MORPHO-TAXONOMY, BEHAVIOUR AND ECOLOGY OF THE RAJAH BROOKE’S BIRDWING (Trogonoptera brookiana) IN PENINSULAR MALAYSIA
PHON CHOOI KHIM
FACULTY OF SCIENCE UNIVERSITY OF MALAYA UniversityKUALA LUMPUR of Malaya
2018 MORPHO-TAXONOMY, BEHAVIOUR AND ECOLOGY OF THE RAJAH BROOKE’S BIRDWING (Trogonoptera brookiana) IN PENINSULAR MALAYSIA
PHON CHOOI KHIM
THESIS SUBMITTED IN FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
INSTITUTE OF BIOLOGICAL SCIENCES FACULTY OF SCIENCE UNIVERSITY OF MALAYA KUALA LUMPUR University of Malaya
2018 UNIVERSITY OF MALAYA
ORIGINAL LITERARY WORK DECLARATION
Name of Candidate: Phon Chooi Khim
Registration/Matric No: SHC110022
Name of Degree: Doctor of Philosophy
Title of Project Paper/Research Report/Dissertation/Thesis (“this Work”):
MORPHO-TAXONOMY, BEHAVIOUR AND ECOLOGY OF THE RAJAH BROOKE’S BIRDWING (Trogonoptera brookiana) IN PENINSULAR MALAYSIA
Field of Study: Ecology and Biodiversity
I do solemnly and sincerely declare that:
(1) I am the sole author/writer of this Work; (2) This Work is original; (3) Any use of any work in which copyright exists was done by way of fair dealing and for permitted purposes and any excerpt or extract from, or reference to or reproduction of any copyright work has been disclosed expressly and sufficiently and the title of the Work and its authorship have been acknowledged in this Work; (4) I do not have any actual knowledge nor do I ought reasonably to know that the making of this work constitutes an infringement of any copyright work; (5) I hereby assign all and every rights in the copyright to this Work to the University of Malaya (“UM”), who henceforth shall be owner of the copyright in this Work and that any reproduction or use in any form or by any means whatsoever is prohibited without the written consent of UM having been first had and obtained; (6) I am fully aware that if in the course of making this Work I have infringed any copyright whether intentionally or otherwise, I may be subject to legal action or any other action as may be determined by UM.
Candidate’s Signature Date: University of Malaya Subscribed and solemnly declared before,
Witness’s Signature Date:
Name:
Designation:
ii MORPHO-TAXONOMY, BEHAVIOUR AND ECOLOGY OF THE RAJAH
BROOKE’S BIRDWING (Trogonoptera brookiana) IN PENINSULAR
MALAYSIA
ABSTRACT
The Rajah Brooke’s Birdwing, Trogonoptera brookiana is a well-known ecotourism
icon. However, habitat degradation and over collecting have threatened the existence of
this birdwing, necessitating an in-depth study of the species. This research investigated
the taxonomy, behaviour and ecology of the subspecies albescens and mollumar in
Peninsular Malaysia, with the intention of enhancing conservation efforts. The status of
the eastern peninsular population of Trogonoptera brookiana, described as a new
subspecies mollumar D’Abrera has in the past been debated, resulting in it being
synonymised with the Sumatran subspecies, trogon Vollenhoven. The findings from
this work elucidate without doubt distinguishable taxonomic characters that support the
re-instatement of its subspecific status. The distribution of subspecies albescens and
mollumar was assessed and seen to be shrinking, reflecting a population decline, when
contrasting the recent (year 2000 onwards) and earlier (before year 2000) extent of its
distribution. To assist in understanding this insect and in its conservation, behavioural
studies were conducted. The subspecies albescens was studied in two selected sites, that
is Ulu Geroh village and Kenaboi Forest Reserve. Detailed information on daily activity
rhythms, such as flight and puddling behaviour in relation to environmental variables Universitywas revealed by the study. The interactions of between malesMalaya and females were studied ad libitum during all the field works. The birdwings preferred Bauhinia flowers over
hibiscus, but more interestingly, there were differences between sexes in their nectaring
rhythm in relation to feeding bouts and social feeding behaviour. All the interactions
between males and females were initiated by males, and competition for mates was a
common phenomenon. Different mate locating strategies were applied by the males in
iii different localities. Some acceptance and rejection behaviours by females were
documented. A novel monitoring method using digital photography, was developed for
the puddling males of subspecies albescens in Ulu Geroh, and tested for 24 months. It
showed for the first time that puddling activity can be used as a population index. The
birdwing population fluctuated greatly during monitoring, with peaks in May for both
years and an additional peak in the month of August in the second year. The population
counts were significantly associated with monthly mean temperature. Additionally,
transect walk and single-point observation methods were utilised to monitor birdwings
in flight in Kenaboi Forest Reserve, generating 29-months of monitoring data at this
site. The results reported here will aid in the formulation of conservation measures for
this charismatic birdwing species that has great potential as an ecotourism icon in this
region.
Keywords: Taxonomy, behaviour, ecology, Trogonoptera brookiana, Peninsular
Malaysia.
University of Malaya
iv MORFO-TAKSONOMI, KELAKUAN DAN EKOLOGI ‘RAJAH BROOKE’S
BIRDWING’ (Trogonoptera brookiana) DI SEMENANJUNG MALAYSIA
ABSTRAK
Kupu-kupu ‘Rajah Brooke’s Birdwing’, Trogonoptera brookiana merupakan ikon eko-
pelancongan yang terkenal. Namun, kehilangan habitat dan pengutipan berleluasa telah
mengancam kewujudannya yang menggesakan kajian terperinci dijalankan. Kajian ini
merangkumi aspek: taksonomi, kelakuan dan ekologi untuk subspesies albescens dan
mollumar di Semenanjung Malaysia, demi menampung usaha pemuliharaan. Populasi
kupu-kupu ‘birdwing’ yang dijumpai di bahagian timur Semenanjung Malaysia telah
dinamakan subspesies mollumar D’Abrera tetapi statusnya kerap menjadi pendebatan
yang mengakibatkan namanya disinonimkan dengan subspesies Sumatra iaitu trogon
Vollenhoven. Hasil kajian ini dapat menjelaskan kesahihan perbezaan ketara ciri
taksonomi di antara dua populasi tersebut dan seterusnya menyokong pengembalian
status subspesies mollumar. Keputusan penelitian taburan subspesies albescens dan
mollumar selepas tahun 2000 dan sebelum tahun 2000 mempamerkan corak populasi
yang berkurangan. Bagi tujuan meningkatkan kefahaman dan usaha pemuliharaan kupu-
kupu ‘birdwing’ ini, kajian kelakuan telah dijalankan. Subspesies albescens telah dikaji
di dua lokasi terpilih iaitu Kampung Ulu Geroh dan Hutan Simpan Kenaboi.
Pengumpulan data terperinci telah dijalankan tentang ritma aktiviti harian untuk aksi
terbang dan aksi kelompokan (‘puddling’) berserta hubungkait pembolehubah faktor Universitypersekitaran. Interaksi di antara jantan dan of betina kupu-kupuMalaya ‘birdwing’ ini juga dikaji secara ad libitum apabila kerja lapangan dijalankan. Kupu-kupu ‘birdwing’ lebih suka
bunga Bauhinia berbanding dengan Hibiscus, tetapi lebih menarik perhatian adalah
terdapat perbezaan dalam ritma pengambilan nektar di antara kedua-dua jantina yang
boleh dikaitkan dengan persaingan pemakanan dan kelakuan pemakanan sosial. Semua
interaksi di antara kupu-kupu jantan dan betina ‘birdwing’ ini dimulakan oleh jantan
v dan persaingan untuk mendapatkan pasangan merupakan fenomena yang biasa. Strategi
pencarian pasangan oleh kupu-kupu jantan ‘birdwing’ ini berbeza mengikut lokasi.
Sesetengah kelakuan penerimaan dan penolakan pasangan oleh kupu-kupu betina
‘birdwing’ ini telah didokumentasikan. Satu kaedah pemantauan terbaru menggunakan
fotografi telah dibangunkan untuk kupu-kupu jantan subspesies albescens yang
‘puddling’ di Kampung Ulu Geroh. Kaedah tersebut telah diuji selama 24 bulan. Ini
merupakan kali pertama aktiviti ‘puddling’ digunakan sebagai indeks populasi. Populasi
kupu-kupu ‘birdwing’ meningkat dan menurun dengan ketara sepanjang tempoh
pemantauan, ia berpuncak pada bulan Mei dalam kedua-dua tahun dan bulan Ogos
dalam tahun kedua. Populasi kupu-kupu ‘birdwing’ ini berkaitrapat dengan suhu purata
bulanan. Tambahan juga, dua kaedah iaitu transek berjalan kaki dan pemerhatian pada
satu tempat telah digunakan untuk memantau kupu-kupu ‘birdwing’ yang berterbangan
di Hutan Simpan Kenaboi selama 29 bulan. Hasil kajian ini akan membantu dalam
penggubalan garispanduan pemuliharaan untuk spesies kupu-kupu ‘birdwing’
berkarisma ini yang berpotensi menjadi ikon eko-pelancongan di rantau ini.
Kata kunci: Taksonomi, kelakuan, ekologi, Trogonoptera brookiana, Semenanjung
Malaysia.
University of Malaya
vi ACKNOWLEDGEMENTS
I am greatly indebted to my supervisor for her constant encouragement and guidance
throughout my graduate study. Special thanks to my superior in FRIM, Dr. Laurence
Kirton who was my co-supervisor before he retired but continued to guide me in my
research, writing and advise me on statistical analysis. My gratitude is also extended to
Prof. Dato' Dr. Mohd Sofian Azirun, Prof. Rosli Hashim, Assoc. Prof. Rosli Ramli and
Prof. Hiroyuki Takaoka for their insightful comments of my work. I gratefully
acknowledge Veronica Khoo for her friendship and company. I would like to express
my warmest appreciation to all listed here for their helps and contributions in making
this work successful: FRIM’s Entomology Branch–Nafaruding, Fadir, Shaiful, Faizal,
Noor Azmy, Shaina & Adi Purba; Butterfly enthusiasts–Chong CY, Goh LC, Khew SK,
Liew NL, Raymond Lim, Adam Cotton, Cheng KW, Chin FS & Kuennie Lee; BMNH–
Huertas B, Martin G, Giusti A & Zilli A; Naturalis– De Vos R and Gassó Miracle ME;
MZUM–Sasekumar A, Thary Gazy & Amni Bazilah Sulaiman;CIS–Hazmi IR &
Aruchunnan, G; LKCNHM–Hwang WS, Lua HK & Peter Ng; LIPI–Peggie D; MCZ–
Pierce N, Eastwood R & Bittleston L; FRIM’s botanists–Yao TL, Kamarudin Saleh &
Khairuddin Kamaruddin; FRIM’s conservationists–Wendy Yong & Hamidah Mamat;
Ulu Geroh villagers– Tok Batin Ngah Sidin, the late Ahha, Bah Insan, Sani, Ah Ngah,
Kak Rohani & Kak Sharifah; FRIM’s sociologists–Lim HF, Intan, Sharkilah & Dadu;
MNS–Sonny Wong; FRIM’s Recreational Unit–Noor Azlin Yahya & Azahari Mohd. UniversityYusoff. This project was funded by MFRDB of (GPP-0609-FA-02), Malaya EPU, NRE and IPPP (PV131/2012A). We appreciate for the approvals from Head Quarters of Forestry
Department and Forestry Department of each state to conduct our research in forest
reserves. Last but not least, I would like to express my appreciation and love to my
family for their tremendous support throughout my graduate education.
vii TABLE OF CONTENTS
ABSTRACT ...... iii
ABSTRAK ...... v
ACKNOWLEDGEMENTS ...... vii
TABLE OF CONTENTS ...... viii
LIST OF FIGURES ...... xiii
LIST OF TABLES ...... xvii
LIST OF SYMBOLS AND ABBREVIATIONS ...... xix
LIST OF APPENDICES ...... xx
CHAPTER 1: GENERAL INTRODUCTION...... 1
CHAPTER 2: LITERATURE REVIEW ...... 7
2.1 Genus Trogonoptera Rippon, [1898] ...... 7
2.2 Issues between Subspecies T. b. trogon and T. b. mollumar ...... 9
2.3 Life History of the Trogonoptera brookiana ...... 10
2.3.1 Trogonoptera brookiana brookiana ...... 10 2.3.2 Trogonoptera brookiana albescens ...... 11 2.3.3 Trogonoptera brookiana trogon ...... 15 2.3.4 Sex Ratio: Field Sightings ...... 17
2.4 The Host Plant and Aristolochic Acid ...... 17
2.5 Habitat...... 19
2.6 Puddling Behaviour ...... 19
University2.7 Flying Behaviour ...... of Malaya 24
2.8 Nectaring Behaviour ...... 26
2.9 Courting and Mating Behaviour ...... 30
2.10 Monitoring of Butterflies ...... 35
2.11 Capture-recapture Method ...... 36
viii 2.12 Transect Walk Method or Pollard Walk...... 38
2.13 Other Methods ...... 39
2.14 Criterion to be Considered in Choosing Monitoring Method...... 41
2.15 Seasonality ...... 42
2.16 Threat and Protection in Malaysia ...... 44
CHAPTER 3: TAXONOMY STATUS OF TROGONOPTERA BROOKIANA MOLLUMAR ...... 46
3.1 Introduction ...... 46
3.2 Literature Review ...... 47
3.3 Methodology ...... 51
3.3.1 Collections Visited and Specimens Examined ...... 51 3.3.2 Description of Characters Examined ...... 52 3.3.3 Analyses...... 55
3.4 Results...... 55
3.4.1 Female...... 55 3.4.2 Male...... 59 3.4.3 Determining Subspecies and Posterior Probabilities ...... 62 3.4.4 Diagnoses of Subspecies ...... 63
3.5 Discussion ...... 71
3.6 Conclusion ...... 77
CHAPTER 4: BIOGEOGRAPHICAL DISTRIBUTION OF THE TROGONOPTERA BROOKIANA IN PENINSULAR MALAYSIA ...... 78
University4.1 Introduction ...... of Malaya 78
4.2 Literature Review ...... 78
4.3 Methodology ...... 80
4.3.1 Collection Records ...... 80 4.3.2 Literature Surveys and Interviews ...... 81 4.3.3 Active Site Surveys ...... 82
ix 4.3.4 Georeferencing ...... 84 4.3.5 Mapping...... 84
4.4 Results...... 85
4.4.1 Collection Records, Literature Surveys and Interviews, and Active Site Surveys...... 85 4.4.2 Comparison between Historical and Present Distributions ...... 90
4.5 Discussion ...... 92
4.6 Conclusion ...... 97
CHAPTER 5: BEHAVIOURAL STUDIES OF THE TROGONOPTERA BROOKIANA ALBESCENS ...... 99
5.1 Introduction ...... 99
5.2 Literature Review ...... 99
5.3 Methodology ...... 101
5.3.1 Puddling Rhythm ...... 101 5.3.2 Flight Rhythm ...... 102 5.3.2.1 Ulu Geroh Village ...... 102 5.3.2.2 Kenaboi Forest Reserve ...... 103 5.3.3 Nectaring Rhythm ...... 104 5.3.3.1 Ulu Geroh Village ...... 104 5.3.3.2 Kenaboi Forest Reserve ...... 106 5.3.4 Preferred Nectar Plant Studies in Ulu Geroh Village and Kenaboi Forest Reserve...... 107 5.3.5 Courting and Mating Incidences ...... 108
5.4 Results...... 109 University5.4.1 Puddling and Flight Rhythms ...... of Malaya 109 5.4.1.1 Ulu Geroh Village ...... 109 5.4.1.2 Kenaboi Forest reserve ...... 114 5.4.2 Nectaring Rhythms ...... 117 5.4.2.1 Ulu Geroh Village ...... 117 5.4.2.2 Kenaboi Forest Reserve ...... 124 5.4.3 Preferred Nectar Plants ...... 130 5.4.3.1 Ulu Geroh Village ...... 130
x 5.4.3.2 Kenaboi Forest Reserve ...... 134 5.4.4 Courting and Mating Incidences ...... 136
5.5 Discussion ...... 141
5.5.1 Puddling and Flight Rhythms ...... 141 5.5.2 Nectaring Behaviour ...... 145 5.5.3 Courting and Mating Behaviour ...... 149
5.6 Conclusion ...... 152
CHAPTER 6: DEVELOPMENT OF A NEW MONITORING METHOD AND MONITORING PROGRAMME USING THREE METHODS FOR THE TROGONOPTERA BROOKIANA ALBESCENS AND ITS RELATIONSHIP WITH ENVIRONMENTAL PARAMETERS ...... 153
6.1 Introduction ...... 153
6.2 Literature Review ...... 154
6.3 Methodology ...... 156
6.3.1 Development of a New Monitoring Method based on Counts of Puddling Trogonoptera brookiana albescens ...... 156 6.3.1.1 Study Site and Camera Set Up ...... 156 6.3.1.2 Comparison of Transect Counts and Puddling Counts ...... 158 6.3.1.3 Sample Requirements for Puddling Counts ...... 158 6.3.1.4 Application of the Technique ...... 160 6.3.1.5 Relationship between the Birdwings Population and the Monthly Environmental Parameters ...... 161 6.3.2 Monitoring using other Methods ...... 161 6.3.2.1 Study Site and Study Period ...... 162 6.3.2.2 Simplified Transect Walk ...... 162 6.3.2.3 Single-point Observation ...... 162 6.3.2.4 Relationship between the Birdwings Population and Environmental UniversityParameters ...... of Malaya 163
6.4 Results...... 163
6.4.1 Development of a New Monitoring Method based on Counts of Puddling Trogonoptera brookiana albescens ...... 163 6.4.1.1 Comparison of Transect Counts and Puddling Counts ...... 164 6.4.1.2 Sample Requirements for Puddling Counts ...... 165 6.4.1.3 Application of the Technique ...... 167
xi 6.4.1.4 Relationship between the Birdwings Population and the Monthly Environmental Parameters ...... 170 6.4.2 Monitoring using Simplified Transect Walk and Single-point Observation and the Relationship between the Birdwings Population and the Environmental Parameters ...... 172 6.4.2.1 Simplified Transect Walk ...... 172 6.4.2.2 Single-point Observation ...... 172 6.4.2.3 Relationship between the Birdwings Population and the Environmental Parameters ...... 174
6.5 Discussion ...... 176
