and came to an asymptote by the end of the sampling period (Fig. 2 7 ), indicating that sampling effort was suffici ent to accurately measure species richness. The stabilization of rarefaction curves also indicates that diversity and richness measures for these communities may be directly compared. 49

PAGE 50

The cumulative species richness for all transects was 21 species of frugivorous nymphalid butterflies. Transect 1 (secondary forest garden), the most disturbed site, held the highest observed species richness, with 20 species. Observed species richness decreased with disturbance level. Transect 2 (post extraction secondary forest) held an observed species richness of 18 species. Transect 3 (old growth forest) held the lowest observed richness with 16 species. Despite the observed decrease in richness from Transect 1 to Transect 3, analysis of variance showed no significant dif ferences in species richness between transects (F2,12 = 0.8, P = 0.4719) On all transects, the observed species abundance was dominated by a few species (Table 22). On Transect 1, the satyrines Mycalesis tagala Melanitis leda and My. georgi were the domi nant species, accounting for 55% of the individuals sampled. On Transect 2, My. tagala Me. boisduvalia and Me. leda were the dominant species, accounting for 64% of the individuals sampled. On Transect 3, My. tagala Me. boisduvalia and Me. atrax were dom inant and accounted for 66% of the individuals sampled. Levels of total abundance on each transect were not significantly different (F2,12 = 1.068, P = 0.3743). Simpsons Reciprocal Index takes into account both species richness and species abundance (evenness) and considers communities that are primarily dominated by few very common species less diverse than communities with similar richness and increased evenness in species abundances. Simpsons Reciprocal Index was calculated as a measure of butterfly di versity collected at trap sites on each transect. As with species richness, diversity was highest on Transect 1 (the most disturbed site) and lowest on Transect 3 (the most undisturbed site). Simpsons Reciprocal Index was 50

PAGE 51

6.385, 5.821 and 4.789 for Transect 1, Transect 2 and Transect 3, respectively. Analysis of variance showed no significant differences between transects (F2,12 = 2.6448, P = 0.1118). Community Structure The proportional rank abundance (Fig. 28 ) and log/rank distributions (Fig. 29 ) for e ach habitat showed curves indicative of fairly similar communities. Both Transect 1 and Transect 3, and to a lesser extent, Transect 2 showed tails indicating the presence of rare species only collected on a few occasions. The steep slopes of these curves are indicative of communities dominated by one or a few very abundant species. Both curves show that on all transects, much of the species abundance is held by a few very common species. This is particularly apparent in the slope of Transect 3. Mycalesis t agala was the most abundant species on all transects and was represented cumulatively by 476 individuals (approximately one third of all butterflies collected). This species was particularly abundant on Transect 3 and contributes to the steep slope observed in both the proportional rank abundance and log/rank curves. This species alone accounts for approximately 27% of the individuals collected on Transect 1 and 2 and >40% of the species collected on Transect 3. On all transects, the three most abundant species were members of the genera Mycalesis and Melanitis in the subfamily Satyrinae. While My. tagala was the most abundant species on all transects, the second and third most abundant species were different for each transect (Table 22). Because these curv es were created based on data from baited butterfly traps, care must be given in their interpretation. Attraction to baits can vary between butterfly species. These curves reflect observations on the abundance and richness of species that visited the traps rather than the true richness and abundance of the natural community of 51

PAGE 52

frugivorous butterflies. However, there is a high degree of species overlap in these communities and the communities are located relatively close to one another. For these reasons, t hese curves can be used to compare the community structure of these species sampled by the traps between transects. Vertical Stratification Eleven of the 21 butterfly species collected at all trap sites were significantly vertically stratified when transect data was pooled (Table 23). Of these, only three species were primarily canopy inhabitants ( Ptychandra negrosensis Rhinopalpa polynice and My. teatus ). Two species of the subfamily Charaxinae were present at the sites ( Charaxes amycus and C. solon ). T hese species would be expected to be primarily canopy inhabitants, but the low number of captures for both species did not produce a significant result in the exact binomial tests. On each transect, My. teatus a small satyrine, was collected most frequent ly in canopy traps, while the other three Mycalesis species were collected primarily in understory traps. Fishers exact test was used to identify differences in vertical stratification of each species between transects. All species identified as significa ntly vertically stratified using binomial exact tests on pooled data and on individual transects were further investigated using this analysis. Species exhibiting significant vertical stratification but low sample sizes were excluded from this analysis. Th ere was little contrast in the vertical stratification of species between transects. The proportion of individuals of a species collected in understory to canopy differed significantly between transects only for My. georgi and My. tagala. For My. georgi th is proportion was significant only between Transect 2 and 3 (P = 0.0032). For My. tagala there was no difference between Transect 1 and 2 (P = .0675). 52

PAGE 53

Comparisons of this species on Transect 1 and 3 (P = .0374) and Transect 2 and 3 (P < .00004) were signif icantly different. Similarity Between Communities The cophenetic correlation of the trap site cluster analysis was .7076, indicating an acceptable representation of pairwise similarities between individual trap sites. The results of cluster analysis on tr ap sites showed weak grouping of sites by transect (Fig. 2 10). Cluster analysis arranges sites into groups based on similarities in species composition. In this analysis, the Bray Curtis distance was used to calculate similarity between sites. It would be expected that if the type of forest in which a transect is located has a strong effect on the butterfly diversity sampled at each trap site then trap sites would be clustered by transect (e.g. Trap Sites 15, 6 10, 1115, representing Transects 1, 2 and 3 respectively, would cluster together). Three of the five Transect 1 trap sites group together with similarities >75%. Trap site 3 (Transect 1) grouped on the same branch as Trap Sites 8 and 10 (Transect 2). Trap Site 4 (Transect 1) was grouped singly, in dicating that it had the least similarity in pairwise comparisons of the individual sites. This particular site was located near a small gap in forest where eucalyptus, bananas and other plantation crops were grown. The presence of this forest gap may have had an effect on the species composition that was observed at this trap site, potentially resulting in its high degree of dissimilarity. None of the Transect 1 trap sites were grouped with any Transect 3 sites. Transect 2 trap sites were grouped with sites from both Transect 1 and 3, while Transect 3 sites were only grouped with sites from Transect 2. Trap Sites 12 and 14 (Transect 3) were grouped together and formed a larger group with Trap Sites 11 and 15 (Transect 3), but also Trap Site 6 (Transect 2). Trap Site 13 (Transect 3) was most similar to Trap Site 9 (Transect 2) and formed a 53

PAGE 54

larger group with three Transect 2 sites and one Transect 1 site. A separate cluster analysis was performed to group transects to one another based on similarity (Fig 211) This analysis grouped Transect 1 and 2 on the same branch, with Transect 3 on its own. However, the cophenetic correlation was low (0.5526), indicating weak similarity between transects. A similar pattern of weak grouping by transect was observed using NMDS (2D, Stress = .2209). This method grouped trap sites on a selected number of axes using the Bray Curtis distance. In this ordination, trap sites were plotted relative to each other in a 2 dimensional space. The distance between points is an indication of the similarity between sites. Points that are closer together in the NMDS plot are more similar than points that are farther. Ordination using NMDS reveals no distinct clus tering by transect (Fig. 2 12). However, loose clusters can be recognized with s ome overlap (outlined in figure). The dissimilarity of Trap Site 4 from all other trap sites is reflected in its distance from other points. Discussion The loss of tropical forests is one of the greatest threats to biodiversity today (Brooks et al. 2002). The degree of forest loss is particularly severe in the Philippines where the vast majority of old growth forests have been removed or converted to agricultural land or secondary forests. Few studies have investigated the effects of these largescale modif ications to the Philippine landscape on the Philippines forest communities. This is especially true of insect and invertebrate communities. The results of studies on the effects of habitat disturbance on butterflies in Southeast Asia and elsewhere contain little consensus. Positive and negative effects on richness, abundance and diversity have been observed with similar incidence (Koh 2007). In part, 54

PAGE 55

this may be the result of the various spatial scales at which butterfly sampling has been conducted. Studies conducted over small spatial scales seem to be more likely to observe positive effects on butterfly richness, diversity and abundance as the result of habitat disturbance, while the reverse is true about studies that use a sampling regime which covers a larger spatial scale (Hamer and Hill 2000). This study extensively sampled frugivorous butterflies over a small spatial scale at the interior margin of Mount Kanlaon National Park for a period of almost one year. Fifteen trap sites were established along t hree transects, each set within a habitat that held varying disturbance histories. One transect was set in a secondary forest garden, post extraction secondary forest and old growth forest. The results of this study suggest that frugivorous butterfly communities occurring in the secondary forests that form the margin of Mount Kanlaon National Park are poorly differentiated from communities of the same guild that occur within nearby old growth forests. Although observed species richness and diversity were gr eatest in the most disturbed habitat sampled and decreased towards the most pristine habitat, in line with similar small scale studies (e.g. Spitzer et al. 1997, Lawton et al. 1998, Wood and Gillman 1998), these differences were not significant. There were no strong effects observed, either positive or negative, on the community structure, diversity or vertical stratification of frugivorous butterfly communities occurring in the habitats examined. The proportional rank abundance and log/rank abundance curves show that community structure is relatively similar on all transects. All transects were dominated by My. tagala and other satyrines. My. tagala was particularly dominant on Transect 3, the old growth site, and contributes to the steep slope observed on the proportional rank 55

PAGE 56

abundance curve. The five most abundant species collected in the traps of each transect all belonged to the satyrine genera Mycalesis and Melanitis The placement of the three Melanitis species in the ranked abundance of each transect may be particularly meaningful. Melanitis leda was the most abundant of the three Melanitis species and second most abundant butterfly species on Transect 1, the secondary forest garden traps. Melanitis leda is a widespread species that occurs through Asi a, Africa and Australia. In the post extraction forest, Me. leda is replaced as the second most abundant butterfly species in the traps by Me. boisduvalia a Philippine precinctive. In the old growth forest traps, Me. leda was replaced again as the third m ost abundant butterfly species by another precinctive Melanitis species, Me. atrax Studies on a range of taxa, including butterflies, have noted that species in disturbed habitats often have widespread distributions (Thomas 1991, Stephenson 1993, Lewis 1998, Fjeldsa 1999, Wijesinghe and Brook 2005). During the course of this study, Me. leda was observed in cities and other human dominated landscapes. Posa and Sodhi (2006), in their study on the effects of habitat modification on butterflies and birds, only observed one Melanitis species, My. boisduvalia. In this study, Me. boisduvalia was not observed in modified habitats. Further, Me. boisduvalia was most abundant in closed canopy forest and less abundant in open canopy forest. Of the frugivorous butterfly taxa collected during sampling for this study, the genus Melanitis has the best potential for use as an environmental indicator on Negros, however, further study is required to link disturbance to the changes in abundance of this group observed in this st udy. In addition, the three members of this genus present on Negros can be easily identified. Mycalesis also has 56