6.5.1 New Monitoring Method ...... 176 6.5.1.1 Comparison of Transect Counts and Puddling Counts ...... 176 6.5.1.2 Sample Requirements for Puddling Counts ...... 177 6.5.1.3 Application of the Technique: Two-year Monitoring ...... 178 6.5.1.4 Application of the Technique: Developing Monitoring Programme 179 6.5.1.5 Adapting the Method for Other Subspecies and Species ...... 180 6.5.2 Other Monitoring Methods ...... 182 6.5.3 Seasonal Fluctuation of the Trogonoptera brookiana albescens ...... 184
6.6 Conclusion ...... 186
CHAPTER 7: GENERAL DISCUSSION ...... 188
7.1 Synthesis ...... 188
7.2 Aspects Requiring Futher Research ...... 190
7.2.1 Genetics and Molecular...... 190 7.2.2 Distributions and Monitoring ...... 190 7.2.3 Resources, Habitat and Behaviour ...... 191 7.2.4 Other Subspecies...... 192 University7.3 Conservation ...... of Malaya 193 CHAPTER 8: CONCLUSION ...... 194
REFERENCES ...... 196
LIST OF PUBLICATIONS AND PAPERS PRESENTED ...... 218
xii LIST OF FIGURES
Figure 1.1: The specimens of Trogonoptera brookiana...... 2
Figure 1.2: The puddling males of the Trogonoptera brookiana albescens...... 3
Figure 2.1: Caterpillars of younger (above) and matured stage of Trogonoptera brookiana brookiana...... 11
Figure 2.2: Life cycle of Trogonoptera brookiana albescens and its host plant...... 14
Figure 3.1: Genitalia illustration of Trogonoptera brookiana mollumar and T. b. trogon scanned from original description (D’Abrera, et al., 1976)...... 49
Figure 3.2: Illustrations of measurements and character categories in females...... 54
Figure 3.3: Illustrations of measurements and character categories in males...... 54
Figure 3.4: Two-dimensional plot of correspondence analysis components 1 and 2 for thirty two females...... 59
Figure 3.5: Two-dimensional plot of correspondence analysis components 1 and 2 for two hundred two males...... 62
Figure 3.6: Male of subspecies mollumar, Holotype deposited at BMNH, BMNH(E)1420520...... 66
Figure 3.7: Female of subspecies mollumar, Allotype deposited at BMNH, BMNH(E)1420526...... 67
Figure 3.8: Male of subspecies trogon, deposited at BMNH, BMNH(E)1420534. .... 70
Figure 3.9: Female of subspecies trogon, deposited at BMNH, BMNH(E)1420526...... 71
Figure 3.10: The Type specimens of Trogonoptera brookiana trogon ♂ form walshi...... 75
Figure 3.11: The Type specimens of Trogonoptera brookiana trogon ♂ form Universitywalshoides...... of Malaya 76 Figure 3.12: The normal scales and curled scales on the type specimens of Trogonoptera brookiana trogon ♂ form walshoides...... 76
Figure 4.1: A map illustrating the mountain ranges in Peninsular Malaysia...... 80
Figure 4.2: Distribution records of Trogonoptera brookiana albescens and T. b. mollumar compiled from (A) collections, (B) literature surveys and interviews and (C) active site surveys...... 88
xiii Figure 4.3: (A) Historical and (B) Present Distributions of the Trogonoptera brookiana albescens and T.b. mollumar...... 92
Figure 5.1: Example of the series of floral clumps scored for the studies on birdwing floral visitation behaviour...... 106
Figure 5.2: Daily rhythms of puddling activities at Puddle 1, Puddle 2 and the total from both puddles at Ulu Geroh village...... 111
Figure 5.3: Daily flight rhythms of the male, female and the total of birdwings at Ulu Geroh village...... 112
Figure 5.4: Daily puddling and flight rhythms and the relation with temperature and relative humidity at Ulu Geroh village...... 113
Figure 5.5: Daily flight rhythms from single spot observation and transect walk methods at Kenaboi Forest Reserve and the relationship with monthly averages of temperature, relative humidity and light intensity...... 116
Figure 5.6: The number of male birdwings nectaring on each plant species and the average of all plant species at Ulu Geroh village...... 119
Figure 5.7: The number of female birdwings nectaring on each plant species and the average of all plant species at Ulu Geroh village...... 120
Figure 5.8: Total time spent by a male birdwing at each hour on a plant of each plant species and the average of all plant species at Ulu Geroh village. 121
Figure 5.9: Total time spent by a female birdwing at each hour on a plant of each plant species and the average of all plant species at Ulu Geroh village. 122
Figure 5.10: Total flowers or flower clumps visited by a male birdwing at each hour on a plant of each plant species and the average of all plant species at Ulu Geroh village...... 123
Figure 5.11: Total flowers or flower clumps visited by a female birdwing at each hour on a plant of each plant species and the average of all plant Universityspecies at Ulu Geroh village. of...... Malaya 124 Figure 5.12: The number of male birdwings nectaring on each plant species and the average of all plant species at Kenaboi Forest Reserve...... 125
Figure 5.13: The number of female birdwings nectaring on each plant species and the average of all plant species at Kenaboi Forest Reserve...... 126
xiv Figure 5.14: Total time spent by a male birdwing at each hour on a plant of each plant species and the average of all plant species at Kenaboi Forest Reserve...... 127
Figure 5.15: Total time spent by a female birdwing at each hour on a plant of each plant species and the average of all plant species at Kenaboi Forest Reserve...... 128
Figure 5.16: Total flowers or flower clumps visited by a male birdwing at each hour on a plant of each plant species and the average of all plant species at Kenaboi Forest Reserve...... 129
Figure 5.17: Total flowers or flower clumps visited by a female birdwing at each hour on a plant of each plant species and the average of all plant species at Kenaboi Forest Reserve...... 130
Figure 5.18: Monthly averages of the number of flowers or flower clumps, percentage of birdwing nectaring on each flower or flower clump, total flowers visited and total time spent by a birdwing on each plant in second in Ulu Geroh Village...... 132
Figure 5.19: The total number of flowers or flower clumps on each plant, percentage of birdwing nectaring on each flower or each flower clump, average of total flowers visited by a birdwing on each plant and average of total time spent by a birdwing on each plant in Kenaboi Forest Reserve...... 135
Figure 5.20: The total number of flower clumps on each plant, percentage of birdwing nectaring on each flower clump, average of total flowers visited by a birdwing on each plant and average of total time spent by a birdwing on each plant in Kenaboi Forest Reserve...... 137
Figure 6.1: Camera setup to capture images on larger puddling group...... 157
Figure 6.2: Camera setup to capture images on smaller puddling group...... 157
Figure 6.3: Puddling Trogonoptera brookiana albescens in study site...... 164
UniversityFigure 6.4: Correlation of the average numbers of of birdwingsMalaya in flight and puddling in six different months...... 165
Figure 6.5: Relationship between the monthly average numbers of birdwings puddling at 1600 hours and 0900–1600 hours in six different months. . 166
Figure 6.6: Correlation between counts at 1100 hrs and 1600 hrs...... 167
xv Figure 6.7: The 24-month birdwing count fluctuation pattern (photography method) and weather fluctuation from weather station data...... 169
Figure 6.8: The 29-month birdwings population fluctuation using simplified transect walk and single-point observation methods...... 173
University of Malaya
xvi LIST OF TABLES
Table 2.1: The list of previously described, Trogonoptera brookiana subspecies with their general distributions...... 8
Table 3.1: Character descriptions. Scores given for each categorical data are given in parentheses. Numerators and denominators for measured characters are defined in Figures 3.2 and 3.3...... 53
Table 3.2: Measurements and ratios for Sumatran and Peninsular Malaysian females ...... 56
Table 3.3: Numbers and percentages (in parentheses) of females with each character from fourteen Sumatran and eighteen Peninsular Malaysia females...... 58
Table 3.4: Analysis of indicator matrix using single correspondence analysis of thirty two females...... 58
Table 3.5: Measurements and ratios for Sumatran and Peninsular Malaysian males.... 60
Table 3.6: Numbers and percentages (in parentheses) of males with each character from one hundred eighty Sumatran and twenty two Peninsular Malaysia males...... 61
Table 3.7: Analysis of indicator matrix using single correspondence analysis of two hundred two males...... 61
Table 3.8: The posterior probability calculation for specimen FM1...... 63
Table 5.1: Averages temperature and pH readings at Ulu Geroh village, measured on 25 February 2010 at 0900, 1200 and 1500 hrs...... 111
Table 5.2: Correlation matrix of puddling and flight rhythms with relative humidity and temperature at Ulu Geroh village ...... 114
Table 5.3: Correlation matrix of flight rhythms recorded from transect walk and single spot observation methods at Kenaboi Forest Reserve with Universitytemperature, relative humidity andof light intensityMalaya ...... 117 Table 5.4: The number of flowers or flower clumps on each plant in each month in Ulu Geroh Village...... 131
Table 5.5: Comparisons between three flowering plant species on the percentage of birdwing nectaring on each flower or flower clump, total flowers visited and total time spent by a birdwing using Kruskal-Wallis analysis in Ulu Geroh Village...... 133
xvii Table 5.6: Comparisons between three flowering plant species on the percentage of birdwing nectaring on each flower or flower clump, total flowers visited and total time spent by a birdwing using Kruskal-Wallis analysis in Kenaboi Forest Reserve...... 138
Table 5.7: Interactions between females and males in Ulu Geroh Village and Kenaboi Forest Reserve...... 139
Table 6.1: Comparison of the number of consecutive images needed to obtain a reliable count of puddling birdwings, showing one-, three- and five- image averages of the counts and their 95% confidence intervals for two different times of the day...... 166
Table 6.2: Correlation matrix for the first, second and third day of the monthly birdwing counts at 1600 hrs ...... 167
Table 6.3: General linear model for counts of puddling birdwing in response to monitoring month, relative humidity and brightness after removing non- significant terms (photography method)...... 169
Table 6.4: Linear regression model of all month physical parameters collected from weather station and monthly count of Trogonoptera brookiana albescens (photography method)...... 171
Table 6.5: Linear regression model of mean temperature collected from weather station and monthly count of Trogonoptera brookiana albescens (photography method)...... 171
Table 6.6: Linear regression model of all the physical parameters and the number of flowering plants recorded and birdwings count (simplified transect walk method)...... 175
Table 6.7: Linear regression model of all the physical parameters and the number of flowering plants recorded and birdwings count (single-point observation method)...... 176
University of Malaya
xviii LIST OF SYMBOLS AND ABBREVIATIONS
BMNH : Natural History Museum London
CIS : The Centre for Insect Systematics
EPU : Malaysian Economic Planning Unit
FR : Forest Reserve
FRIM : Entomological Reference Collection of Forest Research Institute
Malaysia
GPS : Global Positioning System
IPPP : Postgraduate Research Grant of University of Malaya
LKCNHM : Lee Kong Chian Natural History Museum
LIPI : Indonesian Institute of Sciences Indonesia
MCZ : Museum of Comparative Zoology of Harvard University
MFRDB : Malaysian Forest Research and Development Board
MNM : Malaysia National Museum
MNS : Malaysian Nature Society
MRR : Mark-Release-Recapture
MZUM : Museum of Zoology, University of Malaya
Naturalis : Naturalis Biodiversity Center
NRE : Ministry of Natural Resources and Environment
RMNH : Dutch Rijksmuseum van Natuurlijke Historie UniversityUKM : Universiti Kebangsaan Malaysiaof Malaya ZMA : Zoological Museum Amsterdam
xix LIST OF APPENDICES
Appendix A: Part of correspondence analysis results for females. List of specimens grouped with identical coordinates ...... 220
Appendix B: Part of correspondence analysis results for males. List of specimens grouped with identical coordinates ...... 221
Appendix C:The male specimens examined in the study. Specimens without museum’s codes were given the localities and the dates of collections ...... 222
Appendix D:The female specimens examined in the study. Specimens without museum’s codes were given the localities and the dates of collections ...... 224
Appendix E: List of literature surveys in distribution study (Chapter 4) ...... 225
Appendix F: Daily average of the total number of birdwings nectaring on each plant in each month in Ulu Geroh Village. Asterisk indicates the plant was not flowering in that specific month ...... 227
University of Malaya
xx
CHAPTER 1: GENERAL INTRODUCTION
Trogonoptera brookiana (Wallace, 1855) under family Papilionidae is named as the
Rajah Brooke's Birdwing. The species is distributed in Malaysia, Sumatra and Borneo
(Braby et al., 2005). There are three subspecies that occur in Malaysia that is, brookiana
(Wallace, 1855), albescens Rothschild, 1895 and mollumar D’Abrera, 1976. In
Malaysia, subspecies brookiana can be found in Sabah and Sarawak while albescens
and mollumar is confined in Peninsular Malaysia.
The species is a well-known ecotourism icon of Malaysia (Phon & Kirton, 2009). It
is a strikingly large-sized butterfly with metallic green markings and a touch of
iridescent azure blue set against black wing (Figure 1.1). Its beautiful features and
puddling behaviour have attracted people to this lovely butterfly. However, only
albescens and brookiana males often puddle (Figure 1.2). They aggregate at moist
places along forest paths and river banks to drink water from which nutrients are
obtained.
This forest dwelling butterfly has been said to be declining in numbers, but the exact
status of the populations is unknown. The two major factors that are believed to place
great pressure on populations are habitat loss and an extraordinarily high demand for
this butterfly by collectors and commercial dealers (Kirton, 1991). For instance, the
habitat of subspecies mollumar is rapidly depleting in the lowlands of Johor (Phon & UniversityKirton, 2009). Destruction of the puddling of habitat Malaya for the birdwing had further threatened the survival of the populations (Phon et al., 2011; Chan, 2010; Chiew, 2010).
Commercial collections have been mentioned in Cameron Highlands (Batten & Batten,
1970a) and in Ulu Langat (Larsen, 2000). Some premises in Batu Caves were turned
into butterfly and other arthropods factory in 1982 (Malayan Nature Society, 1982). In
Malaysia, 37,259 of Trogonoptera brookiana was under trade and mainly collected
1
from the wild since 1985 to 2005 (UNEP-WCMC, 2007). The total reported export
trade of wild specimens between 2001 and 2010 was 5,060 individuals (UNEP-WCMC,
2012).
University of Malaya
Figure 1.1: The specimens of Trogonoptera brookiana. Top: subspecies albescens (male); Middle: subspecies albescens (female); Bottom: subspecies mollumar (female). Scale bar is equivalent to 2 cm.
2
Figure 1.2: The puddling males of the Trogonoptera brookiana albescens.
Conservation of this birdwing is needed especially it is an additional ecotourism
attraction that contributes to the country’s revenue. But, much scientific information on
the birdwing is still very much lacking although it is considerably a well known and
popular butterfly. Information available so far is the taxonomy, general distribution,
their host plant, life history of subspecies albescens, common flowering plants visited
and general description on puddling behaviour. There are many gaps to be filled, which
is the intention of this current work, in order to conserve this birdwing effectively.
The name of mollumar (sometimes considered synonym of trogon) is not always
acceptable by taxonomists and was discussed between Haugum and Low (1982) and
D’Abrera (2003). The uncertainty of the taxonomic status had complicate the
conservation of this subspecies in Malaysia. Although the general distribution of the
Trogonoptera brookiana is always being mentioned in major butterfly books (e.g.