PAGE 57

potential as an environmental indicator of habitat quality, but difficulty in the taxonomic identification of species makes the genus problemat ic. Of the habitats sampled, the secondary forest garden had experienced the greatest level of disturbance. Like the post extraction secondary forest, this area had been logged approximately 30 years prior to the study. This area is easily accessible to t he human populations of nearby communities and is used as a low maintenance garden. Agricultural operations in this area are largely passive, and plantation crops are untended. Transect 1, established through this area, held the greatest species richness, abundance and diversity. There were two species unique to this transect. My. igoleta and Charaxes solon were only observed only on Transect 1. Both species show preferences for open areas. My. igoleta is a small satyrine often observed at forest edges and road sides, while C. solon is a canopy species. Disturbed forests such as this have been noted to provide suitable habitat for edge, gap and canopy species (Brown and Hutchings 1997, Ghazoul 2002). The influx of these species following a disturbance to for est structure can inflate species richness, at least at a local scale. The transect established through old growth forest collected the least number of species. Because this study was conducted over a small spatial scale, the observed species richness may be lower than the true species richness of the habitat. These results agree with those of studies at similar spatial scales in Southeast Asia which report increased species diversity in disturbed forests and comparatively lower diversity in pristine forest s (Spitzer et al. 1997, Lien and Yuan 2003). Species richness at all sites was lower than expected. Cumulatively, only 21 species were collected after close to one year of sampling. A number of additional 57

PAGE 58

frugivorous species are known to occur at MKNP, but were never collected by the traps or observed during the sampling period (Treadaway 2012). For example, Polyura schreiber Elymnias sansoni E. kanekoi Zeuxidia semperi and Amathuxidia amythaon come readily to rotting fruit baits and are known from the v icinity, but were never observed or sampled (Schroeder and Treadaway 2005, Treadaway 2012). It is possible that these species have been extirpated from the area or that these species are particularly sensitive to habitat disturbance and the secondary fores t trap sites, possibly even the old growth sites, were too disturbed for some or all of these species listed above. In addition, the transect placed in the old growth forest were several hundred meters from secondary forest sites. It is possible that this proximity to secondary forest and disturbed areas affected the presence of these missing species. The results of the cluster analyses, NMDS ordination and absence of significant differences in between transect comparisons of diversity, richness and abunda nce suggest that the species structure of trapvisiting butterflies is similar between the secondary forest garden, post extraction secondary forest and old growth forest. While this does not establish that secondary forests are valuable for old growth frugivorous butterfly communities, it provides evidence that such forests are not avoided by these butterflies and may be useful in augmenting current conservation measures. The secondary forests sampled in this study were approximately thirty years old and contiguous with a tract of old growth forest. The similarities between the samples, in large part, may be due to the lack of barriers (e.g. stretches of agricultural land) between old growth and secondary forests. Future studies should investigate if the sa me similarities arise in isolated tracts of secondary forests. Further, studies 5 8

PAGE 59

conducted in other tropical locations have observed the recovery of forest communities in as little as 20 40 years following a significant disturbance to forest structure (Guar iguata and Ostertag 2001). The age of the secondary forests sampled in this study has likely also contributed to the lack of significant differentiation between habitats and 30 years may be a sufficient period of time to bring back old growth levels of div ersity. Butterfly communities in tropical forests are vertically stratified (DeVries 1997). When forest structure is disturbed by timber extraction, the vertical stratification of butterfly species can be disrupted and canopy species have been observed moving towards the understory (Devries 1988, Schulze et al. 2001, Fermon et al. 2005). Eleven of the 21 butterfly species were significantly vertically distributed. In these species, there were significant differences in the proportion of individuals collect ed in the understory to canopy. However, in comparisons of the vertical stratification of species between habitats only two species ( My. tagala and My. georgi ) were significantly different. Mycalesis tagala and My. georgi are both primarily understory spec ies. While both species were most frequently collected in understory traps throughout the habitats, the ratio between understory and canopy captures was reduced in the old growth sites. In the old growth sites, traps were placed as near to the top of the c anopy as possible. However, this often resulted in the traps being hung from the highest branches, but still shaded by branches. Understory species may avoid areas of bright light. The sparser canopy and greater light intensity of the secondary forest transects may have resulted in an aversion to the canopy traps by these species. The majority of species that were significantly vertically stratified were understory species ( eight out of eleven species). With the exception of Me. leda all understory species are precinctive to the Philippines. 59

PAGE 60

Species that were significantly associated with the canopy included Rhinopalpa polynice, My. teatus and unexpectedly, Ptychandra negrosensis Mycalesis teatus is a small satyrine that would be expected to be an understory species. In all habitats, canopy traps collected it approximately twice as frequently than understory traps. Ptychandra negrosensis is another small, but fast flying satyrine. On Negros and at the study site, P. negrosensis is sympatric with a similar P tychandra species, P. leucogyne Ptychandra leucogyne did not show significant vertical stratification at the study site and was collected in both understory and canopy traps. It would be expected that both these butterflies would be understory species. The dorsal surfaces of the wings of both species are iridescent blue. It has been proposed that this coloration may be useful to butterfly species that specialize microhabitats with low or complex lighting (e.g. the understory of tropical forests) (Douglas et al. 2007). Niche partitioning could explain the differences observed in the vertical stratification of these two similar species. The two Charaxes species sampled in butterfly traps would also be expected to be canopy species, but small sample sizes for both species did not result in significant vertical stratification. Collectively, these analyses find little distinction between the diversity and community structure of frugivorous butterflies sampled by traps in each habitat. Immediately following a dis turbance to forest structure, the species richness of a secondary forest rapidly recovers to levels found in old growth forests (Guariguata and Ostertag 2001). Although some forest taxa are able to persist, species composition of secondary forest communiti es, at least initially, remains different from that of old growth forests (Dent and Wright 2009). As secondary forests regenerate towards an old growth 60

PAGE 61

state through time, biotic communities recover and become more similar to old growth forests. This process is hastened when an old growth forest, acting as a source habitat, is in close proximity and there are no barriers to movement of forest taxa from an old growth site to a recovering forest (Chazdon 2003). One explanation for the lack of distinction betw een secondary forest frugivorous butterflies communities and old growth communities is the age of the secondary forests combined with the adjacent old growth site potentially acting as a source for forest taxa to recolonization. Understory butterfly commu nities avoid areas of high light and are often associated with particular microhabitats It is likely that following timber extraction, the understory butterflies sampled here were absent from the area. As succession progressed and the understory microhabi tat recovered, these butterflies were able to freely recolonize from the old growth site. The similarities between these communities show that secondary forests, adjacent to old growth forest, are utilized to a similar extent by forest species of frugivor ous butterflies. Further study is required to determine how much this holds true for communities of other forest taxa or whether more isolated secondary forests provide a similar resource for forest butterflies. Limitations in the scope and sampling procedures of this study limit the ability of the inferences that may be drawn on the effects of habitat disturbance on butterflies in the Philippines. This study does not attempt to establish disturbance as the cause of the community level similarities or diff erences that were observed. By comparing the frugivorous butterfly communities of a secondary forest garden and post extraction secondary forest to that of an old growth forest, this study seeks to increase understanding on the potential of secondary fores ts to support similar communities to 61

PAGE 62

old growth forests. Further study is required to understand if the similarities observed for frugivorous butterflies are a reflection of general butterfly communities, or those of other taxa. While frugivorous butterfli es are an ideal study group for this type of study, they suffer from some constraints. In this study, the frugivorous guild of Nymphalidae was dominated by the subfamilies Amathusiinae and Satyrinae. These subfamilies utilize plants from the families Poaceae and Arecaceae as larval host plants. These plants are often well represented in disturbed areas. Thus, these subfamilies may not be representative of butterflies in general, in terms of their relationship to secondary forests. Further, because of specieslevel differences in attraction to fruit baits, it is important to recognize that observations on community structure that are based on data generated by traps may not reflect the true patterns of diversity within the community. Despite these limitations and constraints, this study provides evidence that secondary forests, particularly those at the margins of protected areas, hold potential for augmenting current conservation strategies that target old growth sites. Philippine secondary forests generally receive little protection. The integration of secondary forests, particularly those at the periphery of protected areas may expand the area of habitat available to Philippine forest taxa, thereby bolstering the conservation value of the protected area as a whole. 62

PAGE 63

Table 21. List and raw counts of species sampled by butterfly traps on each transect Species Transect 1 Transect 2 Transect 3 Rhinopalpa polynice panayana Fruhstorfer 1912 2 3 2 Tanaecia lupina howarthi Jumalon 1975 10 14 4 Lexias satrapes am lana Jumalon 1970 5 6 5 Charaxes solon lampedo Hbner (1824) 3 0 0 Charaxes amycus negrosensis Schroeder & Treadaway 1982 3 8 5 Faunis phaon carfinia Fruhstorfer 1911 0 1 2 Discophora dodong Schroeder and Treadaway 1981 25 10 11 Discophora ogina pulch ra Nihira 1987 11 7 8 Amathusia phidippus negrosensis Okano & Okano 1986 4 4 0 Melanitis leda leda Linneaus 1758 86 62 34 Melanitis atrax soloni Okano & Okano 1991 61 46 47 Melanitis boisduvalia boisduvalia Felder & Felder 1863 68 113 78 Zethera musid es Semper 1878 17 18 11 Lethe europa cevanna Fruhstorfer 1911 3 0 0 Lethe chandica canlaonensis Okano & Okano 1991 5 8 17 Ptychandra leucogyne Felder & Felder 1867 4 5 17 Ptychandra negrosensis Banks, Holloway & Barlow 1976 7 2 0 Mycalesis georgi canl aon Aoki and Uemura 1982 75 23 19 Mycalesis teatus Fruhstorfer 1911 14 16 30 Mycalesis tagala mataurus Fruhstorfer 1911 149 127 200 Mycalesis igoleta negrosensis Aoki and Uemura 1982 11 0 0 Total 563 473 490 Table 22. The five most abundant species on each transect, ranked by abundance Transect 1 Transect 2 Transect 3 1. Mycalesis tagala 1. Mycalesis tagala 1. Mycalesis tagala 2. Melanitis leda 2. Melanitis boisduvalia 2. Melanitis boisduvalia 3. Mycalesis georgi 3. Melanitis leda 3. Melanitis at rax 4. Melanitis boisduvalia 4. Melanitis atrax 4. Melanitis leda 5. Melanitis atrax 5. Mycalesis georgi 5. Mycalesis teatus 63

PAGE 64

Table 23. Vertical stratification of butterflies captured by traps and results of exact binomial tests Species Understory Ca nopy P value Significance Rhinopalpa polynice 0 7 0.015 Tanaecia lupina 21 7 0.012 Lexias satrapes amlana 15 1 0.0005 Charaxes solon 0 3 0.25 NS Charaxes amycus 5 11 0.31 NS Faunis phaon 3 0 0.25 NS Discophora dodong 26 20 0.46 NS Disc ophora ogina 20 6 0.009 Amathusia phidippus 6 2 0.29 NS Melanitis leda 110 72 0.005 Melanitis atrax 118 36 <0.0002 Melanitis boisduvalia 182 77 <0.0005 Zethera musides 30 16 0.055 NS Lethe europa 0 3 0.25 NS Lethe chandica 19 11 0.2 NS Ptychandra leucogyne 11 15 0.56 NS Ptychandra negrosensis 1 8 0.039 Mycalesis georgi 77 40 0.0007 Mycalesis teatus 17 43 0.001 Mycalesis tagala 298 178 <0.0004 Mycalesis igoleta 9 2 0.065 NS Butterfly species in bold indicate canopy species. 64

PAGE 65

Figure 21. Study location, Negros, Visayas, Philippines 65

PAGE 66

Figure 22. Remaining old growth forest tracts on Negros. The study site, Mount Kanlaon National Park, is circled in red. 66