Corbet & Pendlebury, 1992; Kirton, 2014; D’Abrera, 2003; Igarashi & Fukuda, 1997),
the distribution has never been mapped. The resolution of the maps produced by
Haugum and Low (1982) were too small and can be misleading. Moreover, the
Universitydistributional range and changes throughout of the yearsMalaya that has occurred are still
unknown. Information on the updated version of their distribution is needed, making
this part of the objectives of this study. The behavioural information is very scarce
except for the puddling behaviour. The rhythmics of nectaring, flying and even the well
known puddling are understudied. The preferred nectar plants are unidentified.
3
Monitoring of the population changes and the effect of environmental parameters on the
population has not been studied before.
Therefore, the rationale and the purpose of conducting the study are to gather
information for an effective conservation programme and to utilise the gathered
information to improve ecotourism sector of the birdwing. However, not all the
subspecies in Malaysia will be studied. The targeted subspecies are albescens and
mollumar. The objectives of this study are:
(1) to resolve the taxonomy of subspecies mollumar;
(2) to document the distributional changes and the range limits for the subspecies
albescens and mollumar;
(3) to conduct the behavioural studies for the subspecies albescens, on aspects of
flight, nectaring plus puddling rhythms, preference of adult food exploitation and
reproductive behaviour; and
(4) to monitor albescens population using a newly developed method for puddling
birdwings and two other established methods at selected sites and to relate the
monitoring data with environmental parameters.
All studies are divided into four chapters that is the taxonomy of subspecies
mollumar (in Chapter 3), distribution of albescens and mollumar (in Chapter 4),
behaviour and monitoring of subspecies albescens (in Chapter 5 and 6, respectively). UniversityEach study chapter contains introduction, of a brief Malaya literature review, materials and methods, results, discussion and conclusion. Extensive literature review can be found in
Chapter 2. A general discussion links all the studies’ discussion and also includes future
studies, this is in Chapter 7. Chapter 8 is the overall conclusion. The conclusion in each
study chapter is more comprehensive compared to the overall conclusion.
4
The delimitation of the study is in the choice of focussed subspecies, in which two
subspecies, albescens and mollumar were selected to study their distribution while only
albescens was selected for behavioural and monitoring studies. Both of the subspecies
can only be found in Peninsular Malaysia and therefore it is extremely important to
conduct in depth studies on these major aspects that will contribute towards the
conservation programme for these iconic subspecies in Malaysia. Moreover, the
population of albescens was sufficiently large that enabled the behavioural and
monitoring studies to be completed within the research duration. However, an initial
plan to include the life history studies of these subspecies had to be abandoned since it
became an impossible task to collect the immature stages of these birdwings although
concerted efforts were made to search for them.
A major limitation emerged during the data collection for monitoring and
behavioural studies that hampered the study progress, where a change in study site was
inevitable. A long term monitoring and behavioural studies that were initially planned to
be conducted in Ulu Geroh village which has the largest puddling birdwings population
and moreover, the females were encountered frequently in the village. The site was also
suitable to study nectaring behaviour because of the presence of flowering preferred
food plants for these species of birdwings in the forest nearby. However, sadly after two
years of intensive data collection, some prominent village residence protested to the
project being conducted there. Although we tried to counter these objections, in the Universityform of persuasions and holding several of discussion Malaya sessions with the leader of the village and villagers but still failed to get a positive response from them. The leader
advised us to leave the village although he himself was very supportive on the studies.
Therefore, a new site has to be chosen to continue the studies in order to achieve the
objectives. The data collected in Ulu Geroh village was included in the thesis but it is
not comparable with the data collected in the new study site, namely Kenaboi Forest
5
Reserve. Unsustainable funding resources, logistic constraints, unpredictable weather
and flowering season also contributed to the delay in research completion. The
difficulties faced during the research had to be overcome, in which I took them as
challenges that were successfully resolved with the completion of this current project.
University of Malaya
6
CHAPTER 2: LITERATURE REVIEW
2.1 Genus Trogonoptera Rippon, [1898]
The genus of Trogonoptera Rippon, [1898] was first used around 1898 (Rippon,
1898) and was described in detail in 1906 (Rippon, 1906). The status of Trogonoptera
as a standalone genus or a subgenus under Troides Hübner, [1819] is always being
debated in the taxonomic world. In the recent literatures, it was treated as a subgenus
under Troides Hübner, [1819] based on adult features (Miller, 1987) and immature
stages (Parsons, 1996). Late John Eliot in Corbet and Pendlebury (1992) followed the
nomenclature as suggested by Miller (1987). However, it was re-stated as a distinct
genus because of the presence of Morpho (flute)-type scales (Eliot, 2006; Tilley &
Eliot, 2002) unique to this genus and absent in other Troides butterflies. In 2003, Kondo
et al. used sequences of the mitochondrial gene ND5 to analyse molecular systematic on
three genera namely Trogonoptera, Troides and Ornithoptera. They found that an
ancestral species gave rise to Trogonoptera and the ancestor of Troides plus
Ornithoptera. They were all monophyletic (Kondo et al., 2003). A more recent
phylogenetic analysis of the ND5 sequences suggested that Trogonoptera and Troides
do not form a close sister relationship (Braby et al., 2005). A similar result was found
by Condamine et al. (2015). Male’s juxta was found to be different from Troides as well
as Ornithoptera (Hancock, 1991). The genus was reported to be endemic to a limited
area in the Oriental region, from Malaysia, Sumatra, Borneo to Palawan (Braby et al.,
2005). Haugum and Low (1982) and Ek-Amnuay (2012) listed Mawlamyine in Burma Universityand Yala in Thailand as the occurrence locations of of theMalaya genus, respectively.
There are two species of Trogonoptera, namely brookiana (Wallace, 1855) and
trojana (Staudinger, 1889). Species trojana occurs in Palawan and Philippine
Archipelago (Rippon, 1902). The first description of brookiana was based on a male
collected from Rejang River, Sarawak by Wallace (1855) while in the same year, a
7
second description on the male was done by Hewitson based on his collection which
was deposited in the British Museum (Rippon, 1906). About 26 years later, the female
was described in 1881 based on a female collected from Perak in Peninsular Malaysia
by Philip H. Gosse from his collection (Gosse, 1881).
Under Trogonoptera brookiana, there are a few subspecies described and listed
below (Table 2.1) with their general distribution patterns:
Table 2.1: The list of previously described, Trogonoptera brookiana subspecies with their general distributions.
Subspecies General distribution
brookiana (Wallace, 1855) Borneo, Balabac (Haugum & Low, 1982) and Pulau Labuan (Haugum & Low, 1982; Kirby, 1896).
albescens Rothschild, 1895 Perak, Pahang, Selangor (Corbet & Pendlebury, 1992), Negeri Sembilan (De Worms, 1973), Mawlamyine (Haugum & Low, 1982) and Yala (Ek-Amnuay, 2012).
trogon (Vollenhoeven, 1860) Sumatera (Haugum & Low, 1982).
mollumar D’Abrera, 1976 Southeastern part of Johor and Pahang and eastern (synonym of trogon) part of Terengganu (D’Abrera et al., 1976).
natunensis Rothschild, 1908 Natuna Islands (Haugum & Low, 1982; D’Abrera, 2003).
cardinaali Haugum and Low, Singkep Island (Haugum & Low, 1982); Lingga 1982 Islands and Riau Islands (D’Abrera, 2003).
jikoi Kobayashi, 1986 Tuangku Island (D’Abrera, 2003).
haugumei Parrott, 1991 Southern Borneo (D’Abrera, 2003). Universitytoshikii Kobayashi, 1991 Siberut of Island (D’Abrera,Malaya 2003). mariae Neukirchen, 1993 Batu Islands (D’Abrera, 2003).
8
2.2 Issues between Subspecies T. b. trogon and T. b. mollumar
Subspecies mollumar occurs in south-eastern part of Johor to Pahang and eastern part
of Terengganu (D’Abrera et al., 1976) while subspecies trogon occurs in Sumatera
(Haugum & Low, 1982). The population for mollumar was first discovered by the late
John Eliot and Charles Cowan around 1937–1938 in the swampy forest of south-eastern
Johor (Corbet & Pendlebury, 1956; Eliot, 1972). However, the sighting of a population
in Johor was probably earlier than that based to Distant (1882–1886) who included
Johor as one of the habitats for species brookiana. Corbet and Pendlebury (1956)
regarded this population as subspecies trogon which was subsequently used by Fleming
(1975).
D’Abrera (1975) suspected some degree of differences between Sumatran trogon and
Peninsular Malaysia trogon which he later separated and described mollumar from
trogon (D’Abrera et al., 1976). According to D’Abrera et al. (1976) mollumar warranted
a subspecies status because of the significant differences on the male genitalia and the
external adult morphology. Unfortunately, the number of specimens and details of
specimens examined were not mentioned in the original description except for the
designated holotype and allotype. The holotype and allotypre are kept in Natural
History Museum, London (BMNH). The holotype labelled as “Ulu Sedili, Johore, W.
Malaysia (V. Doggett), 17th February, 1974”. While the allotype labelled as “Ulu Sedili,
Johore, W. Malaysia (N. Parker), 29th August, 1971”. However, Haugum and Low University(1982) treated mollumar as a locality form of (f. loc.) Malaya of trogon because they think this eastern population of the Peninsula is not adequately distinct from the Sumatran
population to warrant a subspecies status. They mentioned that the number of specimens
examined by D’Abrera in D’Abrera et al. (1976) perhaps was not enough because in
their own series of specimens, they found overlapping characters between subspecies
trogon and mollumar. Haugum and Low (1982) also argued that the male’s genitalia of
9
both subspecies prepared in the original description were not properly compared. There
was a refutation against Haugum and Low (1982) which was reported in D’Abrera
(2003). The late John Eliot in Corbet and Pendlebury (1992) synonymised mollumar
under trogon probably followed Haugum and Low (1982).
2.3 Life History of the Trogonoptera brookiana
The life histories of subspecies brookiana, albescens and trogon were recorded. The
life history of subspecies brookiana was briefly recorded in a magazine by Chey (1997).
A thorough life history of subspecies albescens was studied by Goh (1994). All five
instars caterpillar, egg, pupa and adult male and female were photographed and a short
description was given in Igarashi and Fukuda (1997). Straatman and Nieuwenhuis
(1961) studied the morphology and habits of the immature stages while Dahelmi et al.
(2008) studied the duration of immature stages of the subspecies trogon.
2.3.1 Trogonoptera brookiana brookiana
The egg was described as a round pale reddish-grey, which was laid on the foliage of
Aristolochia sp, climbing vine (Chey, 1997). After hatching, the caterpillar fed greedily.
The caterpillar was about 2 inches long, with chocolate brown body. The body had
orange tubercles (Figure 2.1). Both head and thoracic legs were black. The tubercles
colour varied in instar, it became paler as the caterpillar matures (Figure 2.1). After
several weeks, the caterpillar was transformed into apple-green chrysalis which was
about 1.5 inches in length. The chrysalis stood upright and supported by girdles and Universitycremaster onto a twig. The chrysalis with of vein-like Malaya markings camouflaged well with host plant leaf structures. On its back, it had several short spine-like structures (Chey,
1997).
10
Figure 2.1: Caterpillars of younger (above) and matured stage of Trogonoptera brookiana brookiana. The photos were scanned from Chey University(1997). of Malaya
2.3.2 Trogonoptera brookiana albescens
In a study conducted by Goh (1994) on oviposition behaviour of the birdwing, where
two females were collected from Cameron Highlands and observed at hourly intervals
on their oviposition rate on 21–30 August 1992 for the first female and 30 August–5
11
September 1992 for the second female in an enclosure measured 9.0x6.0x2.0 m in
Penang Butterfly Farm at Teluk Bahang, Penang. All the eggs were oviposited singly on
the host plant Aristolochia foveolata and more than 50% of the eggs were deposited
between 1200 and 1800 hours (Goh, 1994).
In the life history study done by the same author (Goh, 1994), another female
collected from Sahom Village near Chenderiang, Perak was observed for oviposition
behaviour. Observation was conducted for more than a week from 26 September to 4
October 1990 on a female which produce a total of 139 eggs. All the eggs collected
each day were kept in plastic containers for life history study. The photo of each stage is
shown in Figure 2.2. The eggs were spherical and about 2.0 to 2.22 mm in diameter.
The freshly laid eggs were yellowish green in colour and on the fifth day, the colour
changed to yellowish orange with black rings on the upper part of the eggs. The
incubation period of the egg was about 6–7 days. The first, second, third and fourth
instars lasted about 5 days, 4 days, 5 days and 7 days respectively. The last instar lasted
about 9 days, therefore the caterpillar duration was about 30 days. On the early morning
of the 31st day, the caterpillar stopped feeding and wandered around to look for a place
to pupate. Most likely, it preferred to pupate on the host plant stem. In the afternoon on
the same day, the caterpillar attached its anal end on the host plant. On the next day, it
curled the body until the morning of the 34th day and began to transform into a
chrysalis. The pupa stage lasted for 22 to 25 days (Goh, 1994).
UniversityThe ground colour of the first instar caterpillar of wasMalaya brown. There were spines at the 2nd and 3rd thoracic segments and the first abdominal segment. The spines were larger,
with each spine having lighter brown at the base and creamy white at the tip. Similar
spines but in chalky white covered the whole abdominal segments 4, 7, 8, and 9. Shorter
spines were found at the abdominal segments 2, 3, 5 and 6, with the same colour as the
12
body. All these spines branching out into tiny black setae. These black setae were
absent in the subsequent instars. Head capsule width was 0.8 mm and body length at
hatching was 6.0–7.0 mm. The newly hatched larva consumed the whole egg shell and
then it moved on to feed on the host plant. It started eating in the central part of the
young leaf by making a hole, and then moved on to feed towards the peripheral leave
edges.
Based to the images in the paper, despite the absence of the black setae, the second
and the third instars were not much different morphologically from the first instar. The
head capsules of the second and third instars were about 1.6 mm and 2.7 mm,
respectively. While the body lengths at moulting were 14.0–15.0 mm and 19.0–20.0
mm, respectively. The second instar caterpillar did not eat the residual first instar skin
which may be due to the presence of hairs on the skin. As for the subsequent instars, the
caterpillar consumed all the moulted skin, except the head capsules. The fourth instar
body measurements were: head-capsule width of 3.8 mm and a body length of 43.4 mm.
In the fifth instar, the tubercles pointed backward instead of vertically. It had a
transverse white band which started at base of the 5th segment and slanted right up to the
pair of tubercles on the 6th segment. The caterpillar fed from the underside of the leaves.
The pupa had apple green body colour and the antennal lines were greenish white.
There were two pairs of sharp-pointed dorsal processes with the pair at the 5th segment
directed towards the midline. The pupa had a small marginal violet spot towards the
Universityapex. of Malaya
Based on Igarashi and Fukuda (1997), the egg was laid singly on the underside of a
host plant leaf. The egg was 1.80 mm and 1.50 mm in diameter and in height,
respectively. The caterpillar rested on the underside of a host plant leaf while the late
instar of caterpillar was also found on the stem. The body length of the pupa was 48
13
mm. The larvae, eggs and the host plants Aristolochia foveolata were found in dark and
dense forest. However, an oviposition was recorded on a host plant planted in a
residential garden along a trunk road (Igarashi & Fukuda, 1997).
University of Malaya
Figure 2.2: Life cycle of Trogonoptera brookiana albescens and its host plant. 3) egg; 4) 1st instar; 5) 2nd instar; 6) 3rd instar; 7 – 8) 4th instar; 9) 5th instar; 10) prepupa; 11) pupa; and 12) hostplant, Aristolochia foveolata. The photos were scanned from Goh (1994).
14
2.3.3 Trogonoptera brookiana trogon
At Gedong Biara Estate situated in north-eastern part of Sumatra, a few hundreds of
adults were reared within 1952–1954 (Straatman & Nieuwenhuis, 1961). They were
reared from a plant species of Aristolochia, identified by the Rijkscherbarium at Leiden.
It shared the host plant with Troides amphrysus ruficollis (Butler, 1879).
The egg described as large, round and pale reddish-grey and the incubation duration
was about 10–12 days (Straatman & Nieuwenhuis, 1961). A large number of eggs were
laid on new shoots of the host plants, some on dry twigs or on other nearby plants
within an area of 20 cm distance apart. The first instar caterpillar was whitish, with
small black dots on the abdominal segments. The body had elongated thin tubercles
which were black and yellow colour. The tubercles had many branched spines on all the
segments. Most of the birdwings went through five caterpillar instars but certain female
caterpillars went through six instars. The intermediate instars between first and fifth
instars had similar appearance. The head black and the ground colour of the body varied
from leaden grey to nearly black. A lighter grey saddle mark present in most specimens.