PAGE 67

Figure 23. Sampling design. Three transects were established, one in a habitat classified as post extraction secondary forest, old growth forest and secondary forest garden. Each transect was placed at least 60 m from the edge of the habitat. Five trap sites were placed at 90 m intervals along each transect. Each trap site consisted of an understory butterfly trap, set at a height of approximately 1 m, and a canopy trap set as close as possible to the top of the canopy. 67

PAGE 68

Figure 24. Butterfly trap. Butterfly traps consisted of a gauze cylinder that was closed at the top. A corrugated plastic platform was hung below the trap and held a dish of rotting fruit bait. Photo courtesy of Chris Johns. 68

PAGE 69

Figure 25. Example image of the method used to collect data. Field assistants checked bu tterfly traps twice weekly. Every butterfly collected was photographed against a background indicative of the trap and transect where the butterfly was collected. Exif data associated with the photograph recorded the date of capture. Photo courtesy of Balm erie Villar. 69

PAGE 70

Figure 26. Data collection. Field assistant, Balmerie Villar photographs Melanitis butterflies captured in a butterfly trap. Photo courtesy of Chris Johns. 70

PAGE 71

Figure 27. Coleman Species Rarefaction Curve for each transect. Dashed line represents Transect 1 (secondary forest garden), dotted line represents Transect 2 (post extraction secondary forest) and solid line represents Transect 3 (old growth forest). 71

PAGE 72

Figure 28. Proportional rank abundance curve for each transect. Dashe d line represents Transect 1 (secondary forest garden), dotted line represents Transect 2 (post extraction secondary forest) and solid line represents Transect 3 (old growth forest). 72

PAGE 73

Figure 29. Log rank abundance curve for each transect. Dashed l ine represents Transect 1 (secondary forest garden), dotted line represents Transect 2 (post extraction secondary forest) and solid line represents Transect 3 (old growth forest). 73

PAGE 74

Figure 210. Results of agglomerative cluster analysis by trap site. Showing community similarity performed on trap sites and calculated using the Bray Curtis distance equation. Trap sites highlighted in red indicate Transect 1. Trap sites highlighted in yellow indicate Transect 2. Trap sites highlighted in green indicate Tra nsect 3. Cophenetic correlation = .7076. 74

PAGE 75

Figure 211. Results of agglomerative cluster analysis by transect. Performed on composition and abundance data from transects. Similarity was calculated using the Bray Curtis distance. Cophenetic correlation = 0.5526. 0.75 0.78 0.81 0.84 0.87 0.90 0.93 0.96 0.99Similarity321 75

PAGE 76

Figure 212. Results of nonmetric multidimensional scaling by trap site. Weak clustering by transect is observed. Trap sites belonging to the same transect are outlined (Transect 1 in red, Transect 2 in yellow, Transect 3 in green). Trap Site 4 is dissimilar from all other site and is not included with other Transect 1 sites. Stress value = .2209. 76

PAGE 77

CHAPTER 3 RANGE SIZE AND LEVELS OF PRECINCTION Introduction Earths biodiversity is not distributed uniformly (Kerr 1997). Some regions of the planet contain a disproportionate number of species or an unusually high concentration of precinctive species in comparison to others. Geographic locations that contain a biota rich in precinctives are considered unique and globally irreplaceable. In conservation planning, high concentrations of precinctive species, coupled with critical levels of habitat loss are popularly used as criteria in the prioritization of geographic regions for conservation action This approach identifies r egions of the planet as biodiversity hotspots, based on the presence of an exceptional concentration of precinctive species that is currently under exceptional pressure from habitat loss (Myers et al. 2000). Through this prioritization system, geographic areas can be evaluated in terms of global irreplaceability and vulnerability. Although various systems and criteria (e.g. species richness, number of imperiled species, presence of extensive wilderness areas) are also utilized in the establishment of conser vation priorities (Brooks et al. 2006), the biodiversity hotspot approach has received unusual attention among the conservation community and is widely perceived as a valid conservation strategy (Myers 2003). By prioritizing areas exhibiting high irreplaceability and high vulnerability, the biodiversity hotspot system emphasizes the importance of preserving precinctive species and taxa. Precinctive species inherently occupy a limited geographic area. As a result, they are intrinsically more susceptible to extinction than widespread species because local extirpation is more likely to result in global extinction (Charrette et al. 2006). Limited rangesize has been recognized repeatedly as a correlate of extinction proneness 77

PAGE 78

across a variety of taxa (Purvis et al. 2000, Jones et al. 2001, Koh et al. 2004). Furthermore, there is abundant evidence that rangerestricted species are more susceptible to disturbances or changes in habitat quality than widespread species (Thomas 1991, Hill et al. 1995, Hamer et al. 199 7, Cleary and Genner 2004, Sodhi et al. 2010). Thus, precinctive species are conservation targets and through the prioritization of geographic regions that support high concentrations of precinctives and are also under severe pressure from habitat loss, the conservation community is able to develop and implement conservation strategies that effectively conserve the most vulnerable and irreplaceable elements of the global biota. The Philippine Archipelago is a global conservation priority. The country consistently ranks high in terms of irreplaceability and vulnerability, making it one of the most imperiled biodiversity hotspots (Myers 2000, Brooks et al. 2000, Brummitt and Lughadha 2003, Ovadia 2003, Fonseca 2009). Of the Philippines terrestrial vertebrate and plant species, approximately 50% are precinctive to the archipelago (Heaney and Regaldo 1998). Many of the countrys terrestrial precinctive taxa depend on the onceextensive forests that blanketed the archipelago (de Jong and Treadaway 1993, Kennedy et al. 2000). Over t he past century, the Philippine Islands have experienced massive and widespread forest loss driven largely by the timber industry and the subsequent conversion of loggedover areas to farmland. Historically, the majority of the countrys land area was forested (Liu 1993). Currently, only 20% of the Philippines land area is classified as forest cover (FMB 2010), of which, <1 million ha (<5% of the Philippines land area) represent original old growth forests. Various types of secondary f orest of unknown ecological importance are currently the dominant component of the 78

PAGE 79

Philippines forestlands and account for 83% of land classified as forest cover (Lasco et al. 2001). Secondary forests are defined here after Chokkalingam and de Jong (2001) as a forest that is regenerating largely through natural processes after significant human and/or natural disturbance of the original forest vegetation at a single point in time or over an extended period, and displaying a major difference in forest structure and/or canopy species composition with respect to nearby primary forests on similar sites. Concomitant with the rapid loss of the vast majority of the Philippines old growth forests has been the formation of secondary forests. The impact of these l andscape changes on the Philippines forest biota and precinctive taxa, as well as the potential of secondary forests to mitigate these impacts, is poorly understood. Elsewhere in the tropics, secondary forests have been found to provide habitat for a subs et of the species that occur in nearby old growth forests ( Dent and Wright 2009). As a secondary forest ages and its physical structure recovers, its biotic communities become increasingly similar to those of old growth forests (Pardini et al. 2009). Howev er, the recovery of precinction levels within a secondary forest can require significantly longer periods of time (Dunn 2004). In the Philippines, much of the current work on the response of Philippine forest communities to changes in habitat quality has been focused on charismatic vertebrates, primarily mammals and birds. In tropical forests, insects and other invertebrates hold functional roles and perform many of the essential services that maintain ecosystems (Wilson 1987). Despite this, few studies hav e investigated the response of the Philippines insect fauna to largescale habitat modification. 79

PAGE 80

This study investigates the ecological importance of a secondary forest in the Philippines to precinctive frugivorous butterflies by examining changes between the frugivorous butterfly communities of a secondary forest garden, post extraction secondary forest and old growth forest on the island of Negros. The objective of this study is to identify community level differences in the distribution of butterfly abu ndances, ranked by rangesize between the three forest types. For investigations on the response of insect communities to habitat modification in the Philippines, butterflies provide an ideal study taxon. Butterflies are the best known of the Philippines insect taxa and thus, taxonomically tractable. The Philippine butterfly fauna is diverse and contains a significant precinctive component (40% of all species) (Treadaway 2012). Further, Philippine butterfly assemblages are known to be sensitive to habitat modification and range restricted species have been found to be particularly sensitive to habitat disturbances (Posa and Sodhi 2006). Not all species respond negatively to disturbances or habitat modification. Some are unaffected, while others respond pos itively. In both mainland and insular systems, an association has been observed between a species rangesize and its response to disturbance (Stephenson 1993, Fjeldsa 1999, Wijesinghe and Brook 2005). In butterflies, species that occupy a limited geographic distribution have largely been found to have a lower tolerance for modified or disturbed habitats, and tend to be more closely associated with pristine habitats, in comparison to widespread species (Thomas 1991, Spitzer et al. 1997, Lewis et al. 1998, H ill et al. 2001, Cleary et al. 2009). Further, on Borneo, precinctive butterfly species were observed to be significantly less resilient to habitat disturbance than widespread species (Cleary and Genner 2004, Charrette et al. 80

PAGE 81

2006). These differences in response to disturbance may be a result of differing degrees of ecological specialization between limitedrange and widespread taxa. In butterflies and other taxa, species that occupy a restricted geographic area may be more likely to be ecological specialis ts (Harcourt et al. 2002, Koh et al. 2004). Meanwhile, widespread and cosmopolitan species are more often ecological generalists and able to tolerate a wider breadth of environmental circumstances. Thus, modified habitats tend to be dominated by widespread, generalist species of low conservation concern that replace precinctive species (McKinney and Lockwood 1999). Secondary forests represent a habitat that is intermediate between an old growth forest and a human dominated habitat (e.g. agricultural matrix ). In the Philippines, these forests are now the major component of the countrys forestlands (Lasco et al. 2001). Current government sponsored conservation plans and protected areas are primarily concentrated on remaining old growth sites. However, if sec ondary forests are able to provide habitat for the precinctive elements within Philippine forest communities, they could play a valuable role in augmenting current conservation strategies, particularly if managed to optimize conservation value. In addition considering the extent of forest loss in the Philippines, integration of secondary forests into management plans could bolster the ability to effectively conserve Philippine biodiversity. Unfortunately, little is known on the ecological importance of Phi lippine secondary forests. In order to understand their potential role in Philippine conservation, it is necessary to assess the response of the communities of forest taxa to these habitats. This study investigates how precinction levels of frugivorous but terfly communities differ between secondary forests and old growth forest. In order to fully understand the value of secondary forest 81

PAGE 82

to precinctive forest species, the inclusion of additional taxa will be necessary in future studies. However, as pollinators, butterflies play functional roles within ecosystems, have been used widely as environmental indicators of habitat quality and many Philippine species are categorized as conservation dependent (Danielsen and Treadaway 2004). For these reasons, butterfli es are an ideal study taxon with which to begin an assessment of the conservation value of secondary forests to Philippine forest taxa. Methods Study Site and Sampling Methods The study site and sampling methods are presented in Chapter 2 of this manuscri pt Taxonomic Identification and Geographic Distribution Using vouchers and images of butterflies collected in the traps, species level identifications were made after Okano et al. (1989), Tsukada (1982) for the subfamily Satyrinae, Schroeder and Treadaway (2005) for the morphine tribe Amathusiini and Treadaway (2012). Subspecies level identification relied on distribution records published in Treadaway (2012). Geographic distributions of species and subspecies within the Philippines were determined from r ecords published in Treadaway (2012). Distribution information outside the Philippines relied on Tsukada (1982). Statistical Methods Measures of species richness and diversity of a community or assemblage do not convey any information on the importance of a species to conservation. A community can be exceptionally diverse, but not contain any species relevant to conservation. For example, a number of studies on butterflies in Southeast Asia and 82

PAGE 83

elsewhere find the highest species richness in the most distur bed areas examined (Hamer et al. 1997, Walpole and Sheldon 1999, Willott et al. 2000, Cleary et al. 2005). In most cases, widespread species accounted for much of the high levels of diversity observed in these areas. It is therefore important to identify areas or habitats that are important to the species that are of conservation concern. In the Philippines, precinctive and rangerestricted species are conservation targets. Thus, in this analysis, the species sampled at all transects were ranked by range si ze using categories of precinction (Table 31). The numbers of major Philippine islands that are occupied by a species were used as a measure of range size. Fifteen of the 21 species observed were precinctive to the Philippines. In the ranking system used here, only the ten largest islands in the Philippine Archipelago are considered (Luzon, Mindanao, Negros, Samar, Palawan, Panay, Mindoro, Cebu, Bohol and Leyte). The area of these islands range from 2368 km2 to 109965 km2. Despite this, they are considered here as units to measure the extent of occurrence of species. The precinction categories that were used to rank species by distribution is as follows: 1 10: Species that received a rank between one and ten occupy a global geographic distribution that inc ludes only that number of major Philippine islands. 11: Global distribution of a species is restricted to Southeast Asia. 12: Global distribution of a species is restricted to Asia. 13: Global distribution of a species includes Asia and at least one other continent. In this ranking system, the most rangerestricted butterfly species are given the lowest rank and the most widespread species are given the highest rank. Species were placed in a precinction category based on the global distribution at the species level. In some cases, several species were assigned to the same category. In these cases, any 83

PAGE 84

species sharing a precinction category with another, or multiple species were further ranked based on the distribution of the subspecies present at the study site. The ranked distributions of the communities sampled by trap transects in the secondary forest garden, post extraction secondary forest and old growth forest were compared using the MannWhitney U test. The Mann Whitney U test tests the null hypothes is that the distributions of two groups are the same. In this case, it was used to compare the ranked distributions of butterfly abundances between transect pairs. The same ranking system was used on the subspecies collected at the trap sites (Table 32) a nd separate MannWhitney U tests were performed. These tests follow the methods used by Hill et al. (1995), Hamer et al. (1997) and Willott (2000) to investigate differences in the precinction levels of butterfly communities in Indonesia and Malaysia, but utilize a modified precinctionranking scheme. In order to test if a particular species is more abundant in the sample from a particular transect in relation to its abundance in other samples, the proportional abundances of each species were compared betw een habitats using Fishers exact test. Fishers exact test performs well in the presence of small sample sizes. This test was selected over a chi square test because some butterfly species were collected rarely and had small sample sizes. These tests compare the proportional abundance of a species within a sample (transect) to its proportional abundance within each of the other transects. If a species proportional abundance is greatest on, for example, Transect 3 (old growth forest), this would ideally im ply that its abundance in that community is greater than the other communities sampled. However, because these data have been obtained through fruit baited traps, these comparisons can only test for changes in 84

PAGE 85

abundance within the sample, and may not accur ately reflect true changes in proportional abundance. The purpose of performing hypothesis tests on the proportional abundance of each species is to consider a species habitat preference in light of its precinction ranking. The null hypothesis used in these tests is that a species proportional abundance is equal within each of the three samples. Results were considered significant at the 95% confidence level. Results The results of the MannWhitney U tests comparing the ranked distribution of species lev el butterfly abundances between transects appear in Table 33. Figure 31 through 33 show the distribution of butterflies ranked by abundance for each transect. In these figures, butterfly species farther to the left are more widespread than those to the right. For all transects, the median was 8. Although the median for each transect was 8, if each individual butterfly, ordered by precinction rank, is listed for each sample, the ranks to the left and right of the median are different for each transect. In other words, despite equal medians, the ranks of to either side of the median are not equal. Of the three transects, the ranked distribution of Transect 2 (Fig. 32) contained lower ranks aside from the median, when abundance is accounted for, indicating lower levels of precinction on this transect. The ranks of Transect 1 (Fig. 31) and Transect 2 (Fig. 33) are approximately equal when abundance is accounted for, and are both distributions contain similar higher ranks to the right of the median. This ind icates that Transect 1 and 3 have similar precinction levels that are greater than Transect 2. In comparisons between transects using the MannWhitney U test, the ranked distribution of butterfly species collected on Transect 1 and Transect 2 were signific antly different, with greater levels of precinction observed on Transect 1 (Transect 1: median = 8, n = 563, IQR = 785

PAGE 86

16.5; Transect 2: median = 8 n = 473, IQR = 79; MannWhitney U test: Z = 3.18, P < 0.001). The tests also indicated that the distributions of Transect 2 and Transect 3 were significantly different (Transect 2: median = 8, n = 473, IQR = 79; Transect 3: median = 8, n = 490, IQR = 7 9; MannWhitney U test: Z = 2.5, P = 0.01). However, the test of Transect 1 and Transect 3 found no significant difference between distributions (Transect 1: median = 8, n = 563, IQR = 716.5; Transect 3: median = 8, n = 490, IQR = 7 9; MannWhitney U test: Z = 1.08, P = 0.27). In summary, Mann Whitney U tests found significant differences between the precinctionr anked abundance distributions of Transect 1 and Transect 2 and between Transect 2 and Transect 3. No significant difference was observed between Transect 1 and Transect 3. The results for comparisons of subspecies level ranked distribution butterfly abundances by transect appear in Table 34. The median values for Transect 1 and 2 were both 14, while the median value for Transect 3 was 15, indicating the highest level of precinction was found along this transect. The results of the MannWhitney U tests for subspecies level comparisons produced a pattern similar to those reported at the species level. The ranked distributions for Transect 1 and 2 were significantly different (Transect 1: median = 14, n =563, IQR = 215; Transect 2: median = 14, n = 473, IQR = 2 15; MannWhitney U test: Z = 2.2, P = .03), as were the ranked distribution of Transect 2 and Transect 3 (Transect 2: median = 14, n = 473, IQR = 215; Transect 3: median = 15, n = 490, IQR = 415; MannWhitney U test: Z = 3.45, P < 0.0005). As in the species level comparisons, the MannWhitney U test did not find significant differences in the ranked distributions between Transect 1 and Transect 3 (Transect 1: 86

PAGE 87

median = 14, n =563, IQR = 215; Transect 3: median = 15, n = 490, IQR = 415; MannWhitney U test: Z = 1.05, P = 0.24). Fishers exact test was used to identify species that were significantly associated with particular transects. These tests compared the proportion of individuals collected of a particular species, relative to all individuals of all species collected on a particular transect to the same proportion on the other transects. This test was performed individually on each of the species collected. Ten of the 21 species collected were significantly associated with a particular transect (Table 35). Only two species with a distribution not restricted to the Philippine archipelago were associated with a particular habitat. Melanitis leda the most widespread species, was most frequently collected on Transect 1. Further, its abundance prop ortion decreased from Transect 2 to Transect 3. Lethe chandica, a species that occurs throughout much of Southeast Asia was most frequently collected on Transect 3. Of the eight species precinctive to the Philippines, three species were more frequently col lected on Transect 1, two were more frequently collected on Transect 2 and three were more frequently collected on Transect 3. Of the three species with the smallest geographic distribution ( Mycalesis teatus, My. georgi and Tanaecia lupina, each receiving a precinction rank of 2), My. teatus was most frequently collected on Transect 3, My. georgi was most frequently collected on Transect 1 and T. lupina was most frequently collected on Transect 2. Discussion The results of the MannWhitney U tests show sig nificant differences in the ranked distribution of butterfly species between Transect 1 (secondary forest garden) and Transect 2 (post extraction secondary forest) and between Transect 2 and Transect 3 (old growth forest). There was no significant difference between Transect 1 and 87

PAGE 88

Transect 3. This pattern was observed at both the species and subspecies level. Further, at both taxonomic levels, t he ranks of Transect 2 were lower than Transect 1 and Transect 3, indicating that the lowest precinction levels were observed on this transect. Ranks were similar for Transect 1 and Transect 3, indicating similar precinction levels. Because these data were collected using baited traps as the sampling method, these results must be interpreted with caution. However, if these results are indeed reflective of true richness and abundance patterns at the sampling sites, they suggest that the greatest concentrations of precinctive species occur in the most pristine and most disturbed of the habitats sampled. Many studies com paring logged and unlogged forests or human dominated landscapes and closed canopy forests in insular and mainland Southeast Asia conclude that the greatest concentrations of precinctive species occur in the most undisturbed sites included in each particul ar study (Hill et al. 1995, Hamer et al. 1997, Ghazoul 2002, Fermon et al. 2003, Lien and Yuan 2003). The results of this study show that at the sample site in MKNP, secondary forests are utilized by rangerestricted species. This finding is in line with a smaller number of Southeast Asian butterfly disturbance studies that found precinction levels to be similar in logged and unlogged forests. Willott et al. (2000) found no evidence that there were shifts in the precinction levels of butterfly communities i n logged an unlogged forest. Veddeler et al. (2005) focused largely on secondary forests and concluded that secondary forests were valuable for the conservation of precinctive taxa. Much emphasis has been placed on the importance of old growth forests to the conservation of rangerestricted butterflies and other forest taxa (Fermon et al. 2005, 88

PAGE 89

Barlow et al. 2007, Gardner et al. 2007). While this emphasis is not unjustified, secondary forests, at least those contiguous with old growth tracts, can likely su pplement the habitat available to some rangerestricted species, in particular, those that are gap or canopy specialists, or species with preferences for less dense forests. In this study, five butterfly species were unique to the secondary forest transect s and collected only in traps set within those habitats. Three of these species are widespread throughout Southeast Asia and two were Philippine precinctives. The geographic distributions of Ptychandra negrosensis and My. igoleta are both restricted to the Philippines. Both species were only collected in secondary forest traps, albeit, with low abundances. Fishers exact test showed that the proportional abundances of both species relative to the total number of butterflies collected on each transect, were significantly different ( P. negrosensis : P = 0.021, My. igoleta : P < 0.0001) and both species peaked in proportional abundance in traps set in the secondary forest garden, the most disturbed habitat sampled. Mycalesis georgi another Philippine precinctive peaked in proportional abundance on the post extraction secondary forest transect, but was also found in traps in the other two habitats. There are 377 species (40% of the Philippines butterfly fauna) that are precinctive to the archipelago (Treadaway 2012). Of these species, there are certainly those that have evolved to favor open areas or forest gaps, like those that are common in secondary forests. Although this study was conducted over a small spatial scale, it provides evidence that the precinction levels of frugivorous butterfly communities in secondary forests are not necessarily lesser than those of old growth forests. If the spatial scale of this study were increased, it is possible that P. negrosensis and My. igoleta would have been detected in old growth sites. By 89

PAGE 90

increasing the spatial scale, variation in habitat heterogeneity within these forests could have been further taken into account. While it is unclear if these two precinctive species are present within the old growth forest at the study site. I t is apparent that they are able to utilize the 30year old secondary forests that form the margins of MKNP. Furthermore, the secondary forests sampled contained similar proportional abundances of several other rangerestricted species (i.e. Zethera musides T. lupina Lexias satrapes Discophora dodong and D. ogina ). No species were collected exclusively in old growth forest, however, two precinctive species ( P. leucogyne and My. teatus ) and one widespread species ( Le. Chandica) peaked in abundanc e in the traps placed in this habitat. Proportional abundance patterns of widespread species did not show a common preference for the most disturbed habitats sampled. Melanitis leda the most widespread species collected (distributed throughout much of As ia and parts of Africa and Australia) was significantly more abundant at the secondary sites than the old growth site. This species was most abundant in the secondary forest garden habitat and least abundant in old growth forest. As its abundance decreased in the samples, the abundances of two precinctive Melanitis species ( Me. atrax and Me. boisduvalia) increased. Melanitis leda received a precinction category of 13, while Me. atrax and Me. boisduvalia were placed in category 9 and 10, respectively. Although only three members of the genus were present at the study site, their abundance patterns indicate that the most widespread of the three decreased in abundance as disturbance increased, while the more rangerestricted species increased as disturbance dec reased. While this pattern was not typical of the other taxa collected, the patterns observed in 90