The saddle mark started on 4th abdominal segment and ran antero-laterally to the 3rd
abdominal segment. There were eight tubercles on thoracic segments while the
abdominal segments had six tubercles on each segment. All the tubercles were fleshy
and without spines. With the exceptions of the dorsal tubercles on the 2nd, 3rd, 5th, and
6th abdominal segments, which were black, all the other tubercles were bright orange. UniversityThe ground colour of the last instar varied of from darkMalaya sooty brown to an ashy grey colour and the saddle mark was somewhat more conspicuous than in the younger
instars. The longer and slender tubercles were held in an almost horizontal position. The
bright orange tubercles of the previous instars were now ochraceous while the shorter
dorsal tubercles of the 2nd, 3rd, 5th, and 6th abdominal segments had a similar colour to
the ground colour. On the lateral sides of the body, it had a series of darker oblique
15
stripes, running forward towards the ventral surface. The ventral surface of the
caterpillar and the 4th abdominal segment were much darker than the ground colour. The
head was large and shiny black with a paler epicranial suture. This epicranial suture was
a seam that begins at the top of the head and branches toward the compound eyes,
shaped like an inverted-Y. This full-grown caterpillar had a characteristic mode of
feeding on young shoots. If young shoots were available, the caterpillar cut off the
leaves by biting through the petioles of all the leaves. That might leave the shoot, about
one meter in length, with no leaves. Then the caterpillar returned to the top of the shoot,
eating its way from top to the base of the shoot.
Consequently, these caterpillars of initial instars were predated by spiders, black ants
and tree frogs while large wasps and the “Beo” birds (Gracula religiosa L.) attacked
larger caterpillars. Sometimes, they drown in heavy rain as well. The caterpillars chose
green twigs to pupate and angled their bodies at 40–50° to the horizontal planes. The
pupa was in apple green colour with outer margin narrowly bordered with violet. There
was a small marginal square shape spot towards the both apexes. There were two pairs
of sharp, pointed, dorsal processes directed towards the midline and a yellow saddle
dorsal marking between the wing bases. The pupa produced a hissing sound by moving
the abdominal segments when disturbed. During warm weather, these segments
compressed but relaxed at night or rainy days. The pupa stage lasted from 24–27 days
(Straatman & Nieuwenhuis, 1961).
UniversityDahelmi et al. (2008) studied the duration of of immatureMalaya stages of Trogonoptera brookiana trogon with other swallowtail butterflies. He bred nine eggs on Aristolochia
foveolata, the eggs were collected from Harau (about 140 km north of Padang). The egg
incubation period was about 7 days which was shorter than in Straatman and
Nieuwenhuis (1961). The whole caterpillar duration was about 23.7±1.2 days and the
16
duration for every instar was 2.9±0.8 days in first instar, 3.8±0.4 days in second instar,
4.3±0.5 days in third instar, 5.0±0.5 days in fourth instar and 7.1±0.3 days in fifth instar.
The pupa lasted 23.7±1.0 days. Therefore the duration from egg to emergence was
56.3±2.1 days with range of 54 days to 61 days (Dahelmi et al., 2008).
2.3.4 Sex Ratio: Field Sightings
Herr Künstler mentioned that he caught over a thousand males and only fifteen
females during his five years sampling period, in Perak (Distant, 1882–1886).
Furthermore, he managed to collect over 800 males with three men assistance in just
three months but not a single female. He saw twenty to thirty females flying high on
flowers on high trees within that period (Distant, 1882–1886). Wheeler mentioned that
the ratio was 20 males to 1 female in Malaya and he also concluded that the females
flew much higher than the males (Straatman & Nieuwenhuis, 1961). However, for
subspecies trogon which was studied from bred specimens by Straatman and
Nieuwenhuis (1961), the ratio was 2 males to 1 female. While the study conducted by
Dahelmi et al. (2008) on the same subspecies concluded that the ratio was 5 males to 4
females. Eliot (1972) stated that the sexes of mollumar were about equal proportions.
Jumalon (1967) noted the extreme rarity of female of the Trogonoptera trojana, where
he reported only an average of one female to every 30 males that were sighted.
2.4 The Host Plant and Aristolochic Acid
Trogonoptera brookiana is very host specific because only two species of host plants Universitywere reported so far. The known host plantsof were MalayaAristolochia foveolata (Corbet & Pendlebury, 1992; Goh, 1994; Igarashi & Fukuda, 1997; D’Abrera, 2003) and A. kerrii
(D’Abrera, 2003). Aristolochia kerrii was said to be cultivated by Detani in Bali,
Indonesia (Matsuka, 2001). However, the only reported distribution for Aristolochia
kerrii was in Thailand (Phuphathanaphong, 1987).
17
Aristolochia was considered to be a slender climbing shrub or sometimes as high
lianas (Yao, 2015). There are about 400 species in the world. It is distributed throughout
the tropics and subtropics and rarely in warm temperate regions. Three major chemical
constituents can be found in Aristolochia species, they are aristolochic acid, alkaloids
and serquiterpenes (Yao, 2015).
There are six species in Peninsular Malaysia, they are Aristolochia acuminata Lam.,
A. curtisii King, A. jackii Steud., A. minutiflora Ridl. ex Gamble, A. vallisicola T.L. Yao
and A. foveolata Merr (Yao, 2015). The distribution of Aristolochia foveolata is
northeastern part of Sumatra, Peninsular Malaysia, Borneo, the Philippines and Taiwan.
Within Peninsular Malaysia, it has been found in Pahang, Terengganu and Johor. It
occurs in primary lowland and hill forests to 500 m altitude. It has a scattered
distribution and it is near threatened in Peninsular Malaysia (Yao, 2015).
The role of aristolochic acids present in Aristolochiaceae plants to deter natural
enemies have been proved in a study conducted by Pinto et al. (2011) on Battus
polydamas archidamas (Papilionidae, Troidini). In a study conducted in Santiago, Chile
to investigate the effect of aristolochic acid concentration on the feeding behaviour and
development of Battus polydamas archidamas which feed on Aristolochia chilensis
(Pinto et al., 2009). The study concluded that mortality of first instar caterpillar was
higher after they feed on artificial diet with lower concentration aristolochic acid.
However, lower mortality and heavier in weight of caterpillar were observed when the Universitycaterpillar was fed on artificial diet with higherof concentration Malaya of aristolochic acid (Pinto et al., 2009). The young leaves are softer and have higher concentration aristolochic
acid compared to mature leaves. These characteristics could have caused the majority of
ovipositions occurred on young leaves (Pinto et al., 2009). Third instar caterpillar
moved to mature leaves and fourth and fifth instar feed on mature leaves and also the
18
stems. Stems have higher concentration of the acid compared to mature leaves. This
proved that they have phagostimulant effect (Pinto et al., 2009). This probably explains
the majority oviposition of Trogonoptera brookiana trogon occurred on young shoots in
the study by Straatman and Nieuwenhuis (1961).
2.5 Habitat
In general, the adult of the Trogonoptera brookiana occurs at the altitude around
200–1500 m a.s.l. (Igarashi & Fukuda, 1997). The birdwings are thought to breed only
in low mountains and the emergence occurs all year round but with a slight decrease in
number in dry season. The males often fly along streams and in open spaces in forest
(Igarashi & Fukuda, 1997). Similarly to Igarashi and Fukuda (1997), the males of the
subspecies albescens can be found nearly all year round, principally in March, April,
May and June in showery weather (Distant, 1882–1886). They were seen flying over
muddy streams that flowed from the tin mines with overhanging jungle and they can
also be found on decayed animal matter. However, the females were flying high on tree
flowers and unlike males, they were not attracted to baits (Distant, 1882–1886).
D’Abrera (2003) noted that subspecies albescens could be found within the vicinity of
streams or watercourses, or on pathways near puddles formed after rainfall and the
males were attracted to animal urine or thermal springs. It was reported that Burbidge
observed that the subspecies brookiana in Borneo were most numerous by rivers or in
sunny areas by the dry beds of streams and they were most abundant during the cool Universitywet monsoon (Distant, 1882–1886). of Malaya 2.6 Puddling Behaviour
Many species of Lepidoptera adults often visit sweat, tears, moistened ground,
carrion or dung to get nutrients and water, a behaviour called mud-puddling (Arms et
al., 1974). Puddling of Papilio polytes on exposed coral structures or green algae
19
floating mats on Ipan reef shelf located at Guam Island in Micronesia, USA was first
reported by Pola and García-París (2005). The explanations given for this behaviour
were: to take up salt and/or because they were attracted to the high water temperature
(36–42°C) at the sea reef platform.
The males of Trogonoptera brookiana albescens often congregates in the area of hot
springs, or on paths or river banks, or even spots moistened by urine (Corbet &
Pendlebury, 1992). For all puddling butterflies, generally this behaviour has a role in
nutrient (e.g. sodium) intake (Boggs & Jackson, 1991). Igarashi and Fukuda (1997)
reported that this behaviour started as early as 0800 hours where the males of
Trogonoptera brookiana started to gather on moistened grounds in shaded areas to
drink. They are also sensitive to disturbances and immediately take flight when
disturbed. The group will disperse during noon (Igarashi & Fukuda, 1997). The
subspecies brookiana displayed a distinct behaviour when feeding on seepages, the
males flutter then settle with wings horizontal and motionless (Panchen, 1980).
In 1889, Shelley Denton reported that many butterflies especially the larger ones,
attracted to decoy that resembling themselves in size and colour (Denton, 1889).
Papilionids attractions to decoy were found in a study on Papilio glaucus and Battus
philenor (Otis et al., 2006). The study used artificial puddles moistened with sodium
chloride solution. Both species preferred puddles with a decoy compared to un-baited
puddles. The study also found that Papilio glaucus was only attracted to the same Universityspecies decoys while Battus philenor was notof restricted Malaya to the same species decoys (Otis et al., 2006).
Males of Pieris rapae had higher sodium content compared to females (Adler &
Pearson, 1982). Sodium content in fresh emerged males was higher than older males
whereas it was the same in both fresh and old females. They suggested that feeding
20
from soil helped in restoring sodium content in the body of males (Adler & Pearson,
1982). In an experiment conducted by Arms et al. (1974) using Papilio glaucus in New
York concluded that sand contains sodium ion which attracted butterflies to puddle. The
sodium concentration ranges between 10-4 M and 10-3 M in artificial play sand that were
placed in trays and put near to the natural puddling area of the butterfly (Arms et al.,
1974). However, the sodium content on natural puddling sand was higher than the
threshold. They hypothesized that the low sodium content in caterpillar host plant
resulted in the adult butterflies taking sodium through puddling (Arms et al., 1974).
Different species of butterflies may have different preference on the contents in the
puddles. Representatives of Papilionidae and Pieridae preferred sodium solution while
representatives of Nymphalidae, Hesperiidae and partially of Lycaenidae preferred
protein resources was recorded in a study conducted in two consecutive years, October
to November 1996 and August to September 1997 in Kinabalu National Park in Sabah,
Malaysia (Beck et al., 1999). Resources enhanced with decoy especially decoys that
were offered in groups clearly attracted more butterflies, especially the family groups,
Papilionidae and Pieridae but not butterflies under Lycaenidae and Nymphalidae (Beck
et al., 1999). Trogonoptera brookiana brookiana at the study site, however, never
visited the baits (Beck et al., 1999). It seems the resources needed to attract this
birdwing to puddle are not as simple as sodium or protein only. Boggs and Dau (2004)
demonstrated that different butterfly species prefer different puddling substrates such as Universitymud, carnivore dung and herbivore dung inof their study. Malaya In their work using, Pieris napi showed clear preference for mud than the other two substrates. This was due to the
different concentration of sodium in the substrates because the butterflies were more
attracted to sand trays soaked with sodium that matched their natural puddling substrate
concentration (Boggs & Dau, 2004).
21
A study using Papilio xuthus as a model (Watanabe & Kamikubo, 2005), where four
concentrations of saline were given to the freshly emerged males, i.e. 0.001, 0.01, 0.1,
and 1 mol/L. After two days, hand-pairing was conducted. Although the males preferred
0.1 mol/L saline but the largest production of spermatophore was by the males fed on
0.01 mol/L. However, no significant difference was found between the number of
eupyrene sperm (fertile) and apyrene sperm (infertile) between saline-taking males and
control males that fed on water. Males that fed on 1.0 mol/L saline were reluctant to
couple (Watanabe & Kamikubo, 2005).
A serious drought may induce a non-puddling butterfly to puddle. Launer et al.
(1993) had observed a non-typical puddling species, Euphydryas editha bayensis, Bay
checkerspot butterflies of both males and females puddling on the banks of a seasonal
creek in 1990 in the East Hills near Morgan Hill (Santa Clara County, California). Two
major reasons were given and most of the other reasons were tied to the serious drought
in 1990. The first reason was that; drought may have caused the reduction of resources
such as water, mineral and sodium in nature causing the butterflies to fly a longer
distance in search of alternate resources, or in search of mates and oviposition sites. All
the resources stored in the butterflies’ body might have been used up more quickly. The
water in the creek could be their alternative resources. The second reason was probably
due to the creek was abnormally enriched in salts during the1990 drought period
causing butterflies to puddle (Launer et al., 1993).
UniversityDrummond mentioned that in most lepidopterans,of Malaya the males will pass complex spermatophores to the females through mating and the egg complex consist not only the
sperms but various types of nuptial gifts (Beck et al., 1999). A study conducted by
Boggs and Gilbert (1979) reported that nutrients contributed by males to females were
22
used for egg production in three butterfly species through radiotracer studies. The
nutrients could be utilized for somatic maintenance.
Pivnick and McNeil (1987) showed that, 32% of sodium in males of Thymelicus
lineola, a skipper was transferred to female during first mating event. When the males
take up sodium after first mating event, the chances to re-mate on the following day was
also increased (Pivnick & McNeil, 1987). The female will produce more than 50%
fertile eggs if she mated to the twice-mated males. On the other hand, virgin females
have reduced longevity and fecundity (Pivnick & McNeil, 1987). Sodium transfer
during mating was observed in moth Gluphisia septentrionis (Smedley & Eisner, 1996).
The eggs produced by the females mated with males fed with sodium had two to four
times more sodium contents than the eggs produced by control females. About one third
of the sodium content that the male procured during puddling was transferred to the
eggs of his first mate (Smedley & Eisner, 1996).
The toxic secondary plant metabolites, such as pyrrolizidine alkaloids, were also
transferred to the female through spermatophore of Utetheisa ornatrix, a moth under
Arctiidae (Eisner & Meinwald, 1995). Then, the alkaloids from the male and mated
female were transferred to the eggs to protect them from predators (Eisner & Meinwald,
1995). Another study on the same subject proved that the female also gain benefit from
the alkaloids transferred during mating and the alkaloids were distributed evenly in her
body within the first two hours after mating (González et al., 1999). Lai-Fook (1991) Universityshowed that calcium phosphate, a significant of portion Malayaof the spermatophore of Calpodes ethlius, a skipper was found in the bursa copulatrix of the female and phosphorus was
absorbed through the cuticle of the bursa copulatrix. The phosphorus function is to be
discovered (Lai-Fook, 1991).
23
2.7 Flying Behaviour
Notes reported by naturalists were useful in describing the flight pattern of the
Trogonoptera brookiana. Kirby (1896) wrote down the observations by Sir Hugh Low.
Sir Hugh Low mentioned that the flight of the Trogonoptera brookiana in Gopeng
located in Kinta district of Perak, was slow, straight and heavy. It was probably a male.
They usually flew about 10 feet above the ground and then frequently descended and
landed on damp ground. It was very different from the Trogonoptera brookiana found
in Kinabalu which the flight was fast (Kirby, 1896). Panchen (1980) described that a
male of the subspecies brookiana flew with the forewings forming a straight line, wing
beats rapid but somewhat shallow. The flight was rapid but seems difficult to
manoeuvre. The males were seen on almost every sunny day from mid-morning until
about 4.30 pm. But usually they were seen near noon. The flight of a female has much
greater excursion of the wings at every beat. A specific flight pattern was described, a
female flew down from a height of about 12 m at an angle of about 20–30° to the
horizontal but turned away and disappeared from the view (Panchen, 1980). The
Trogonoptera brookiana trogon can fly rapidly if they were pursued however in normal
situation they fly slowly (de Nicéville & Martin, 1895). But the female of the trogon
was a strong flier. Both male and female appeared very early about 7 am in a day and
appeared again in the late evening around 5 or 6 pm. During hottest hours, they were
rarely sighted (de Nicéville & Martin, 1895). University There was a systematic study on butterflies’of Malaya flight periodicity in Peninsular Malaysia conducted by Orr (1982). He included Trogonoptera brookiana albescens in
his study. The study was conducted in Tapah (a small town located in Perak state) for
five days and in Tanah Rata for three days (located in Cameron Highlands, Pahang
state). Observations were conducted at a vantage point beside a stream at each location.