PAGE 91

Melanitis at the study site follow the results of studies conducted elsewhere (Thomas 1991, Lewis et al. 1998). Globally, there is concern that anthropogenic modification of habitats is causing a process of biotic homogenization that replaces precinctive species with widespread species (McKinney and Lockwood 1999). In this study, abundance patterns of the three Melanitis species present at the study site seem t o follow this trend, but the presence of two precinctive species unique to the secondary forest transects, along with the approximately equal abundances on all transects of seven precinctive species provides evidence that the level of disturbance present i n these secondary forests does not restrict precinctive species to old growth forest. While further study is required to elaborate on the conservation value of secondary forests in the Philippines, these results suggest optimism that secondary forests coul d provide a useful tool in the mitigation of the effects of forest loss on precinctive species in the Philippines. Considering the extent of old growth forest loss in the archipelago, it is increasingly necessary to understand the potential of secondary forests in the conservation of the Philippines precinctive species. Although compositional changes were observed among the butterfly communities within the secondary forest and old growth sites, the data provide little evidence that secondary forests are useless to precinctive components of the Negros frugivorous butterfly fauna. The presence of precinctive species in the two secondary forests sampled does not provide any indication of whether these habitats alone are enough to ensure long term survival of populations of these species. Further study is required before making recommendations on conservation strategies. In particular, future studies 91

PAGE 92

should investigate other taxa and other types of secondary forest. This study took place over a small spatial s cale and only investigated secondary forest tracts that were contiguous with a large fragment of old growth forest. Veddeler (2005) found evidence that even disjunct fragments of secondary forest on Sulawesi were able to support frugivorous butterfly assem blages that contained precinctive species. Such results provide support to the idea that secondary forests, particularly those that are managed for conservation purposes, have potential for mitigating the effects of forest loss on taxa that are of conservation concern (i.e. precinctives). This is not to downplay the importance of protecting old growth forest tracts, but to augment conservation management strategies by increasing the area of lands that hold conservation value. 92

PAGE 93

Table 31. Species level precinction ranking scheme. Species Prec Category Prec. Rank Melanitis leda 13 1 Lethe europa 12 2 Charaxes solon 12 3 Rhinopalpa polynice 12 4 Lethe chandica 11 5 Amathusia phidippus 11 6 Melanitis boisduvalia 10 7 Mycalesis tagala 10 8 Melanitis atrax 9 9 Mycalesis igoleta 8 10 Charaxes amycus 8 11 Ptychandra leucogyne 7 12 Faunis phaon 6 13 Lexias satrapes 6 14 Discophora ogina 4 15 Ptychandra negrosensis 3 16 Zethera musides 3 17 Discophora dodong 2 18 Tanaecia lupina 2 19 Mycalesis teatus 2 20 Mycalesis georgi 2 21 Species are ranked according to size of distribution. Species receiving a score of 10 and below are restricted to the Philippine Archipelago. 93

PAGE 94

Table 32. Subspecies level precinction ranking scheme. Subspecies Prec. Category Prec. Rank Melanitis leda leda 10 1 Melanitis boisduvalua boisduvalia 9 2 Lethe europa cevanna 9 3 Ptychandra leucogyne 7 4 Charaxes solon lampedo 5 5 Mycalesis igoleta negrosensis 4 6 Ptychandra negrosensis 3 7 Zethera mu sides 3 8 Discophora dodong 2 9 Discophora ogina pulchra 2 10 Faunis phaon carfinia 2 11 Lethe chandica canlaonensis 2 12 Lexias satrapes amlana 2 13 Melanitis atrax soloni 2 14 Mycalesis tagala mataurus 2 15 Mycalesis teatus 2 16 Rhinopalpa polyn ice panayana 2 17 Amathusia phidippus negrosensis 1 18 Charaxes amycus negrosensis 1 19 Mycalesis georgi 1 20 Tanaecia lupina howarthi 1 21 Subspecies are ranked according to size of distribution. Subspecies receiving a rank of 9 and below are restric ted to the Philippine Archipelago. Table 33. Species level results of MannWhitney U tests. Transects U value Z score P Significance 1 vs. 2 148423 3.18 < 0.001 1 vs. 3 143242.5 1.08 0.27 NS 2 vs. 3 126652 2.5 0.01 Asterisk indicates signific ance at P < 0.05, NS indicates not significant. Table 34. Subspecies level results of MannWhitney U tests. Transects U value Z score P Significance 1 vs. 2 143692 2.2 0.03 1 vs. 3 132769.5 1.05 0.24 NS 2 vs. 3 10091 3.45 < 0.0005 Asterisk i ndicates significance at P < 0.05, NS indicates not significant. 94

PAGE 95

Table 35. Results of Fishers exact test comparing proportional abundance of individual species between transects Species Prec. Rank P Significance Max. Abund. Melanitis leda 1 <0.0005 1 Lethe europa 2 0.11 NS 1 Charaxes solon 3 0.11 NS 1 Rhinopalpa polynice 4 0.8 NS 2 Lethe chandica 5 0.01 3 Amathusia phidippus 6 0.1 NS 2 Melanitis boisduvalia 7 <0.0002 2 Mycalesis tagala 8 <0.0002 3 Melanitis atrax 9 0.76 NS 1 Mycales is igoleta 10 <0.0002 1 Charaxes amycus 11 0.19 NS 2 Ptychandra leucogyne 12 0.002 3 Faunis phaon 13 0.3 NS 2 Lexias satrapes 14 0.8 NS 2 Discophora ogina 15 0.88 NS 1 Ptychandra negrosensis 16 0.02 1 Zethera musides 17 0.37 NS 2 Discophora d odong 18 0.06 NS 1 Tanaecia lupina 19 0.04 2 Mycalesis teatus 20 0.009 3 Mycalesis georgi 21 <0.0003 1 Asterisk indicates significance at P < 0.05, NS indicates not significant. Far right column indicates the transect where the maximum abundance of each species was observed. 95

PAGE 96

Figure 31. Proportional abundance distribution of butterfly species ranked by range size for Transect 1 (secondary forest garden). Species appearing farthest left have the largest global distribution while species fart hest right have the smallest global distribution. Species 721 are precinctive to the Philippines. Species bars that are the same color share the same precinction rank. 96

PAGE 97

Figure 32. Proportional abundance distribution of butterfly species ranked by range size for Transect 2 (post extraction secondary forest). Species appearing farthest left have the largest global distribution while species farthest right have the smallest global distribution. Species 721 are precinctive to the Philippines. Species bars that are the same color share the same precinction rank. 97

PAGE 98

Figure 33. Proportional abundance distribution of butterfly species ranked by range size for Transect 3 (old growth forest). Species appearing farthest left have the largest global distribut ion while species farthest right have the smallest global distribution. Species 721 are precinctive to the Philippines. Species bars that are the same color share the same precinction rank. 98

PAGE 99

Figure 34. Proportional abundance distribution of butterfly subspecies ranked by range size for Transect 1 (secondary forest garden). Subspecies appearing farthest left have the largest global distribution while subspecies farthest right have the smallest global distribution. Subspecies 221 are precinctive to the Philippines. Subspecies bars that are the same color share the same precinction rank. 99

PAGE 100

Figure 35. Proportional abundance distribution of butterfly subspecies, ranked by range size for Transect 2 (post extraction secondary forest). Subspecies appearing farthest left have the largest global distribution while subspecies farthest right have the smallest global distribution. Subspecies 221 are precinctive to the Philippines. Subspecies bars that are the same color share the same precinction rank. 100

PAGE 101

Figur e 36. Proportional abundance distribution of butterfly subspecies, ranked by range size for Transect 3 (old growth forest). Subspecies appearing farthest left have the largest global distribution while subspecies farthest right have the smallest global distribution. Subspecies 221 are precinctive to the Philippines. Subspecies bars that are the same color share the same precinction rank. 101

PAGE 102

CHAPTER 4 GENERAL CONCLUSIONS The Philippine Archipelago is a global conservation priority and the country represen ts a great challenge to the global conservation community. Historically, the archipelago was almost completely forested (Liu 1993). Over the past century, the Philippine landscape has undergone dramatic changes as logging and conversion of forests to agric ultural land have removed >90% of the countrys original forests (Heaney and Regaldo 1998). Concomitant with this dramatic decline in the Philippines old growth forest has been a sharp increase in the area of land occupied by secondary forest. The Philippines Forest Management Bureau currently classifies 20% of the countrys land area as forest (FMB 2010). Of this, the vast majority (83%) represents various types of secondary forest of unknown conservation value (Lasco et al. 2001). Currently conservation strategies and protected areas are designed to preserve remnant tracts of old growth forest. However, the prevalence of secondary forests in the Philippines may represent an opportunity to expand on these conservation strategies. There is a substantial am ount of evidence suggesting that secondary forests hold potential in mitigating the effects of deforestation on forest taxa (Willott et al. 2000, Veddeler et al. 2005, Wright and Muller Landau 2006, Bowen et al. 2007, Berry et al. 2000, Tabarelli et al. 2010, Pinotti et al. 2012). In a country, such as the Philippines, that has suffered critical levels of old growth forest loss, investigation of secondary forests as a conservation tool is particularly salient. This study investigated frugivorous butterfly communities in two types of secondary forest and an old growth forest. In terms of species richness, abundance and 102

PAGE 103

diversity, this study found weak differentiation between the frugivorous butterfly communities of a secondary forest garden, post extraction secondary forest and old growth forest. Measures of richness, abundance and diversity between samples from each of these habitats did not indicate significant differences in the community structure of these habitats, as sampled by butterfly traps. Cluster analysis and ordination by nonmetric multidimensional scaling showed poor grouping of trap sites by habitat, suggesting that these communities are not acutely distinct from one another. These findings are in contrast to those of several studies that have investigated Southeast Asian butterfly communities in logged and unlogged forests and found unlogged forest to contain significantly higher diversity than logged forests or selectively logged forests (Hill et al. 1995, Ghazoul 2002, Dumbrell and Hill 2005) While no significant differences were observed between the diversity, richness and abundance of these habitats, the highest levels of these three measures were observed in the most disturbed habitat sampled (secondary forest garden) and lowest in old growth forest, in agreement with other studies conducted in Southeast Asia (Walpole and Sheldon 1999, Willott et al. 2000, Lien and Yuan 2003, Cleary et al. 2005, Fermon et al. 2005). The differing spatial scales at which individual studies were conducted may explain the discrepancies in the results of these studies (Hamer and Hill 2001). These results suggest that the frugivorous butterfly communities of these habitats are relatively similar, thus providing evidence that secondary forests, at least those cont iguous with old growth tracts, may be useful habitats for forest taxa. The Philippine Archipelago holds an exceptional concentration of precinctive species. Species with geographic distributions that are restricted to the Philippines are 103