A total of fifty males and eighteen females of Trogonoptera brookiana albescens were
24
observed in eight days observation. Both males and females of the species started flying
as early as 0800 hours, and their last flight occurred at 1700 hours for males and one
hour earlier for females. Highest flight activities occurred between 1000–1400 hours for
both sexes. Higher flight activity at 0800 hours than 0900 hours was observed in female
(Orr, 1982). In a study conducted by Murphy (2008) in Alaska, males of Papilio
machaon aliaska flew earlier in the day than females. The males of Trogonoptera
brookiana, in general, the flight was not very swift but straight and often about 3–7 m
above ground. The females flew more slowly (Igarashi & Fukuda, 1997).
Many Lepidoptera can only fly when the body temperatures reached the thermal
requirements through thermoregulation (Kingsolver, 1985a). There are two mechanisms
of thermoregulation, the physiological and behavioural (Kingsolver, 1985a).
Physiological mechanism involved of muscular contracting during pre-flight or during
flight. This mechanism is usually practiced by moths. Butterflies only gain slight body
temperature through this mechanism (Kingsolver, 1985a). In Queensland, Kemp and
Krockenberger (2004) examined the effect of body size on the heat exchange rates on
Hypolimnas bolina and found that smaller butterflies had greater potential vary the rate
of heat gain compared to larger butterflies but in another hand, they lost heat faster then
the larger ones. An experiment to examine the relationship between habitat
temperatures, solar absorptions and structural or iridescence colour was conducted by
Bosi et al. (2008). In their findings, they found no evidence of relationship between Universitysolar absorption of the wings and temperature of of the Malaya habitat but they found that wings with structural colours had less solar absorption compared to iridescence colours (Bosi
et al., 2008). Melanization of the wings also increased the heat absorption of butterflies
(Kingsolver, 1985a). Kingsolver (1985b) documented that wings of Pieris butterflies
can served as a solar reflector to increase the thorax temperature by basking with wings
positioned at certain angles.
25
Butterflies mainly gain body heat by behavioural mechanism, which involves relative
posture to the sun and behavioural orientation, this is called basking (Kingsolver,
1985a). Flight in Papilio polyxenes was limited to the ambient temperature between 19–
30°C (Rawlins, 1980). At low ambient temperature, the Papilio raised its abdomen
above wings to gain heat from direct sun but hiding the abdomen below the wings
during high ambient temperature (Rawlins, 1980). Papilio polyxenes was unable to
sustain flight when the thorax temperature was less than 23.7°C, while gliding or
soaring flights with lower wing beat frequency were observed when the thorax
temperature was between 23.7–28°C, and vigorous flight was observed above 28°C
(Rawlins, 1980). According to Rawlins (1980) this Papilio would not have survived if
the ambient temperature was above 46°C at low humidity due to overheating.
2.8 Nectaring Behaviour
Information on the nectar plants of the Trogonoptera brookiana specifically on the
subspecies occurs in Peninsular Malaysia was mentioned in several literatures. Corbet
and Pendlebury (1992) reported the plant species Bauhinia (Fabaceae) was exploited as
a nectar plant for subspecies albescens. According to Goh (1994), the adults of
subspecies albescens are known to visit flowers of Crossandra infundibuliformi
(Acanthaceae) and Duranta lorentzii (Verbenaceae). Cole (1998) observed that the male
of subspecies albescens in Fraser’s Hill was taking nectar from the Mussaenda flowers
on the wing and fluttering over the flowers without landing. The sole recorded nectar Universityplant for subspecies trogon is from the Family: of Rubiaceae, Malaya genus Mussaenda (Corbet & Pendlebury, 1992).
De Nicéville and Martin (1895) described the nectaring behaviour of Trogonoptera
brookiana trogon very clearly. The birdwing will not land on flowers but use its
forelegs to snatch the flowers. Then they hover above and softly move their wings for a
26
few seconds then move to another flowers (de Nicéville & Martin, 1895). Straatman
was told by his collectors that Trogonoptera brookiana trogon and other Aristolochia
feeding Papilionids seldom visit flowers in hot weather however, when it rains or when
the weather is misty and wet, the butterflies were seen in numbers flying slowly around
Lantana (Verbenaceae) flowers (Straatman, 1955). The best hours to collect them near
Lantana flowers were between 0600 to 0800 hours and around 1700 hours in the
afternoon, they disappeared into the heavy forest for the rest of the day (Straatman,
1955).
Feeding activity of Trogonoptera trojana was very early, it appeared about 6 am to 3
pm (Jumalon, 1967). Trogonoptera trojana favoured flowers of guava tree and the
Bauhinia sp., however it showed high preference for certain vines compared to others.
Nectar contains mainly water and sugar (dominated by sucrose, fructose and glucose)
but also contain smaller amount of inorganic ion (e.g. K+ and Na+), amino acids,
proteins, lipids, organic acids, phenolics, alkaloids and terpenoids in certain plants
(Nicolson & Thornburg, 2007). Nectars of plantain bananas (Nicolson & Thornburg,
2007) and Nicotiana spp. (Tiedge & Lohaus, 2017) contain amino acids; Bignoniaceae,
Caesalpiniaceae and Saxifragaceae contain lipids (Baker, 1977); Persea americana
contains phenolics (Afik et al., 2006); Gelsemium sempervirens contains terpenoidal
alkaloids (Adler & Irwin, 2012; Manson et al., 2012); Delphinium spp. contain
norditerpene alkaloids (Cook et al., 2013); Coffea spp. and Citrus spp. contain purine
Universityalkaloid (Thomson et al., 2015). of Malaya
Nectar plants play an important role in the life history of butterflies–enhancing
butterfly fecundity (Mevi-Schütz & Erhardt, 2005) and longevity (Murphy et al., 1983).
The presence of sugars stimulated Jalmenus evagoras butterflies to feed (Hill & Pierce,
1989). Although they prefer amino acids over water but the function of amino acids in
27
the biology or physiology of the butterflies was not known. The presence of sugar in the
ratio of 2:2:1 (fructose:glucose:sucrose) at medium level (25% weight/weight)
significantly increased the fecundity and longevity of the Jalmenus evagoras. The body
weight and fat body size were maintained or may even increased (Hill & Pierce, 1989).
On the other hand, Mevi-Schütz and Erhardt (2005) showed that the nectar with
amino acid increased the fecundity of map butterfly, Araschnia levana that bred from
poor quality of host plant, i.e. lacking nitrogen. However, amino acid did not help if the
butterfly bred from good quality of host plant (Mevi-Schütz & Erhardt, 2005). The
effect of mating frequency on longevity of Pieris napi was shown in Mevi-Schütz and
Erhardt (2004), twice-mated female lived longer than once-mated and un-mated female.
This was probably due to the nuptial gifts contributed by the males however, the
compound that helped in longevity was yet to be determined (Mevi-Schütz & Erhardt,
2004).
Tiple et al. (2009) studied the association between adult butterfly feeding and nectar
flowers in India and concluded that butterflies with high wing load indices bias their
feeding to plants with densely packed flowers. Proboscis length is often related to
corolla tube length of flower species visited by Lepidoptera, such that Lepidoptera with
longer proboscis can exploit flowers with deeper corolla tubes (Gilbert & Singer 1975).
Therefore, butterflies pollinated flowers were usually tubular or brush-like (Momose et
al., 1998). Some species show innate preferences for particular colours and differ Universitybetween genera or even between species. Forof example, Malaya Heliconius erato prefers orange (Crane, 1955), H. charitonius preferred orange or red followed by blue or blue-green
(Swihart & Swihart, 1970). Papilio demoleus and P. machaon preferred violets and
blues (Ilse & Vaidya, 1956). Battus philenor preferred yellow followed by blue and
purple (Weiss, 1997). However, Battus philenor can learn to associate floral colours
28
with rewards, this learning behaviour enable the butterfly to adjust their foraging efforts
in response to floral rewards that vary from time to time (Weiss, 1997).
The preference for floral odour was found in hawkmoth Manduca sexta. The nectar
plants of M. sexta were Datura wrightii which is also its larval host plant and Agave
palmeri. Both plants have different odour and nectar composition (Riffell et al., 2008).
Manduca sexta has the innate affinity for Datura wrightii but they were able to shift
their nectar plant to Agave palmeri when D. wrightii was not flowering (Riffell et al.,
2008). Another experiment on the same species, Manduca sexta, showed that the naïve
female preferred Agave palmeri and male preferred Datura wrightii (Alarcón et al.,
2010), whereby, these chosen plants enhanced their energy gained.
Males and females may prefer different flowering plants for nectar. In Switzerland,
Rusterholz and Erhardt (2000) reported that males and females of the Adonis Blue
Butterfly, Lysandra bellargus behaved differently in their nectar foraging behaviour and
nectar composition. The females forage wider area to look for nectar and they preferred
flowers with high glucose and amino acids whereas the males preferred high sucrose
and high amount of total sugar (Rusterholz & Erhardt, 2000). The males of Parnassius
apollo butterfly feed on three species of nectaring plants, namely Armeria arenaria,
Jasione montana and Carduus carpetanus (Baz, 2002). While for the females, they had
additional three nectaring plant species, they are Jurinea humilis, Thymus bracteata and
Anthyllis lotoides. The plant used by the butterflies can differ slightly between sites and
Universitychange within sites through time as well (Baz,of 2002). Malaya
Mevi-Schütz and Erhardt (2004) reported that if a female of Pieris napi remained un-
mated for 5 days or longer, they would shift their preference to take nectar that
contained higher amino acid. This was due to the female did not receive nuptial gift
from male through mating.
29
Nectar concentrations are altered by environmental and biotic conditions. High
temperature, wind, and low humidity lead to evaporation of water from nectar,
concentrating it (Shuel, 1957). Rain can dilute nectar while oil and water nutrient
conditions can also affect nectar concentration (Shuel, 1957). Nectar concentration
varies with altitude in some species, with plants at elevations above 2,400 m producing
less concentrated nectars than plants below 2,400 m (Cruden et al., 1983). This may be
due to differences in abundance of low and high-energy demand pollinators at these
altitudes (Watt et al., 1974).
Nectar production rates vary within and among plants, changing with time of day,
genetic differences, age, nutritional status of the plant, location of flower on the plant,
and other factors (Shuel, 1957; Percival, 1965; Cruden et al., 1983; Pleasants &
Chaplin, 1983). Production of nectar usually starts before onset of pollinator activity
and continues until a critical volume is produced. Production then ceases until the nectar
is removed (Cruden et al., 1983).
2.9 Courting and Mating Behaviour
Courting behaviour descriptions of the Trogonoptera brookiana were found in two
literatures while there seems no description on its mating behaviour. Skertchley (1889)
saw a pair of Trogonoptera brookiana brookiana hovering above an orange-blossom
tree and courting for about 20 minutes in the mountain region of the headwaters of the
River Segama in Sabah. While the male was taking nectar from the flowers, then female Universitysailed down with stately flight and commenced of to woo.Malaya They flew in a circle about 6 inches apart for a long time. The female always uppermost and a little behind then the
male and she pointed her abdomen backwards all the time. The flight was a sailing
motion with a peculiar tremor of the wings very unlike the quivering while feeding.
During the courting, they ignored the feeble attacks by a Ornithoptera flavicollis (a
30
synonym of Troides amphrysus). After a long time, they settled and united high up in a
tree, the female remains uppermost (Skertchley, 1889).
Courting behaviour of the subspecies brookiana was also observed in Gunung Mulu
National Park, Sarawak in August 1978 (Panchen, 1980). At about 2 pm, an old male
was feeding on the Ixora flowers. A female suddenly appeared hovering immediately.
Initially the male continue to feed. The female, moved up and down at a maximum
height of about 30 cm above him. Rhythmically, she came down to buffet him from
above and then rising again. The male apparently reluctant to partner with her and soon
stopped feeding and abandoned the place (Panchen, 1980).
Before mating begins, the female needs to make sure that the mate is the correct
species and is in good quality and the mate achieved this by releasing pheromone
(Wedell, 2005). This happened during courting which serves to promote mating and to
distinguish mates from predators (Scott, 1972). Some females also release pheromone to
induce males to pursue then causes continue courtship (Scott, 1972).
Female behaviour during courting was depends on the age and virginity. Bergman et
al. (2011) used Pararge aegeria to study the relationship between young or old female
and mated or virgin females on the mating behaviour. In their study, the old virgin
females spent 60% in flight while old mated females only spent 20% in flight to search
for mates. An old virgin female took 13.7 flights within 20 minute but a young female Universityonly took 5.5 flights. A male detected a virginof female Malaya quicker than mate female, it took about 9 minutes and 24 minutes respectively. These strategies helped the females to
shorter the time of being unmated (Bergman et al., 2011). Older females were found by
their mates quicker than the younger females (Bergman et al., 2007).
31
Vision and olfactory played vital roles in courtship. Fluttering female, larger female
and brighter colour of female had more successful in courtship due to they were easier
to be spotted by the males (Scott, 1972). Males of Danaus plexippus with deeper shade
orange on his wings mated more frequently compared to lighter orange but they claimed
other traits such as scent played more important toward the result (Davis et al., 2007).
They also found that larger males mated more compared to smaller males.
The behaviour of butterflies to locate mate by the males can be divided into perching
and patrolling (Wiklund, 2003). The males spend their whole live to search their mates
by patrolling without resting unless to re-charge energy or when the weather is not
permissible to fly. The males perch on a vantage point to wait for a female to enter the
area and the vantage point usually is a place where females always occur (Wiklund,
2003). Hilltopping is a system where the males engage perching and patrolling at a
more specific area such as hilltops or ridges to wait for or search for receptive females
while lekking means males form aggregations that females visit only for mating
(Wiklund, 2003).
Ide and Kondoh (2000) generated a mating system model and explained that the
systems were related to environmental conditions, such as temperature and location of a
lek site or an emergence site. Under three conditions, lek mating system was preferred
than searching mating system: (1) males survived better during lekking, (2) female
survived better when visiting a lek and (3) the searching efficiency is higher at a lek Universitythan a emergence site. Searching mating of system Malaya was preferred when the three conditions above was not met (Ide & Kondoh, 2000). Merckx and Van Dyck (2005)
found that the males of butterfly Pararge aegeria behaved slightly different in
woodland and fragmented landscapes. In fragmented landscapes, the males showed
32
higher levels of displacement and also of aggressive fast take-offs which both are the
indicators of patrolling and territorial perching (Merckx & Van Dyck, 2005).
Males usually will not mate as soon as after eclosion, unlike females. Three reasons
were given by Scott (1972). Firstly, the males usually take the role in mate-searching
therefore it needs to have stronger flight and that need times and energy to achieve it.
Secondly, the males usually emerge few days earlier than the females therefore it has
more time to develop scent and quality spermatophore. Lastly, it takes more advantage
to mate or to fertilise the female as soon as it emerged. In two Papilio species, the
females began to mate after 1 and 2 day(s) after eclosion for species polytes and species
demolion respectively while oviposition started 2 and 3 day(s) after eclosion (Dahelmi,
2002).
Territoriality of butterflies usually exhibited in defending mate opportunities (Baker,
1983 reported in Freitas et al., 1995). The two behaviours of mate-locating, perching
and patrolling are also related to territorial defense. A nymphalid Caligo idomenaeus
patrolled around their territorial to interact with intruders and then expelled them
(Freitas et al., 1995) while Favonius taxila, a lycaenid always perched on almost the
same leaf or a shoot and expelled the intruder in specific flight (Takeuchi & Imafuku,
1999). Males of Celaenorrhinus approximate, Astraptes galesus cassius and Mesosemia
asa asa also behaved like Favonius taxila (Alcock, 1988). These species had their Universitylandmark territories (Alcock, 1988). of Malaya Territorial behaviour can be different between species. Eurytides orthosilaus did not
use perching and flying in circles behaviours to defend it territory but they used
patrolling and hovering (Pinheiro, 1990). Papilio thoas used perching, flying in circles,
hovering and patrolling while Battus polydamas used perching, hovering and patrolling
(Pinheiro, 1990).
33
Some territorial behaviour is very aggressive. Eurytides orthosilaus defended the
territory aggressively by grappling and falling to the ground (Pinheiro, 1990). In the
finding by Lehnert et al. (2013), when Papilio homerus defended their territory,
physical contact between resident and intruder butterflies occurred during circular
flights and caused damages on the wings. In their study, resident butterflies always
become the winner, but with a few exceptional incidences (Lehnert et al., 2013).
In a contest, the resident usually returned to his territory most of the time (Takeuchi
& Imafuku, 1999; Rosenberg & Enquist, 1991). Territorial contest at a territory was not
affected by male body mass or the age (Bergman et al., 2007). The winners (residents)
had more success in getting mated than the losers (non-residents) in the territory but if it
was outside of a territory, the non-residents were mated more successful (Bergman et
al., 2007). Kemp (2000) also found that body size did not affect the result of a combat
but age and residency of Hypolimnas bolina did. The males who were the first resident
and the older males always became the winner of a combat. But, whether the winners
were intrinsically better competitors or just had a better opportunity to find territory was
unknown (Kemp, 2000). However, bigger males of Limenitis weidemeyerii under
Nymphalidae were more successful in a combat (Rosenberg & Enquist, 1991).
Territorial of the Trogonoptera was reported by several authors however the
relationship between territorial and courting was not determined. D’Abrera (2003)
mentioned that the males of albescens were strongly territorial which flew in defined Universityflight corridors. Cole (1998) thought that theof Trogonoptera Malaya brookiana albescens male was territorial because during his stay in Fraser’s Hill in 1997, an observation spot
consists of Mussaenda frondosa plant, a male flew up each time a new butterfly
entering the spot to check its identity. Around the same spot, a male was also patrolling
back and forth, sometimes alongside with a Troides helena. The female occurred, about
34
two or three times over a period of an hour in the same spot visiting the Mussaenda
flowers (Cole, 1998).
A solitary male of subspecies brookiana was observed patrolling up and down near a
rock shelter of the cliff face on sunny mornings around 9.30 am throughout the day until
4.30 or 5.00 pm when there was no direct sunlight (Panchen, 1980). Another territorial
behaviour described by Panchen (1980) occurred in the forest fringe with some
flowering Ixora plants. There were two males “nesting” on a particular leaf of the
particular branch on each of the two trees near two Ixora plants. The height of the leaf
stations were about 2.5 m and 2.0 m, respectively. The male from the each nesting
station left their “nesting” stations to patrol along the dry river for several hundred
metres or to visit the “home” Ixora flowers to feed. The earliest and latest times of this
behaviour happened were unknown but it was seen from about 11.00 am to about 4.00
pm (Panchen, 1980).
2.10 Monitoring of Butterflies
Hellawell (1991) mentioned that a monitoring programme is needed to monitor the
effectiveness of policy or legislation, to monitor performance for management purposes
and to detect incipient change. Among insect groups, butterflies are considered to be the
more suitable insect for a monitoring programme because butterflies are well-
documented, easy to recognize and also popular to the public (van Swaay, 2014). They
are however, sensitive to environmental changes with short generation time making Universitythem also suitable candidates as bio-indicators of (BonebrakeMalaya et al., 2010). This was supported by the work done by Lomov et al. (2006) using butterflies in restoration
monitoring conducted in Australia.
In European countries, butterfly monitoring is well established. Some examples of
monitoring scheme establishment are, United Kingdom since 1976, Germany since
35
1989, The Netherlands since 1990 and several others (van Swaay, 2014), Illinois since
1987 (IBMN, 2015) and Ohio since 1995 in United States (PollardBase, 2016). In the
tropics, there is no long-term butterfly monitoring scheme established until several
years ago in some forest dynamics plots monitored by Center for Tropical Forest
Science (CTFS), such as in Panama, Thailand and Papua New Guinea (Basset et al.,
2011).
2.11 Capture-recapture Method
The analysis of capture-recapture data can provide robust predictions of animal
population size. From the population samplings done, there are a number of equations
available such as: Chapman estimator, Bayesian estimator and Jolly-Seber model to
arrive at a population estimate, however the Lincoln Index is the simplest way and using
the calculation:
a b A B
A = Number of butterflies in the population
B = Number of animals marked on the first visit
a = Number of animals captured on the second visit
b = Number of recaptured animals that were marked
UniversityIn order to use this method, some assumptions of must Malaya be fulfilled (Southwood, 1971):
(1) The marked animals are not affected by being marked and the marks will not
lose;
(2) The marked animals become completely mix in the population;
36
(3) The population is sampled randomly with respect to its mark status;
(4) Sampling must be at discrete time intervals and the actual time involved in
taking the samples must be small in relation to the total time.
When using the simple Lincoln Index, it is also assumed that:
(5) The population is a closed one or, if not, immigration and emigration can be
measured or calculated;
(6) The are no births or deaths in the period between sampling or, if there are,
allowance must be made for them;
(7) Being captured one or more times does not affect an animal’s subsequent
chance of capture.
However, it is time-consuming to mark a large number of butterflies as stated by
Douwes (1976) and Nowicki et al. (2008). The technique is also not suitable for sparse
populations or elusive species where the numbers of captures may be too low (Haddad
et al., 2008; Murphy, 1988). The handling will sometimes affect negatively on the
populations and the behaviour (Nowicki et al., 2008; Murphy, 1988). The handling
effect will cause the butterflies from returning to the captured sites, and according to
Mallet et al. (1987), the effect lasted for two days after capture. Also, it can lead to
increased mortality rates (Murphy, 1988), increased emigration rates (Gall, 1984
Universityreported in Gall, 1985) and changes in activityof patterns Malaya (Mallet et al., 1987). Studies
conducted on Graphium sarpedon in Sarawak showed that the recapture rates were
increased when the butterflies were marked without handling them (Singer & Wedlake,
1981). When the Graphium sarpedon accumulate on river bank, a person laid flat on
stomachs and holding a marker with both hands out in front and moved like a worm
37
towards the butterflies and marked them without handling. The position of holding
marker was to un-cap the marker without much movement (Singer, personal
communication, November 6, 2014). The technique was applied on Polyommatus
icarus, Lysandra coridon and Cupito minimus in the temperate zone and at least
Polyommatus icarus and Lysandra coridon showed difference between handled and
unhandled specimens (Morton, 1984).
2.12 Transect Walk Method or Pollard Walk
Transect walk method or Pollard walk was developed by Pollard (1977). His
recording is based to the individual butterfly sighted within an imaginary 5x5x5 m box
in front of the observer walking at a uniform pace along a fixed route. Surveys are
conducted once every week during flight season under defined weather conditions and
time of day. Mean weekly count of each species is summed up to get index of
abundance for each generation. This method only provides relative abundance indices.
This method is well established in European countries. Pollard walk has been
implemented in Butterfly Monitoring Scheme at 100 sites in United Kingdom for the
past 17 years (Harding et al., 1995).
However, the detection probabilities were shown to be imperfect, which fully
depended on the time of sampling and also seasonal abundance of butterfly species
(Pellet, 2008). Site characteristics and behaviour of adult butterflies, e.g. Lasiommata
megera also affected detection probabilities (Harker & Shreeve, 2008). In tropical Universityforest, conducting Pollard walk can be problematicof Malaya because the difficulty in species identification in the field. The abstract of Walpole and Sheldon (1999) concluded that
less than 50% of butterflies recorded using transect walks could be identified to species
level in a study conducted in Kalimantan, Indonesia.
38
Another study suggested that when the study plot is small, timed survey generated
more meaningful result compared to Pollard walk (Kadlec et al., 2012). The whole
study plot was surveyed and the survey path was not fixed which varied among visits,
and focussed samplings on patches with abundance resources i.e. flowers patches.
Another similar method to transect walk, an area census method was tested to be able
to provide sufficiently accurate estimate of population size and the estimates were
highly correlated with absolute population size using capture-recapture method
(Douwes, 1976). The butterflies were counted in a 50 x 100 m area by a zig-zag walk
pattern. The area census method took 25 min to finish while the mark-release-recapture
took 2–3 man hours each time.
The transect walk method was evaluated for monitoring of moth, Luperina nickerlii
leechi (Spalding, 1997). In the study, although the number of observed moths using
transect walk method and the recapture rate using capture-recapture method were low,
the transect walk method was still recommended for sedentary moth species when the
index of abundance was incorporated for the monitoring programme of annual
population index.
2.13 Other Methods
Trapping methods, such as Malaise traps and bait traps can be used in monitoring.
However, Malaise traps can only cover a small area and it is technically difficult to Universityhandle (Nowicki et al., 2008). Malaise of trap was Malaya not recommended for butterfly sampling in a study comparing efficiency of each trap for several insect groups
(Lamarre et al., 2012). But intercept traps together with bait traps were recommended
because more butterflies were collected from intercept traps compared to Malaise trap.
Many butterflies collected were from subfamily Satyrinae, therefore bait traps can be
complemented to intercept traps.
39
The bait traps design is variable. Generally, it contains a cylinder of nylon netting
with a plastic dish cover the top and the bottom part is open to allow butterflies to go in
to the trap. A tray containing bait is at the bottom of the trap and it is tied to the
cylinder. These traps are placed at variable heights of the trees. The common bait used
is rotting or fermented bananas mixed with sugar water and alcohol, sometimes baits
such as rotting fish or shrimps are used as replacements for fruit bait. Hamer et al.
(2006) found that carrion attracted more butterflies species compared to fruit bait. Bait
traps attracted certain butterfly groups only (Hughes et al., 1998), its area coverage is
unknown (Nowicki et al., 2008) and the attractiveness of the bait is not known too
(Nowicki et al., 2008; Hughes et al., 1998). However, bait trapping is still a useful
method to monitor species abundance changes over time series (Hughes et al., 1998).
Besides that, the method can be used to compare spatial dan temporal patterns of
species composition and abundances between sites (DeVries et al., 1999; DeVries &
Walla, 2001), and also to track the movement of butterflies when the method combines
with mark-recapture method (Tufto et al., 2012).
The potential of using immature stages (eggs, late-instar larvae and the feeding
damage inflicted by larvae) as measures of population status for monitoring purposes
was explored by Lindzey and Connor (2011) for Mission blue butterfly (Icaricia
icarioides missionensis Hovanitz). They showed that immature stages and feeding
damage were highly correlated with adult counts especially the late instars larval counts Universityand foliar feeding damage. They also claimed of that immature Malaya stages had higher detection probabilities than the adults and the method is repeatable and can be conducted in a
variety of weather conditions.
40
2.14 Criterion to be Considered in Choosing Monitoring Method
Butterflies detectability is always a concern in a monitoring programme (Kéry &
Plattner, 2007; Pellet, 2008). Detectability varied among butterfly species (Kéry &
Plattner, 2007) it is because different butterflies behave differently and occur in
different habitat (Pellet, 2008). Relative abundance may influence detectability as well.
Each butterfly species has its own period of seasonal abundance as demonstrated clearly
in a study conducted by Pellet (2008). Other factors that should be taken into account
are variations in observers, transects and total surveys that can affect detectability (Kéry
& Plattner, 2007). Furthermore, Pellet et al. (2012) mentioned that the detectability
tends to vary in space (among sites) and time (among years). Therefore, the monitoring
method should be chosen based to their behaviour and habitat, e.g. transect method is
not suitable for species that occurs primarily in the tree canopy (Harker & Shreeve,
2008) and capture-recapture method is not suitable for small-sized butterfly species
(Murphy, 1988).
Peak abundance of the year should be chosen for conducting monitoring (Pellet,
2008). In a study conducted in Switzerland in 2003 under Biodiversity Monitoring
Programme, it clearly showed that between May and August 2003, more species were
counted in July compared to other months (Kéry & Plattner, 2007). However, the
abundance fluctuates each year due to climatic variability and uncertainty (Zonneveld et
al., 2003). In temperate countries with clear annual flight periods, such as in California, Universitythe weather did not seem to be a major factorof affecting Malaya butterfly flight in a study on seasonal variation in the four species of butterflies (Pellet, 2008).
Harker and Shreeve (2008) found that transect count of Lasiommata megera
conducted in different time of a day generated different population pattern during their
flight period. Transect count conducted in morning (1045–1200 BST) and midday
41
(1200–1400 BST) produced very different pattern compared to Jolly-Seber population
estimates using mark-release-recapture method, while the afternoon (1400–1545 BST)
produced a better match. They found that the butterflies were more encountered during
afternoon hours, and probably contributed to the match.
2.15 Seasonality
Seasonal changes in species diversity and abundance in a tropical country is not as
simple as in a temperate country (Owen et al., 1972; Hamer et al., 2005). Seasonality of
species richness and abundance of fruit-feeding butterflies was affected spatially and
temporally suggesting that a longer term monitoring at different vertical position was
needed in order to understand butterfly communities and to better conserve them in
tropics (DeVries & Walla, 2001; Molleman et al., 2006; DeVries et al., 2011). Many
butterfly species in the tropics can potentially breed throughout the year but they
reached a seasonal peak of abundance during a restricted period. In the study, they
found that each species of Acraeidae butterfly had different abundance peak throughout
sampling months from 1 Oct 1968 to 31 Dec 1970 in a specific locality on Mount
Aureol near Freetown, Africa. The factors that contributed to seasonal changes were
rather complex, which comprised of: the combination of wet and dry seasons, larval
food plants availability and avoidance of adult food plants competition (Owen et al.,
1972).
Seasonality pattern of abundance and richness of fruit feeding butterflies was studied Universityfrom September 2005 to August 2006 inof Brazil (NobreMalaya et al., 2012). The butterfly abundance and richness dropped significantly during the dry season due to plant
defoliation and lack of fruits. However, the decline in population only began after the
first month of dry season, this was because the vegetation and fruits were still available
and capable of supporting the butterfly fauna within the limited period of time.
42
Consequently, during the initial month of wet season, the butterfly population did not
increase immediately because available food resources were still low in density (Nobre
et al., 2012). They also found that, when the season became too hot, the butterflies tend
to search for cooler areas. However, in a study using carrion baits, conducted in
Ecuadorian Amazonia from April 2002 to April 2003 found higher captures during
periods of high temperatures and low rainfall (Checa et al., 2009).
From a 15-year population data collected in British Butterfly Monitoring Scheme,
positive associations were found between butterfly abundance with warm summer
temperature and low rainfall during the current and previous year (Roy et al., 2001). A
study on drought and rainfall effect on Ragadia makuta population was conducted in
Danum Valley Field Centre and Ulu Segama Forest Reserve in Sabah from 1996 to
1999 (Hill et al., 2003). The results revealed a significant drop in butterfly abundance in
the year 1997 compared to the subsequent years of 1996, 1998 and 1999. This was due
to the effect of El Niño Southern Oscillation in 1997. However, in the same study,
monthly abundance changes were conducted from March 1999 to February 2000
indicated that the abundance was significantly and negatively related to rainfall in the
month before the survey was conducted. This indicated that the effect of rainfall
towards the butterfly abundance was not linear and it was adversely affected by both
high rainfall and drought (Hill et al., 2003).
However, the seasonality of the butterflies is not always influenced by the weather. UniversityNowicki et al. (2009) argued that the influence of of environmentalMalaya is not always correct such as weather and mowing process effect was weak towards the population dynamics
of Maculinea alcon and M. teleius butterflies. They claimed that the 12-year population
dynamic of the butterflies was influenced greatly by density-dependent regulation.
43
Yamamoto et al. (2007) conducted a study on the Urato islands (includes Mahanashi
island, Katsura island, Sabusawa island, Nono island and Hoh island) in Matsushima
Bay, Japan. The study demonstrated that the relative abundance of resources (host
plants) has a highly significant effect on the relative abundance of consumer species.
Their results indicated that the amount of butterflies’ resources strongly influences the
abundance of butterflies and, consequently, butterfly biodiversity patterns (Yamamoto
et al., 2007). In northern Western Ghats, population of a smaller Lycaenid showed rapid
increased at the time when the plants were in suitable phenophase for growth of the
caterpillars (Kunte, 1997). A study conducted in Western Uganda between May 2000
and December 2011 found that butterflies species richness and abundance peaked about
2 and 3 months respectively after the greenness peak due to the availability of the
butterflies host plants (Valtonen et al., 2013).
2.16 Threat and Protection in Malaysia
This butterfly is protected under Malaysian law (Laws of Malaysia, 2010) and listed
under Convention on International Trade in Endangered Species of Wild Flora and
Fauna Appendix II (CITES, 2017). Collection without a permit is prohibited.
Unfortunately, the extraordinarily high demand by collectors for these spectacularly
beautiful butterflies has created a market that is placing great pressure on already
dwindling populations (Phon, 2008). Middlemen pay indigenous people–who have
traditional rights to hunt wild animals–a very low price even for a good specimen, Universitywhich is then mounted for display or made ofinto decorative Malaya souvenirs. These commercial products fetch prices more than ten times of the original purchase value (Phon, 2008).
For every beautifully preserved specimen or souvenir that ends up being sold, many
others are discarded because damage by improper collection techniques renders them
worthless (Phon, 2008). According to UNEP-WCMC (2007), from 1985–2005, about
44
37,259 Trogonoptera brookiana specimens were under trade, mainly collected from the
wild in Malaysia.
Over-collecting could cause the population to decrease, however, an even greater
threat is deforestation for timber exploitation, agriculture and development, which
destroys their habitat and food plants (Collins & Smith, 1995). Protected forests in
Malaysia do not sufficiently protect the habitat of the birdwing. Much of its habitat lies
outside of the range of totally protected forests. Forests in which it occurs are vulnerable
to commercial exploitation. For example, the habitat of subspecies mollumar is rapidly
depleting in the lowlands of Johor. And a project to salvage an old water pipeline
almost destroyed one of the largest puddling sites of the birdwing in Ulu Geroh (Chan,
2010; Chiew, 2010; Phon et al., 2011).