PAGE 104

conservation targets. In the Philippines, secondary forests will only be effective in mitigating the effects of forest loss on forest taxa if they are able to support populations of Philippine precinctives. This study also investigated if precinction levels within frugivoro us butterfly communities differed between the secondary forest garden, post extraction secondary forest and old growth forest. MannWhitney U tests showed significant differences in the precinctionranked species abundance distributions between the secondary forest garden and post extraction secondary forest samples and between the post extraction secondary forest and old growth samples. No significant difference was observed between the secondary forest garden and the old growth samples. This pattern was observed at the species and subspecies taxonomic levels. Precinction levels in the secondary forest garden and old growth samples were similar. Two precinctive species were unique to the secondary forest habitats and were never encountered in the old growth sample. In addition, six precinctive species had similar proportional abundances in the sample of each habitat. These results provide evidence that secondary forest tracts contiguous with old growth sites are utilized to a similar degree as old growth for ests by precinctive species. Although this study was conducted at a small spatial scale, these results provide reason for optimism that secondary forests may be useful to the precinctive component of the frugivorous butterfly fauna of Negros. Further study is required before recommendations can be made on the conservation value of these forests. In particular, these patterns should be investigated across a range of taxa and include additional types and ages of secondary forest. 104

PAGE 105

LIST OF REFERENCES Alcala, A. C. 2004. Biodiversity research in the Philippines from 19982003. Asean Biodiversity 4 :26 31. Allen, D., C. Oliveros, C. Espanola, G. Broad, and J. C. T. Gonzalez. 2004. A new species of Gallirallus from Calayan island, Philippines. Forktail 20 :1 7. A ndersen, A. N., B. D. Hoffman, W. J. Muller, and A. D. Griffiths. 2002. Using ants as bioindicators in land management: simplifying assessment of ant community responses. Journal of Applied Ecology 39:8 17. Anderson, S. 1994. Area and endemism. The Quarterly Review of Biology 69:451471. Bankoff, G. 2007. One island too many: reappraising the extent of deforestation in the Philippines prior to 1946. Journal of Historical Geography 33:314 334. Barlow, J., W. L. Overal, I. S. Araujo, T. A. Gardner, and C. A Peres. 2007. The value of primary, secondary and plantation forests for fruit feeding butterflies in the Brazilian Amazon. Journal of Applied Ecology 44:10011012. Benedick, S., J. K. Hill, N. Mustaffa, V. K. Chey, M. Maryati, J. B. Searle, M. Schilthui zen, and K. C. Hamer. 2006. Impacts of rain forest fragmentation on butterflies in northern Borneo: species richness, turnover and the value of small fragments. Journal of Applied Ecology 43:967 977. Berry, N. J., O. L. Phillips, S. L. Lewis, J. K. Hill, D. P. Edwards, N. B. Tawatao, N. Ahmad, D. Magintan, C. V. Khen, M. Maryati, R. C. Ong, and K. C. Hamer. 2010. The high value of logged tropical forests: lessons from northern Borneo. Biodiversity Conservation 19 :985 997. Bhat, D. M., K. S. Murali, and N. H. Ravindranath. 2001. Formation and recovery of secondary forests in India: a particular reference to Western Ghats in south India. Journal of Tropical Forest Science 13:601620. Black, S. H., M. Chepard, and M. M. Allen. 2001. Endangered invertebrates: the case for greater attention to invertebrate conservation. Endangered Species Update 18:42 50. Bonebrake, T. C., L. C. Ponisio, C. L. Boggs, and P. R. Ehrlich. 2010. More than just indicators: a review of tropical butterfly ecology and conservation. Bi ological Conservation 143 :18311841. Bossart, J. L., and C. E. Carlton. 2002. Insect conservation in America: status and perspectives. American Entomologist 48 :82 92. 105

PAGE 106

Bowen, M. E., C. A. McAlpine, A. P. N. House, and G. C. Smith. 2007. Regrowth forests o n abandoned agricultural land: a review of their habitat values for recovering forest fauna. Biological Conservation 140 :273296. Bowman, D. M. J. S., J. C. Z. Woinarski, D. P. A. Sands, A. Wells, and V. J. McShane. 1990. Slash and burn agriculture in the wet coastal lowlands of Papua New Guinea: response of birds, butterflies and reptiles. Journal of Biogeography 17: 227239. Brooks, T. M., T. D. Evans, G. C. L. Dutson, G. Q. A. Anderson, D. C. Asane, R. J. Timmins, and A. G. Toledo. 1992. The conservati on status of the bird of Negros, Philippines. Bird Conservation International 2 :273302. Brooks, T. M., R. A. Mittermeier, C. G. Mittermeier, G. A. B. da Fonseca, A. B. Rylands, W. R. Konstant, P. Flick, J. Pilgrim, S. Oldfield, G. Magin, and C. H. Hilton Taylor. 2002. Habitat loss and extinction in the hotspots of biodiversity. Conservation Biology 16:909 923. Brooks, T. M., R. A. Mittermeier, G. A. B. da Fonseca, J. Gerlach, M. Hoffman, J. F. Lamoreux, C. G. Mittermeier, J. D. Pilgrim, and A. S. L. Rodr igues. 2006. Global biodiversity conservation priorities. Science 313 :58 61. Brown, J. H. 1984. On the relationship between abundance and distribution of species. The American Naturalist 124 :255 279. Brown, K. S. 1997. Diversity, disturbance, and sustainable use of Neotropical forests: insects as indicators for conservation monitoring. Journal of Insect Conservation 1 :25 42. Brown, K. S., and R. W. Hutchings. 1997. Disturbance, fragmentation, and the dynamics of diversity in Amazonian forest butterflies. Pages 91110 in W. F. Laurance and R. O. Bierregard, editors. Tropical forest remnants: ecology, management and conservation of fragmented communities. Chicago University Press, Chicago, Illinois. Brown, K. S., and A. J. L. Freitas. 2000. Atlantic forest butterflies: indicators for landscape conservation. Biotropica 32:934956. Brown, S., and A. E. Lugo. 1990. Tropical secondary forests. Journal of Tropical Ecology 6 :1 32. Brummitt, N., and E. N. Lughadha. 2003. Biodiversity: Wheres hot and wheres not. Conservation Biology 17:14421448. Castelletta, M., N. S. Sodhi, and R. Subaraj. 2000. Heavy extinctions of forest avifauna in Singapore: Lessons for biodiversity conservation in Southeast Asia. Conservation Biology 14:18701880. 106

PAGE 107

Charrette, N. A., D. F. R. Cleary, and A. O. Mooers. 2006. Rangerestricted, specialist Bornean butterflies are less likely to recover from ENSO induced disturbance. Ecology 9 :23302337. Chazdon, R. L. 2003. Tropical forest recovery: legacies of human impact and natural disturbances. Perspectives in Plant Ecology: Evolution and Systematics 6 :51 71. Chokkalingam, U., and W. de Jong. 2001. Secondary forest: a working definition and typology. International Forestry Reviews 3 :19 26. Cleary, D. F. R., and M. J. Genner. 2004. Changes in rain forest butterfly diversity following major ENSO induced fires in Borneo. Global Ecology and Biogeography 13:129 140. Cleary, D. F. R., and A. Grill. 2004. Butterfly response to severe ENSO induced forest fires in Borneo. Ecological Entomology 29:666 676. Cleary, D. F. R., T. J. B. Boyle, T. Setyawati, and S. B. J. Menken. 2005. The impact of logging on the abundance, species richness and community composition of butterfly guilds in Borneo. Journal of Applied Entomology 1 :52 59. Cleary, D. F. R. M. J. Genner, L. P. Koh, T. J. B. Boyle, T. Setyawati, R. de Jong, and S. B. J. Menken. 2009. Butterfly species and traits associated with selectively logged forest in Borneo. Basic and Applied Ecology 10:237245. Colwell, R. K. 2005. EstimateS: statisti cal estimation of species richness and shared species from samples. Version 7.5. Users Guide and application published at http://purl.oclc.org/estimates Colwell, R. K., G. Brehm, C. L. Cardelus, A. C. Gilman and J. T. Longino. 2008. Global warming, elevational range shifts, and lowland biotic attrition in the wet tropics. Science 322 :258 261. Danielsen, F., and C. G. Treadaway. 2004. Priority conservation areas for butterflies (Lepidoptera, Rhopalocera) in the Philippines islands. Animal Conservation 7 :79 92. de Jong, R., and C. G. Treadaway. 1993. The Hesperiidae (Lepidoptera) of the Philippines. Zoologische Verhandelingen 288 :1 139. Dent, D. H., and S. J. Wright. 2009. The future of tropical species in secondary forests: a quantitative review. Biological Conservation 142 :28332843. Devictor, V., R. Juliard, and F. Jiguet. 2008. Distribution of specialist and generalist species along spatial gradients of habitat disturbance and fragmentation. Oikos 117:507514. 107

PAGE 108

DeVries, P. J. 1988. Stratification of fruit feeding nymphalid butterflies in a Costa Rican rainforest. Journal of Research on the Lepidoptera 26:98 108. DeVries, P. J., D. Murray, and R. Lande. 1997. Species diversity in vertical, horizontal, and t emporal dimensions of a fruit feeding butterflies community in an Ecuadorian rainforest. Biological Hournal of the Linnean Society 62:343364. Dinerstein, E., E. D. Wikramanajake, and M. Forney. 1995. Conserving the resevoirs and remnants of tropical mois t forest in the IndoPacific region. Pages 140175 in R. B. Primack and T. E. Lovejoy, editors. Ecology, Conservation and Management of Southeast Asian Rainforests. Yale University Press, New Haven, Connecticut. Douglas, J. M., T. W. Cronin, T. Chiou, and N. J. Dominy. 2007. Light habitats and the role of polarized iridescence in the sensory ecology of neotropical nymphalid butterflies (Lepidoptera: Nymphalidae). The Journal of Experimental Biology 210:788799. Dumbrell, A. J., and J. K. Hill. 2005. Impac ts of selective logging on canopy and ground assemblages of tropical forest butterflies: implications for sampling. Biological Conservation 125 :123131. Dunn, R. R. 2004. Recovery of faunal communities during tropical forest regeneration. Conservation Biology 18:302 309. Dunn, R. R. 2005. Modern insect extinctions, the neglected majority. Conservation Biology 19:10301036. Edwards, D. P., T. H. Larsen, T. D. S. Docherty, F. A. Ansell, W. W. Hsu, M. A. Derhe, K. C. Hamer, and D. S. Wilcove. 2011. Degraded lands worth protecting: the biological importance of Southeast Asias repeatedly logged forests. Proceedings of the Royal Society B 278:82 90. Edwards, D. P., A. R. Blackhouse, C. Wheeler, C. V. Khen, and K. C. Hamer. 2012. Impacts of logging and rehabili tation on invertebrate communities in tropical rainforests of northern Borneo. Journal of Insect Conservation 16 :591 599. Fearnside, P. M., and W. F. Laurance. 2004. Tropical deforestation and greenhousegas emissions. Ecological Applications 14:982 986. Fermon, H., M. Waltert, R. I. Vane Wright, and M. Muhlenberg. Forest use and vertical stratification in frui feeding butterflies of Sulawesi, Indonesia: impacts for conservation. Biodiversity and Conservation 14:333 350. Fernandez, J. M., M. A. E. Selma, F. R. Aymerich, M. T. P. Saez, and M. F. C. Fructoso. 2005. Aquatic birds as bioindicators of trophic changes and ecosystem deterioration in the Mar Menor lagoon (SE Spain). Hydrobiologia 550 :221 235. 108

PAGE 109

Fjeldsa, J. 1999. The impact of human forest disturbance on the endemic avifauna of the Udzungwa Mountains, Tanzania. Bird Conservation International 9 :47 62. Fonseca, C. R. 2009. The silent mass extinction of insect herbivores in biodiversity hotspots. Conservation Biology 23:15071515. Food and Agriculture Organization (FAO). 2010. Global forest resources assessment 2010 main report. Forestry paper 163. FAO, New York. Available from http://www.fao.org/docrep/013/i1757e/i1757e.pdf (accessed Dec ember 2012). Forest Management Bureau (FMB). 2010. Forestry Statistics 2010. Forest Management Bureau. Quezon City, Philippines. Available from http://forestry.denr.gov.ph/2010PFS.pdf (accessed November 2012). Foster, S. D., and P. K. Dunstan. 2010. The analysis of biodiversity using rank abundance distributions. Biometrics 66 :186 195. Frank, J. H., and E. D. McCoy. 1990. Introduction to attack and defense: behavioral ecology of predators and their prey Endemics and epidemics of shibboleths and other things causing chaos. The Florida Entomologist 73 :1 9. Galang, R. 2004. A critical review of wildlife conservation in the Philippines. Philippine Spotted Deer Conservation Foundation, Melbourne, Australia. Gardner, T. A., J. Barlow, L. W. Parry, and C. A. Peres. 2007. Predicting the uncertain future of tropical forest species in a data vacuum. Biotropica. 39:25 30. Gardner, T. A., J. Barlow, R. Chazdon, R. M. Ewers, C. A. Harvey, C. A. Peres, and N. S. Sod hi. 2009. Prospects for tropical forest biodiversity in a human modified world. Ecology Letters 12:561 582. Ghazoul, J. 2002. Impact of logging on the richness and diversity of forest butterflies in a tropical dry forest in Thailand. Biodiversity and Cons ervation 11:521 541. Gibb, H., and D. F. Hochuli. 2002. Habitat fragmentation in an urban environment: large and small fragments support different arthropod assemblages. Biological Conservation 106 :91 100. Gotelli, N. J., and R. K. Colwell. 2001. Quantif ying biodiversity: procedutes and pitfalls in the measurement and comparison of species richness. Ecology Letters 4 :379 391. Guariguata, M. R., and R. Ostertag. 2001. Neotropical secondary forest succession: changes in structural and functional characteris tics. Forest Ecology and Management 148 :185 206. 109

PAGE 110

Hamann, A., E. B. Barbon, E. Curio, and D. A. Madulid. 1999. A botanical inventory of a submontane tropical rainforest on Negros Island, Philippines. Biodiversity and Conservation 8 :10171031. Hamer, K. C. J. K. Hill, L. A. Lace, and A. M. Langan. 1997. Ecological and biogeographical effects of forest disturbance on tropical butterflies of Sumba, Indonesia. Journal of Biogeography 24:67 75. Hamer, K. C., and J. K. Hill. 2000. Scaledependent effects of ha bitat disturbance on species richness in tropical forests. Conservation Biology 14:14351440. Hammer, ., D. A. T. Harper, and P. D. Ryan. 2001. PAST: paleontological statistics software package for education and data analysis. Palaeontologia Electronica 4 :1 9. Harcourt, A. H., S. A. Coppeto, and S. A. Parks. 2002. Rarity, specialization and extinction in primates. Journal of Biogeography 29:445 456. Heaney, L. R., and J. C. Regaldo. 1998. Vanishing Treasures of the Philippine Rainforest. The Field Museum Chicago, Illinois. Hill, J. K., K. C. Hamer, L. A. Lace, and W. M. T. Banham. 1995. Effects of selective logging on tropical forest butterflies on Buru, Indonesia. Journal of Applied Ecology 32:754 760. Hill, J. K., and K. C. Hamer. 1998. Using species abundance models as indicators of habitat disturbance in tropical forests. Journal of Applied Ecology 35:458460. Hill, J. K. 2001. Butterfly spatial distribution and habitat requirements in a tropical forest: impacts of selective logging. Journal of Appli ed Ecology 36:564 572. Horner Devine, M. C., G. C. Daily, P. R. Ehrlich, and C. L. Boggs. Countryside biogeography of tropical butterflies. 2003. Conservation Biology 17:168177. Howard, P. C., P. Viskanic, T. R. B. Davenport, F. W. Kigneyi, M. Baltzer, C J. Dickinson, J. S. Lwanga, R. A. Mathews, and A. Balmford. 1998. Complementarity and the use of indicator groupds for reserve selecation in Uganda. Nature 394 :472 495. Hughes, J. B., G. C. Daily, and P. R. Ehrlich. 2000. Conservation of insect diversity : a habitat approach. Conservation Biology 14:17881797. International Union for the Conservation of Nature (IUCN). 2010. Guidelines for using the IUCN red list categories and criteria. Version 8.1. IUCN Standards and Petitions Subcommittee, Gland, Switzerland. Jablonski, D. 1991. Extinctions: a paleontological perspective. Science 253 :754. 110

PAGE 111

Jones, M. J., M. S. Sullivan, S. J. Marsden, and M. D. Linsley. 2001. Correlates of extinction risk of birds from two Indonesion islands. Biological Journal of the Lin nean Society 73 :65 79. Juliard, R., J. Clavel, V. Devictor, F. Jiguet, and D. Couvet. 2006. Spatial segregation of specialists and generalists in bird communities. Ecology Letters 9 :1237 1244. Kadlec, T., M. A. A. M. Kotela, I. Novak, M. Konvicka, and V. Jarosik. 2009. Effect of land use and climate on the diversity of moth guilds with different habitat specialization. Community Ecology 10:152158. Kennedy, R. S., P. C. Gonzales, E. C. Dickinson, H. C. Miranda Jr., and T. H. Fisher. 2000. A Guide to the Birds of the Philippines. Oxford University Press, New York, New York. Kerr, J. T. 1997. Species richness, endemism, and the choice of areas for conservation. Conservation Biology 11:10941100. Kitahara, M., and K. Fujii. 1994. Biodiversity and community structure of temperate butterfly species within a gradient of human disturbance: an analysis based on the concept of generalist vs. specialist strategies. Researches on Population Ecology 36:187 199. Koh, L. P., N. S. Sodhi, and B. W. Brook. 2004. Ecolog ical correlates of extinction proneness in tropical butterflies. Conservation Biology 18:15711578. Koh, L. P. 2007. Impacts of land use change on Southeast Asian forest butterflies: a review. Journal of Applied Ecology 44:703713. Kotze, D. J., and R. B OHara. 2003. Species decline: but why? Explanations of carabid beetle (Coleoptera, Carabidae) declines in Europe. Oecologia 135 :138 148. Kremen, C. 1994. Biological inventory using target taxa: a case study of the butterflies of Madagascar. Ecological Applications 4 :407422. Kummer, D. M. 1992. Measuring forest decline in the Philippines: an exercise in historiography. Forest and Conservation History 36:185189. Kummer, D. M., and B. L. Turner II. 1994. The human causes of deforestation in Southeast A sia. Bioscience 44:323328. Lasco, R. D., R. G. Visco, and J. M. Pulhin. 2001. Secondary forests in the Philippines: formation and transformation in the 20th century. Journal of Tropical Forest Science 13:652 670. Laurance, W. F. 1999. Reflections on the tropical deforestation crisis. Biological Conservation 91 :109 117. 111

PAGE 112

Laurance, W. F. 2006. Have we overstated the tropical biodiversity crisis? Trends in Ecology and Evolution 22:65 70. Lawrence, A. 1997. Kaingin in the Philippines: is it the end of the for est? Network paper 21f. Rural Development Forestry Network, London, United Kingdom. Lawton, J. H., D. E. Bignell, B. Bolton, G. F. Bloemers, P. Eggleton, P. M. Hammond, M. Hodda, R. D. Holt, T. B. Larsen, N. A. Mawdsley, N. E. Stork, D. S. Srivastava, and A. D. Watt. 1998. Biodiversity inventories, indicator taxa and effects of habitat modification in tropical forest. Nature 391 :72 76. Lewis, O. T., R. J. Wilson, and M. C. Harper. 1998. Endemic butterflies on Grande Comore: habitat preferences and conservat ion priorities. Biological Conservation 85:113 121. Lewis, O. T. 2001. Effects of experimental selective logging on tropical butterflies. Conservation Biology 15:389 400. Lien, V. V., and D. Yuan. 2003. The differences of butterfly (Lepidoptera, Papilion oidea) communities in habitats with various degrees of disturbance and altitudes in tropical forests of Vietnam. Biodiversity and Conservation 12:10991111. Liu, D. S., L. R. Iverson, and S. Brown. 1993. Rates and patterns of deforestation in the Philippines: application of geographic information system analysis. Forest Ecology and Management 57:1 16. Losey, J. E., and M. Vaughan. 2006. The economic value of ecological services provided by insects. BioScience 56 :311 323. Magurran, A. 2004. Measuring Biol ogical Diversity. Blackwell Publishing Company, Malden, Massachusetts. Mather, A. S. 1992. The forest transition. Area 24 :367 379. Mather, A. S. 2001. The transition from deforestation to reforestation in Europe. Pages 3552 in A. Angelsen and D. Kaimowi tz, editors. Agricultural Technologies and Deforestation. Center for International Forestry Research. CABI Publishing, Wallingford, United Kingdom. Mather, A. S. 2007. Recent Asian forest transitions in relation to forest transition theory. International F orestry Review 9 :491502. McGeoch, M. A. 1998. The selection, testing and application of terrestrial insects as bioindicators. Biological Reviews 73:181201. McKinney, M. L., and J. L. Lockwood. 1999. Biotic homogenization: a few winners replacing many l osers in the next mass extinction. Trends in Ecology and Evolution 14 :450 453. 112

PAGE 113

Miller, J. C. 1993. Insect natural history, multi species interactions and biodiversity in ecosystems. Biodiversity and Conservation 2 :233241. Miller, S. W., P. Budy, and J. C Schmidt. 2010. Quantifying macroinvertebrate responses to instream habitat restoration: applications of metaanalysis to river restoration. Restoration Ecology 18:8 19. Millsap, B. A., J. A. Gore, D. E. Runde, and S. I. Cerulean. 1990. Setting prioriti es for conservation of fish and wildlife species in Florida. Wildlife Monographs 11 :3 57. Mittermeier, R. A., N. Myers, J. B. Thomsen, G. A. B. da Fonseca, and S. Olivieri. 1998. Biodiversity hotspots and major tropical wilderness area: approaches to sett ing conservation priorities. Conservation Biology 12:516 520. Molleman, F., A. Kop, P. M. Brakefield, P. J. de Vries, and B. J. Zwaan. 2006. Vertical and temporal patterns of biodiversity of fruit feeding butterflies in a tropical forest in Uganda. Biodiv ersity and Conservation 15 :93 107. Munday, P. L. 2004. Habitat loss, resource specialization, and extinction on coral reefs. Global Change Biology 10:16421647. Myers, N. 1988. Threatened biotas: hot spots in tropical forests. The Environmentalist 8 :18 7 208. Myers, N. 1993. Tropical forests: the main deforestation fronts. Environmental Conservation 20 :9 16. Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. da Fonseca, and J. Kent. 2000. Biodiversity hotspots for conservation priorities. Nature 403 :853858. Myers, N. 2003. Biodiversity hotspots revisited. BioScience 53 :916 917. Nagendra, H. 2002. Opposite trends in response for the Shannon and Simpson indices of landscape diversity. Applied Geography 22:175186. New, T. R. 1993. Angels on a pin: dimensions of the crisis in invertebrate conservation. American Zoologist 33:623630. New, T. R., R. M. Pyle, J. A. Thomas, and C. D. Thomas. 1995. Butterfly conservation management. Annual Review of Entomology 40:57 83. New, T. R. 1999. Untangling the web: spiders and the challenges of invertebrate conservation. Journal of Insect Conservation 3 :251 256. New, T. R. 2011. Launching and steering flagship Lepidoptera for conservation benefit. Journal of Threatened Taxa 3 :18051817. 113