University of Malaya
45
CHAPTER 3: TAXONOMY STATUS OF TROGONOPTERA BROOKIANA
MOLLUMAR
3.1 Introduction
Trogonoptera brookiana (Wallace, 1855) is a large and iconic birdwing butterfly of
conservation importance that is listed under Convention on International Trade in
Endangered Species of Wild Flora and Fauna Appendix II (CITES, 2017). Despite
being well known, there remain questions about the taxonomy of its geographically
separate populations, particularly the subspecific status of the eastern Peninsular
Malaysian population that occurs from south-eastern Johor to south eastern Pahang and
more rarely in the southern half of eastern Terengganu. This population, which does not
appear to interbreed with subspecies albescens on the western side of the central
mountain range, was originally regarded as the Sumatran subspecies trogon
(Vollenhoeven, 1860) by Corbet and Pendlebury (1956), but later described as a new
subspecies, mollumar D’Abrera (in D’Abrera et al., 1976). However, Haugum and Low
(1982) examined in detail the characters used in its description and disputed its
subspecific status, treating it only as a locality form. Subsequently, Eliot, in Corbet and
Pendlebury (1992), formally synonymised mollumar with trogon, probably based on
Haugum and Low’s views. Despite this, Matsuka (2001) and Ohya (in Matsuka, 2001)
retained usage of mollumar as a subspecies. In response to Haugum and Low, D’Abrera
(2003) clarified its authorship, argued against synonymy and formally reinstated Universitymollumar as a subspecific taxon. of Malaya D’Abrera used six characters on the wings of males and three on the wings of
females to differentiate between mollumar and trogon, in addition to differences in the
male genitalia. Haugum and Low disagreed with almost all the differences described by
D’Abrera. They concluded that the males do not differ and that the females are almost
indistinguishable, but are likely to have a darker ground colour and occasionally an
46
extended green area on the hindwing. They stated that the characters described for
mollumar could also be found in some specimens of trogon, and that D’Abrera may
have examined too few specimens of trogon to make a valid comparison. Haugum and
Low’s sample size is difficult to determine accurately because the number of specimens
stated as examined in the text differs somewhat from the number listed in their section
on material examined. Assuming the latter is more reliable, they examined many
Sumatran males (212 specimens) but much fewer Sumatran females (10 specimens),
and very few specimens of either sex from Peninsular Malaysia (3 males and 4
females). The number of specimens examined by D’Abrera was not stated.
Previous attempts to diagnose character differences between subspecies trogon and
the nominal taxon mollumar have been limited by inadequate examples of females of
trogon and both sexes of mollumar. In addition, the characters used are sometimes
difficult to apply objectively. Therefore the specific objectives for this taxonomic study
(based on morphology) are:
1. to examine differences between the Sumatran and Peninsular Malaysian
populations; and
2. to discuss issue on whether the nominal taxon mollumar merits subspecific status,
based on the analysis.
3.2 Literature Review UniversityThere was a refutation against Haugum of and Low Malaya (1982) arguments in D’Abrera (2003). D’Abrera et al. (1976) think that mollumar warrant a subspecies status because
of the significant differences on the male genitalia and the external adult morphology.
The characters described in D’Abrera et al. (1976) are listed below. Male: (1) The green
discal area on the upperside hindwing is more extensive in mollumar than trogon, it is
47
extended to more than half the distance from the base of the hindwing to the dorsum;
(2) the distal margin of the green disc on the upperside hindwing is very noticeably
convex in mollumar while specimens of trogon are either straight or concave; (3) green
scalling suffused in space 7, just above vein 7 of upperside hindwing of mollumar and
this character is very occasional and weakly developed in trogon; (4) the vein 8 in the
hindwing of mollumar is more bowed close to the costa and vein 7 is less bowed along
its length; (5) the apical margin of the hindwing of mollumar is more sharply angled
away from the costa, and (6) mollumar do not have scalloped or incurved section of the
dorsum between vein 7 and 8 on the hindwing. Female: (1) The green discal area of the
upperside hindwing mollumar is more extensive than the trogon; (2) the whitish sub-
apical area of the upperside forewing is better developed on mollumar than the trogon,
and (3) the sub-marginal white spots on the hindwing also better developed in
mollumar. Genitalia: (1) Spatulate harpe with tapering neck and elongate body was
observed on the holotype genitalia while this spatulate harpe is poorly developed and
just barely in-relief to the clasper of trogon; (2) The valvae of mollumar elongate with
better defined mid-marginal tooth than on trogon; (3) the vinculum mollumar is bulkier
and blunter than on trogon; (4) the apex angularis of mollumar is short, blunt and
sharply downcurved while it is long slender and very gently bent on trogon, and; (5) the
aedeagus head of mollumar is tubiform and prognathic at its lower extremity while the
aedeagus head of trogon is narrow and characterised by two lateral delta-shaped Universityprocesses with a finely scooped-out apex. of Malaya Unfortunately, the number of specimens and details of specimens examined were not
mentioned in the original description. The holotype and allotypre are kept in BMNH.
The holotype labelled as “Ulu Sedili, Johore, W. Malaysia (V. Doggett), 17th February,
1974” with forewing length 85.0mm. While the allotype labelled as “Ulu Sedili, Johore,
W. Malaysia (N. Parker), 29th August, 1971” with forewing length 69.5mm. The genital
48
description of mollumar for the original description was prepared by Huggins, a staff in
the BMNH and during the preparation it suffered slight damages (D’Abrera, 2003). The
genitalia were examined by D’Abrera and published the drawings (Figure 3.1) of both
subspecies mollumar and trogon in the original description (D’Abrera et al., 1976). The
genital was re-treated by Vane-Wright later and the description was published in
Haugum and Low (1982). The re-treated genital was then compared with the original
description in the book.
UniversityFigure 3.1: Genitalia illustration of Trogonopteraof Malaya brookiana mollumar and T. b. trogon scanned from original description (D’Abrera et al., 1976). (a) genitalia of T. b. mollumar; (b) opposite clasper of T. b. mollumar; (c) genitalia of T. b. trogon; (d) opposite clasper of T. b. trogon.
Almost all the characters especially the genitalia description by D’Abrera et al.
(1976) were disagreed by Haugum and Low (1982). They claimed that the mollumar
49
females almost cannot be distinguished from trogon but likely to have a rather darker
ground colour and occasionally extended green areas in the hindwings while the males
do not have any differences. They examined probably about 212 males and 10 females
from Sumatra and 3 males and 4 females from Johore were listed. Their comments on
D’Abrera et al. (1976) description are listed below (Haugum & Low, 1982). Male:
Subspecies mollumar is very constant and not constantly distinguishable from trogon,
both upperside and underside of wings. They found some trogon males also have
relative extensive green area on the upperside hindwing as in mollumar and green
scalling may well extend up past vein 7. Female: They agreed with D’Abrera et al.
(1976) that the females have more extended areas of iridescent green scaling on the
upperside hindwing. The consistency of this character convinced Haugum and Low
(1982) that the character can be used to differentiate this subspecies from the other
subspecies of brookiana. Additional to that, mollumar has darker ground colour. Other
than these two characters, they are not differing from trogon. They did not agree with
D’Abrera et al. (1976) on the whitish sub-apical area character. Genitalia: The holotype
genitalia were claimed to be distorted, not properly cleared and covered in scales.
Furthermore, it was partly damaged and was depicted in a tilted position therefore
giving a distorted projection in the original description. The holotype genitalia was re-
treated and re-examined by Vane-Wright in BMNH and provided his diagnosis to
Haugum and Low (1982). He probably examined two males of mollumar (holotype and
other specimens from Mersing, Johore) and two males of trogon (both from Sumatra Universitywith one specified from Banissa Range). Hisof diagnosis Malaya is listed here: (1) the valvae are similar and there isn’t mid-marginal tooth in both subspecies genitalia, the mid-
marginal tooth mentioned in original description is due to damage (he claimed that the
bases of both valves and valve margins are split; (2) the valve (clasper) of mollumar is
not notably produced or elongate; (3) the harpe of mollumar is neither spatulate nor
50
produced; (4) the aedeagus, pseudouncus, penis, saccus and basal areas of the valves are
similar for both subspecies; (5) the sculptured saccus mentioned in the original
description was claimed to be due to part of the lamella being destroyed in the
dissection; (6) there is a little, asymmetrical right lateral ‘lobe’ (=ostium folds ?) at the
tip of the aedeagus in both subspecies, and this probably is a specific character
represented in all races of Trogonoptera brookiana; (7) the subspecies mollumar has a
longer and more pointed dorsal projection to the clasper; (8) a thickened process below
the anellus of the juxta was observed on both subspecies but it is thicker and slightly
more massive in the two males of mollumar than in two males of trogon, and; (9) there
is a dorsal spine of the left side of the left apex in the specimen from Mersing, Johore.
The last three characters were not included in the original description. Based to the
above comments, Haugum and Low (1982) concluded to put mollumar as a locality
form of subspecies trogon.
3.3 Methodology
3.3.1 Collections Visited and Specimens Examined
A total of 182 males and 14 females from Sumatra, 22 males and 18 females from
Johor, south-eastern Pahang and Terengganu in Peninsular Malaysia were examined.
The specimens were from the collections of the Natural History Museum London
(BMNH), Naturalis Biodiversity Centre, Leiden comprises collections from the Dutch
Rijksmuseum van Natuurlijke Historie (RMNH) and Zoological Museum Amsterdam University(ZMA), Entomological Reference Collection of of ForestMalaya Research Institute Malaysia, Kepong (FRIM) and the Museum of Zoology in University of Malaya (MZUM).
Additional specimens examined from photographs, were from the Lee Kong Chian
Natural History Museum, Singapore (LKCNHM), the Chong-Arshad collection, the Dr.
Laurence Kirton’s collection and the Museum of Comparative Zoology, Harvard
51
University (MCZ). Several unlabelled specimens and a small, bred female were
excluded.
3.3.2 Description of Characters Examined
Several wing characters mentioned by D’Abrera et al. (1976) and Haugum and Low
(1982) were quantified, some of which were disputed by the latter. Characters in which
no perceptible differences were found between populations were excluded. Characters
that were chosen demonstrated the relative size and extent of the green disc on the
hindwing in both sexes using ratios, the shape of the green disc in the male, the
distalmost extent of the green scaling on both sides of the forewing in the male, and the
ground colour in the female. Wing size was an additional variable. The characters used
are defined in Table 3.1 and are illustrated in Figures 3.2 and 3.3. Measurements and
scores were generally made with a ruler on physical specimens or image analysis
software on photographed specimens on the right wings, and on the left wings only
when damage to the former precluded its use. When both wings were damaged, affected
characters were excluded from analysis.
University of Malaya
52
Table 3.1: Character descriptions. Scores given for each categorical data are given in parentheses. Numerators and denominators for measured characters are defined in Figures 3.2 and 3.3.
Character Sex Character Calculation or abbreviation categorisation FwL ♂ Forewing length (cm) measured on the AB ♀ underside from the base of the forewing costa to the distalmost margin of the apex. UpHwSp5GrRatio ♂ On the hindwing upperside, a ratio formed CD / CE ♀ by the distance to which the green scaling in space 5 reaches (D) on a straight line between the base of the cell (C) and end of vein 6 (E), divided by the entire length of the straight line. UpHwSp4GrRatio ♂ On the hindwing upperside, a ratio formed FG / FH ♀ by the distance to which the green scaling in space 4 reaches (G) on a straight line between the base of vein 5 (F) and end of vein 5 (H), divided by the entire length of the straight line. UpHwSp2GrRatio ♂ On the hindwing upperside, a ratio formed CI / CJ ♀ by the distance to which the green scaling in space 2 reaches (I) on a straight line between the base of the cell (C) and end of vein 3 (J), divided by the entire length of the straight line. GColour ♀ On the upperside of both wings, the ground (1) Brown, or colour of the dark distal areas, in (2) black (as comparison to the black basal half of the black or forewing cell. nearly as black as the cell) UpFwV7Gr ♀ On the forewing upperside, the density of (1) None, (2) postdiscal green scales just below vein 7. sparse, or (3) dense UnFwSp4Gr ♀ On the forewing underside, the density and (1) None, (2) extent of discal green scales in space 4. sparse, or forming a (3) broken patch or (4) dense patch UniversityUpHwGrDisc ♂ On the hindwing of upperside, Malaya the shape of the (0) Indented, distal margin of the green disc. (1) straight, (2) convex UpHwSp7Gr ♂ On the hindwing upperside, the density of (0) None, (2) discal green scales in space 7 just above sparse, or (3) vein 7. dense
53
Figure 3.2: Illustrations of measurements and character categories in females.
University of Malaya
Figure 3.3: Illustrations of measurements and character categories in males.
54
3.3.3 Analyses
Measurements were log transformed to equalise variances prior to analysis. The
Hotelling T-square test was used to test for significant differences in the ratios or
measurements of each sex of the Sumatran and Peninsular Malaysian populations.
Simultaneous confidence intervals were then calculated for each character to determine
which characters were significantly different. Characters that were significantly
different and less variable between populations were then incorporated into a
discriminant function analysis with cross-validation.
Simple correspondence analysis was used for the categorical characters of each sex,
to explore their association with each other and with the Sumatran and Peninsular
Malaysian populations. For characters that formed stronger species associations in the
correspondence analysis, chi-square tests were used to determine if they were
significant, with the significance level taken as the normal α value of 0.05 divided by
the number of tests, to compensate for the likelihood of spurious significance. Minitab®
15.1.1.0 was used in all the analyses.
In order to facilitate recognition of each specimen in a correspondence analysis, each
specimen was given a code. The specimens collected from Sumatra were labelled as
FT1, FT2 and so forth while from Peninsular Malaysia were labelled as FM1, FM2 and
so forth. Similar to the females, each male was given a code such as MT1, MT2 and so
forth for the specimens from Sumatra and MM1, MM2 and so forth for Peninsular
UniversityMalaysia. of Malaya
3.4 Results
3.4.1 Female
A summary of the measurement data for female birdwings from Sumatra and
Peninsular Malaysia is shown in Table 3.2. Hotelling’s T-Square test showed a
55
significant difference between Sumatran and Peninsular females in at least one
measured character (T-Square=1.473, F=8.469, df=4,23, p<0.001). Using the
simultaneous confidence intervals used to determine which characters differed
significantly, there was no significant difference in forewing length, which differed on
average by less than one millimetre. The ratios for the extent of green that reached space
2 and space 5 on the hindwing upperside showed small differences on average between
populations, but with great variation, and were not significant. Sumatran females had on
average slightly more extensive green in space 2, but slightly less extensive green in
space 5, in comparison to Peninsular females. The ratio for extent of green in space 4
along vein 5 differed more greatly and was significantly different between populations.
Peninsular females had on average more extensive green scaling in space 4 than
Sumatran females. However, this character was not fully diagnostic, because the ranges
for the two populations partly overlapped. In a discriminant function analysis for this
character, 25 out of 28 specimens (89.3%) were classified correctly to their original
regions, with a squared distance between populations of 4.51. The discriminant
functions for the Peninsular and Sumatra populations were
Female, Peninsular Malaysia = −16.88 – 118.22 × log(UpHwSp4GrRatio)
Female, Sumatra = −31.48 − 161.42 × log(UpHwSp4GrRatio)
Table 3.2: Measurements and ratios for Sumatran and Peninsular Malaysian females (mean ± standard error, and range). UniversityCharacter abbreviation Sumatra of (n=14) Malaya Peninsular Malaysia (n=14) FwL 7.8±0.1 8.0±0.1 (7.3–8.5) (7.4–8.8) UpHwSp4GrRatio 0.41±0.02 * 0.52±0.01 * (0.35–0.55) (0.45–0.62) UpHwSp2GrRatio 0.73±0.01 0.71±0.02 (0.64–0.81) (0.59–0.78) UpHwSp5GrRatio 0.67±0.01 0.71±0.01 (0.56–0.75) (0.58–0.78) * significantly different (Hotelling T-square test, α = 0.0125)
56
Table 3.3 summarises the data for the categorical characters of females, Table 3.4
shows the result of simple correspondence analysis and Figure 3.4 shows a two-
dimensional plot of the first two components of simple correspondence analysis.
Sumatran and Peninsular specimens were distributed on opposite ends of the two-
dimensional plot, primarily because of the first component. Sumatran specimens were
strongly associated with a brown ground colour (GColour_Brown), less green scaling in
space 4 of the forewing underside (UnFwSp4Gr_None or Sparse), and less green
scaling just below vein 7 on the forewing upperside (UpFwV7_None or _Sparse).