PAGE 114

Okano, M., T. Okano, an d P. Pizarras. Butteflies taken on Bohol Island, the Philippines (II). Artes Liberales 44:131149. Olden, J. D., N.L. Poff, M. R. Douglas, M. E. Douglas, and K. D. Fausch. 2004. Ecological and evolutionary consequences of biotic homogenization. Trends in Ecology and Evolution 19:18 24. Ovadia, O. 2003. Ranking hotspots of varying sizes: a lesson from the nonlinearity of the species area relationship. Conservation Biology 17:14401441. Pardini, R., D. Faria, G. M. Accacio, R. R. Laps, E. Mariano Neto, M. L. B. Paciencia, M. Dixo, and J. Baumgarten. 2009. The challenge of maintaining Atlantic forest biodiversity: a multi taxa conservation assessment of specialist and generalist species in an agroforestry mosaic in southern Bahia. Biological Conservation 142 :11781190. Pearce, J. L., and L. A. Venier. 2006. The use of ground beetles (Coleoptera: Carabidae) and spiders (Araneae) as bioindicators of sustainable forest management: a review. Ecological Indicators 6 :780 793. Perz, S. G., and D. L. Skole. 2003. Secondary forest expansion in the Brazilian Amazon and the refinement of forest transition theory. Society and Natural Resources: An International Journal 16 :277 294. Peterson, A. T., and D. M. Watson. 1998. Problems with areal definitions of endemism: th e effects of spatial scaling. Diversity and Distributions 4 :189 194. Pinotti, B. T., C. P. Pagotto, and R. Pardini. 2012. Habitat structure and food resources for wildlife across successional stages in a tropical forest. Forest Ecology and Management 283 : 119 127. Posa, M. R. C., and N. S. Sodhi. 2006. Effects of anthropogenic land use on forest birds and butterflies in Subic Bay, Philippines. Biological Conservation 129:256270. Posa, M. R. C., A. C. Diesmos, N. S. Sodhi, and T. M. Brooks. 2008. Hope for threatened tropical biodiversity: lessons from the Philippines. BioScience 58:231240. Prendergast, J. R., and B. C. Eversham. 1997. Species richness covariance in higher taxa: empirical tests of the biodiversity indicator concept. Ecography 20:210216. Purvis, A., J. L. Gittleman, G. Cowlishaw, and G. M. Mace. 2000. Predicting extinction risk in declining species. Proceedings of the Royal Society B 267 :19471952. Pyle, R., M. Bentzien, and P. Opler. 1981. Insect conservation. Annual Review of Entomology 26:233258. 114

PAGE 115

Rainio, J., and J. Niemala. 2003. Ground beetles (Coleoptera, Carabidae) as bioindicators. Biodiversity and Conservation 12:487 506. Rosenberg, D. M., H. V. Danks, and D. M. Lehmkuhl. 1986. Importance of insects in environmental impact assess ment. Environmental Management 10 :773783. Rudel, T. K., O. T. Coomes, E. Moran, F. Achard, A. Angelsen, J. Xu, and E. Lambin. 2005. Forest transitions: towards a global understanding of land use change. Global Environmental Change 15 :23 31. Schroeder, H G., and C. G. Treadaway. 2005. Butterflies of the World, Part 20: Nymphalidae IX, Amathusiini of the Philippine Islands. Hillside Books, Canterbury, United Kingdom. Schulze, C. H., K. E. Linsenmair, and K. Fiedler. 2001. Understorey versus Canopy: patter ns of vertical stratification and diversity among Lepidoptera in a Bornean rain forest. Plant Ecology 153:133 152. Schulze, C. H., M. Waltert, P. J. A. Kessler, R. Pitopang, Shahabuddin, D. Veddeler, M. Muhlenberg, S. R. Gradstein, C. Leuschner, I. Steffa n Dewenter, and T. Tscharntke. 2004. Biodiversity indicator groups of tropical landuse systems: comparing plants, birds, and insects. Ecological Applications 14: 13211333. Shahabuddin, G., G. A. Herzner, C. Aponte R., and M. D. C. Gomez. 2000. Persistenc e of a frugivorous butterflies species in Venezuelan forest fragments: the role of movement and habitat quality. Biodiversity and Conservation 9 :1623 1641. Shahabuddin, G., and C. A. Ponte. 2005. Frugivorous butterfly species in tropical forest fragments: correlates of vulnerability to extinction. Biodiversity and Conservation 14:11371152. Sodhi, N. S., L. P. Koh, R. Clements, T. C. Wanger, J. K. Hill, K. C. Hamer, Y. Clough, T. Tscharntke, M. R. C. Posa, and T. M. Lee. 2010. Conserving Southeast Asian bio diversity in humanmodified landscapes. Biological Conservation 143 :23752384. Spitzer, K., J. Jaros, J. Havelka, and J. Leps. 1997. Effect of small cale disturbance on butterfly communities of an Indochinese montane rainforest. Biological Conservation 80 : 9 15. Stephenson, P. J. 1993. The small mammal fauna of Reserve Speciale dAnalamazaotra, Madagascar: the effects of human disturbance on endemic species diversity. Biodiversity and Conservation 2 :603 615. Suarez, R. K., and P. E. Sajise. 2010. Deforestation, swidden agriculture and Philippine biodiversity. Philippine Science Letters 3 :91 99. 115

PAGE 116

Tabarelli, M., A. V. Aguiar, M. C. Ribeiro, J. P. Metzger, and C. A. Peres. 2010. Prospects for biodiversity conservation in the Atlantic Forest: lessons from aging humanmodified landscapes. Biological Conservation 143 :23282340. Tangah, J, J. K. Hill, K. C. Hamer, and M. M. Daawood. 2004. Vertical distribution of fruit feeding butterflies in Sabah, Borneo. Sepilok Bulletin 1 :17 27. Thomas, C. D. 1991. Habitat use and geographic ranges of butterflies from the wet lowlands of Costa Rica. Biological Conservation 55 :269 281. Thomas, J. A. 2005. Monitoring change in the abundance and distribution of insects using butterflies and other indicator groups. Philosophical Transactions of the Royal Society B 360 :339 357. Treadaway, C. G. 2012. Revised checklist of the butterflies of the Philippine Islands (Lepidoptera, Rhopalocera). Nachricten des entomologischen Vereins Apollo Supplement 14:7 118. Tsukada, E. 1982. Butterflies of South East Asian Islands. Volume III, Satyridae Libytheidae. Plapac Company, Tokyo, Japan. UeharaPrado, M., K. S. Brown, and A. V. L. Freitas. 2007. Species richness, composition and abundance of fruit feeding butterflies in the Brazilion Atlantic F orest: comparison between a fragmented and a continuous landscape. Global Ecology and Biogeography 16:43 54. UeharaPrado, M., J. D. O. Fernandes, A. D. M. Bello, G. Machado, A. J. Santos, F. Z. Vaz deMello, and A. V. L. Freitas. 2009. Selecting terrestr ial arthropods as indicators of small scale disturbance: a first approach in the Brazilian Atlantic Forest. Biological Conservation 142 :12201228. Vallejo, B. 2011. The Philippines in Wallacea. Biodiversity, Biogeography and Nature Conservation in Wallacea and New Guinea 1 :27 42. VaneWright, R. I. 1990. The Philippines key to the biogeography of Wallacea? Pages 1934 in W. J. Knight and J. D. Holloway, editors. Insects and the Rain Forests of Southeast Asia (Wallacea). Royal Entomological Society of London, London, United Kingdom. Veddeler, D., C. H. Schulze, I. Steffan Dewenter, D. Buchori, and T. Tscharntke. 2005. The contribution of tropical secondary forest fragments to the conservation of fruit feeding butterflies: effects of isolation and age. Bi odiversity and Conservation 14 :35773592. Vitug, M. D. 1993. Power from the forest: the politics of logging. Philippine Center for Investigative Journalism. Manila, Philippines. 116

PAGE 117

Vitug, M. D. 1998. The politics of logging in the Philippines. Pages 122138 in P. Hirsch and C. Warren, editors. The Politics of Logging in Southeast Asia. Asia Research Centre on Social, Political and Economic Change. Routledge, London, United Kingdom. Walpole, M. J., and I. R. Sheldon. 1999. Sampling butterflies in tropical rai nforest: an evaluation of a transect walk method. Biological Conservation 87:85 91. Warren, M. S., J. K. Hill, J. A. Thomas, J. Asher, R. Fox, B. Huntley, D. B. Roy, M. G. Telfer, S. Jeffcoate, P. Harding, G. Jeffcoate, S. G. Willis, J. N. Greatorex Davie s, D. Moss, and C. D. Thomas. 2001. Rapid response of British butterflies to opposing forces of climate and habitat change. Nature 414 :65 69. Welton, L. J., C. D. Siler, D. Bennett, A. Diesmos, M. R. Duya, R. Dugay, E. L. B. Rico, M. V. Weerd, and R. M. B rown. 2010. A spectacular new Philippines monitor lizard reveals a hidden biogeographic boundary and a novel flagship species for conservation. Biology Letters 6 :654 658. Wijesinghe, M. R., and M. de L. Brooke. 2005. Impact of habitat disturbance on the di stribution of endemic species of small mammals and birds in a tropical rain forest in Sri Lanka. Journal of Tropical Ecology 21:661668. Willott, S. J., D. C. Lim, S. G. Compton, and S. L. Sutton. 2000. Effects of selective logging on butterflies of a Bor nean rainforest. Conservation Biology 14:10551065. Wilson, E. O. 1987. The little things that run the world (the importance and conservation of invertebrates). Conservation Biology 4 :344346. Wilson, K. A., M. F. McBride, M. Bode, and H. P. Possingham. 2 006. Prioritizing global conservation efforts. Nature 16:337340. Wood, B., and M. P. Gilman. 1998. The effects of disturbance on forest butterflies using two methods of sampling in Trinidad. Biodiversity and Conservation 7 :597 616. Wright, S. J. 2005. Tro pical forests in a changing environment. Trends in Ecology and Evolution 20 :553 560. Wright, S. J. 2010. The future of tropical forests. Annals of the New York Academy of Sciences 1195:1 27. Wright, S. J., and H. C. Muller Landau. 2006. The future of tro pical forest species. Biotropica 38:287 301. 117

PAGE 118

BIOGRAPHICAL SKETCH Lawrence grew up in New York, NY, Chapel Hill, NC and Gainesville, FL. He received his bachelors degree in conservation biology from the State University of New Yorks College of Environm ental Science and Forestry in 2006. Since 2009, Lawrence has been working to promote conservation in the Philippines and began studies at the University of Florida in 2010. In 2012, he was awarded a graduate research fellowship through the National Science Foundation. 118