Peninsular specimens were strongly associated with a black ground colour
(GColour_Black), a dense and complete or broken patch of green scales in space 4 of
the forewing underside (UnFwSp4Gr_BrokenPatch or _DensePatch), and dense green
scaling just below vein 7 on the forewing upperside (UpFwV7Gr_Dense). The
difference in ground colour was sufficiently distinct to be discerned and was diagnostic
for the samples of the two populations. All Sumatran females were categorised as
brown and all Peninsular Malaysian females as black. Differences between the green
scaling below vein 7 on the hindwing upperside and in space 4 of the forewing
underside were significant (Chi square test, Table 3.3). About 89% of the Peninsular
females had dense green scaling below vein 7, compared to 29% of Sumatran
specimens. All Peninsular females had a patch or broken patch of green scales in space
4 of the forewing underside, whereas 93% of Sumatran females had either no scaling or Universitysparse green scales, with only 7% having a ofbroken patch.Malaya
57
Table 3.3: Numbers and percentages (in parentheses) of females with each character from fourteen Sumatran and eighteen Peninsular Malaysia females.
Characters (abbreviated) and categories GColour UpFwV7Gr # UnFwSp4Gr * Region Brown Black None Sparse Dense None Sparse Broken One patch dense patch Sumatra 14 0 3 7 4 2 11 1 0 (100) (0) (21) (50) (29) (14) (79) (7) (0) Peninsular 0 18 0 2 16 0 0 15 3 Malaysia (0) (100) (0) (11) (89) (0) (0) (83) (17) # Chi Square=14.31, df=2, p=0.001, α=0.025 * Chi Square=36.38, df=2, p<0.001, α=0.025 (‘Sparse’ and ‘None’ combined)
Table 3.4: Analysis of indicator matrix using single correspondence analysis of thirty two females.
Cumulative Component Inertia Percentage Percentage Histogram 1 0.8434 48.19 48.19 ****************************** 2 0.3224 18.42 66.62 *********** 3 0.2529 14.45 81.07 ******** 4 0.2013 11.50 92.57 ******* 5 0.1175 6.71 99.29 **** 6 0.0125 0.71 1.00 Total 1.75
University of Malaya
58
4.0 Component 2 Inertia = 0.3224 Percentage = 18.42 3.0 UnFwSp4Gr_None UpFwV7Gr_None FT13 2.0
FT2 FT15 1.0 FT8 UnFwSp4Gr_ Sumatra BrokenPatch Peninsular Malaysia GColour_Brown UpFwV7Gr_Dense GColour_Black 0.0 UnFwSp4Gr_ -1.5 -1.0 FT-B -0.5 0.0 FM-C 0.5 FM-A 1.0 1.5 FM-B DensePatch FT-A UpFwV7Gr_Sparse Component 1 UnFwSp4Gr_Sparse -1.0 Inertia = 0.8434 Percentage = 48.19
-2.0 Characters Sumatran specimens Peninsular specimens -3.0
Figure 3.4: Two-dimensional plot of correspondence analysis components 1 and 2 for thirty two females. Specimens with identical coordinates are grouped together (eg. FT- A = female, Sumatra, group A; FM-A = female, Peninsular Malaysia, group A). The list of grouping is shown in Appendix A.
3.4.2 Male
Table 3.5 summarises the measurement data for the male birdwings from the
Peninsula and Sumatra. Hotelling’s T-Square test showed a significant difference
between Peninsular and Sumatran males in at least one character (T-Square=0.532,
F=25.656, df=4,193, p<0.001). All the characters differed significantly between Universitypopulations based on the simultaneous confidence of intervals.Malaya On average, the forewing length of Peninsular males was about 0.5 cm longer than in Sumatran males, but the
range overlapped widely. Peninsular Malaysian males also had on average greater green
scaling ratios in space 2, 4 and 5 of the hindwing upperside than Sumatran males. The
differences were very slight in spaces 2 and 5. In space 5, the range for the Sumatran
males encompassing the entire range of the Peninsula males while in space 2, the ranges
59
of both populations overlapped greatly. However, the difference between populations
was greater in space 4, with partly overlapping instead of fully encompassing ranges.
The discriminant functions for the Peninsular and Sumatran males based on the ratio
of green scaling in space 4 of the hindwing upperside, which differed the most between
populations, are given below. The percentage of correct classification was 78.8% with a
squared distance between groups of 2.77.
Male, Peninsular Malaysia = –10.3 – 42.0×log(UpHwSp4GrRatio)
Male, Sumatra = –19.3 – 57.4×log(UpHwSp4GrRatio)
Table 3.5: Measurements and ratios for Sumatran and Peninsular Malaysian males (mean ± standard error, and range).
Character abbreviation Sumatra (n=177) Peninsular Malaysia (n=21) FwL 7.4±0.02 * 7.9±0.1 * (6.5–8.4) (6.8–8.4) UpHwSp4GrRatio 0.22±0.00 * 0.33±0.01 * (0.09–0.36) (0.22–0.44) UpHwSp2GrRatio 0.64±0.00* 0.67±0.01 * (0.55–0.76) (0.61–0.78) UpHwSp5GrRatio 0.61±0.00 * 0.66±0.01 * (0.49–0.74) (0.59–0.71) * Significantly different within characters (Hotelling T-square test, α = 0.0125)
Data for the categorical characters is summarised in Table 3.6. There was only University28.02% of the inertia is represented along theof Component Malaya 1 and 21.21% for Component 2 (Table 3.7). Figure 3.5 shows the first two components of correspondence analysis.
Peninsular Malaysian and Sumatran males were more scattered in the graph (Figure 3.5)
compared to females (Figure 3.4), but males from the Peninsula were still located on the
right and males from Sumatra on the left. Sumatran males were most associated with
little or no green scales in space 7 of the hindwing upperside (UpHwSp7Gr_Sparse and
60
_None) and with a straight green disc on the hindwing upperside
(UpHwGrDisc_Straight). Peninsular males were most associated with a convex and
indented green hindwing disc (UpHwGrDisc_Convex and _Indent) and with dense
green scales in space 7 of the hindwing upperside (UpHwSp7Gr_Dense), but they
shared these characters to some extent with Sumatran males. Differences between the
populations were significant for both characters (Chi Square test, Table 3.6). A large
proportion of Peninsular males had either a convex distal margin of the upperside
hindwing green disc or an indented margin, but only 9% had a straight margin. In
contrast, about 57% of Sumatran males had straight margin. Among Peninsular males
77% had dense scaling in space 7 of the hindwing upperside and only a small proportion
had sparse scaling. Sumatran males, however, ranged from having no green scaling to
having dense scaling.
Table 3.6: Numbers and percentages (in parentheses) of males with each character from one hundred eighty Sumatran and twenty two Peninsular Malaysia males.
Characters (abbreviated) and categories Region UpHwGrDisc # UpHwSp7Gr * Convex Indented Straight None Sparse Dense Sumatra 46 32 102 25 89 66 (25) (18) (57) (14) (49) (37) Peninsular 13 7 2 0 5 17 Malaysia (59) (32) (9) (0) (23) (77) # Chi Square=20.370, df=2, p=0.001, α=0.025 * Chi Square=15.838, df=2, p=0.001, α=0.025
Table 3.7: Analysis of indicator matrix using single correspondence analysis of two hundred two males. UniversityComponent Inertia Percentage Cumulative of Histogram Malaya Percentage 1 0.4670 28.02 28.02 ****************************** 2 0.3535 21.21 49.23 ********************** 3 0.3396 20.37 69.60 ********************* 4 0.3078 18.47 88.07 ******************* 5 0.1988 11.93 1.00 ************ Total 1.6667
61
1.5 Component 2, Inertia = 0.3535, Percentage = 21.21 Characters MM-C MT-C UpHwGrDisc_Convex Sumatran specimens 1.0 Peninsular specimens MT-G UpHwSp7Gr_Sparse 0.5 MM-A MT-D MM20 Peninsula MT-A UpHwSp7Gr_None 0.0 -1.0 -0.5Sumatra 0.0 0.5 1.0 1.5 2.0 2.5 3.0 MT-F MM5 MT-E Component 1, Inertia = 0.467 UpHwGrDisc_Straight MM8 Percentage = 28.02 MT-B-0.5 MT-I UpHwSp7Gr_Dense MM-B -1.0 MT-H UpHwGrDisc_Indent
-1.5 Figure 3.5: Two-dimensional plot of correspondence analysis components 1 and 2 for two hundred two males. Specimens with identical coordinates are grouped together (eg. MT-A = male, Sumatra, group A; MM-A = male, Peninsular Malaysia, group A). The list of grouping is shown in Appendix B.
3.4.3 Determining Subspecies and Posterior Probabilities
The discriminant functions generated in Sections 3.4.1 and 3.4.2 can be used to
classify specimens of uncertain origin by calculating the posterior probability. For
example, we could ascertain with a probability of 0.77 based on the green scaling ratio
in space 4 on the upperside hindwing character alone that a female labelled
BMNH(E)1420530 in the BMNH, which lacks locality information, is from Sumatra. In
order to get the probability, a log of one half is added to the discriminant function
because of the unknown relative abundance of population. The green scales ratio on
Universityspace 4 upperside hindwing of this specimenof is Malaya 0.43, this figure was put into
discriminant function of Sumatra and Peninsular Malaysia as below. The specimen will
be classified to the highest discriminant function, which is Sumatra.
Female, Peninsular Malaysia = −16.88 – 118.22 × log(0.43) + log0.5
= 26.15
62
Female, Sumatra = −31.48 − 161.42 × log(0.43) + log0.5
= 27.38
After the classification has been done, the posterior probability of the classification
can be calculated firstly by exponent the both functions, then calculate the probability in
usual way as in Table 3.8. Therefore, the probability of BMNH(E)1420530 classified as
Sumatra population was 0.77.
Table 3.8: The posterior probability calculation for specimen FM1.
Region Discriminant Exponent of Probability function discriminant function Peninsular 26.15 2.2748E+11 =(2.2748E+11)/[(2.2748E+11)+( Malaysia 7.81509E+11)]
=0.23 Sumatra 27.38 7.81509E+11 =0.0128/(1.3476+0.0128)
=0.77
3.4.4 Diagnoses of Subspecies
The females of both subspecies mollumar and trogon are differentiable from other
subspecies by the very much reduced subapical white maculation on the forewing and
by more extensive green scaling on the upperside of the forewing and hindwing. The
upperside forewing usually has at least five green arrow shaped markings, the sixth
arrow shaped marking sometimes reduced to just streaks above and below vein six, and
the seventh reduced to a streak below vein seven, or sometimes non-existent. The males
Universityof these two subspecies are less readily differentiable of Malayafrom other subspecies.
63
The following diagnoses separate mollumar and trogon.
Trogonoptera brookiana mollumar, D’Abrera, 1976
Original description: D’Abrera, B. in D'Abrera et al., 1976: 22–23, genitalia figured, p.
24. Holotype ♂: West Malaysia, Johore, Ulu Sedili. 17th February 1974. V. Doggett.
BMNH(E)1420520. In BMNH. Examined. Paratype (allotype ♀): West Malaysia,
Johore, Ulu Sedili. 29th August 1971. N. Parker. BMNH(E)1420526. In BMNH.
Examined.
Trogonoptera brookiana trogon (Vollenhoeven, 1860)–Eliot, 1959: 372.
Trogonoptera brookiana trogon (Vollenhoeven, 1860)–Eliot, 1972: 175–176.
Trogonoptera brookiana trogon (Vollenhoeven, 1860)–Fleming, 1975: 15; female
figured, Plate 2.
Trogonoptera brookiana trogon (Vollenhoeven, 1860)–Haugum and Low, 1982: 56–60;
genitalia figured, Text-fig. 19B, p. 45, Text-fig. 29D–F, p. 57, Text-fig. 30, p. 58;
distribution, p. 56 (locality form).
Troides (Trogonoptera) brookiana trogon (Vollenhoeven, 1860)–Eliot, J.N. in Corbet
and Pendlebury, 1992: 59–60; checklist, p. 394; female figured, Plate 1 (synonymy).
Trogonoptera brookiana mollumar–D’Abrera, 1982: 4; holotype and allotype figured,
p. 5.
Trogonoptera brookiana mollumar–Tsukada and Nishiyama, 1982: 235–237; male and
female figured, Plate 32, p. 56. UniversityTrogonoptera brookiana mollumar–Matsuka, of 2001: Malaya 210; checklist, p. 347; male and female figured.
Trogonoptera brookiana mollumar stat.rev.–D’Abrera, 2003: 172–174; holotype,
allotype and one male figured, p. 175; genitalia figured, p. 314.
64
Male (Figure 3.6): Upperside hindwing green disc on average more extensive than in
subspecies trogon, with the outer margin usually convex or slightly indented. Green
scales always present in space 7 of the hindwing upperside, and usually dense. Not
reliably distinguishable from subspecies trogon. Female (Figure 3.7): Identifiable by
the wholly black ground colour on the upperside in which the black distal areas of both
wings barely contrast with the veins and with the black forewing cell. Hindwing
upperside with more extensive green scaling than in subspecies trogon, extending to
about half to two thirds of the length of space 4. Underside forewing space 4 with a
broken patch or sometimes entire patch of green scales. Upperside forewing below vein
7 with a green streak, its scaling usually dense.
Range: Southeastern Johore to southernmost of Pahang, and central to southeastern
Terengganu in Peninsular Malaysia.
Materials examined: Males (Appendix C); Females (Appendix D).
University of Malaya
65
Figure 3.6: Male of subspecies mollumar, Holotype deposited at BMNH, UniversityBMNH(E)1420520. Scale bar is equivalent ofto 2 cm. Malaya
66
Figure 3.7: Female of subspecies mollumar, Allotype deposited at BMNH, BMNH(E)1420526. Scale bar is equivalent to 2 cm.
UniversityTrogonoptera brookiana trogon (Vollenhoeven, of 1860) Malaya
Original description: Vollenhoeven, S.C. Snellen van. (1860): 69–70; male figured,
Plate. 6. Syntype (2 males): Sumatra. 1859. Ludeking. Deposited in Naturalis.
Examined.
67
Ornithoptera brookeana (Wallace, 1855)–Distant, 1882–1886: 330–332.
Ornithoptera brookeana (Wallace, 1855)–Fickert, 1889: 749–753; female figured, Plate
XXI, Figure 8.
Ornithoptera brookeana (Wallace, 1855)–Hagen, 1894: 18.
Papilio brookeanus (Wallace, 1855)–C. & R. Felder, 1864: 292 and 334.
Papilio brookeana (Wallace, 1855)–Snellen and Snelleman, 1892: 6 and 24.
Papilio brookiana trogon (Vollenhoeven, 1860)–Jordan, 1908: 17–18; female figured,
Plate 7c (instate trogon as subspecies).
Troides brookianus brookianus (Wallace, 1855)–Rothschild, 1895: 198–199.
Trogonoptera brookeana (Wallace, 1855)–Rippon, 1906: 2–4, 6B.
Trogonoptera brookeana (Wallace, 1855)–Kirby, 1896: 259–263.
Trogonoptera brookiana trogon (Vollenhoeven, 1860)–Tsukada and Nishiyama, 1982:
235–237; male and female figured, Plate 32, p. 56.
Trogonoptera brookiana trogon (Vollenhoeven, 1860)–D’Abrera, 1982: 4; male
figured, p. 5.
UniversityTrogonoptera brookiana trogon (Vollenhoeven, of 1860)–Haugum Malaya and Low, 1982: 51–56; caterpillar figured, Text-fig. 18, p. 42; male genitalia figured, Text-fig. 19D, p. 45, Text-
fig. 27, p. 54, Text-fig. 29A–C, p. 57, Text-fig. 30A and C, p. 58; distribution map
figured, Text-fig. 25, p. 51.
68
Trogonoptera brookiana trogon–Matsuka, 2001: 210; checklist, p. 347; male and
female figured.
Trogonoptera brookiana trogon (Vollenhoeven, 1860)–D’Abrera, 2003: 170; male and
female figured, p. 171; genitalia figured, p. 314.
Male (Figure 3.8): Green disc on the upperside hindwing on average less extensive
than in subspecies mollumar, with the outer margin variable, but straight in about half
of males. Green scaling in space 7 on the hindwing upperside variable, not always
present, often sparse. Not reliably distinguishable from subspecies mollumar. Female
(Figure 3.9): Identifiable by the brown to dark brown ground colour on the upperside.
All the distal areas of the dark ground colour are brown, contrasting clearly with the
blacker base of the forewing cell and the blackened veins, especially at the cell end and
the wing apex. Hindwing upperside with less extensive green scaling than in subspecies
mollumar, extending to about one third to half the length of space 4. Underside
forewing space 4 usually with just sparse green scaling and sometimes without green
scales, but occasionally with a broken patch of scales. Upperside forewing below vein 7
without green scaling or, more commonly, with sparse scaling, and sometimes with a
densely-scaled green streak.
Range: Sumatra Island, in the foothills, especially in the north and central regions,
excluding small neighbouring islands.
UniversityMaterials examined: Males (Appendix C);of Females Malaya (Appendix D).
69
Figure 3.8: Male of subspecies trogon, deposited at BMNH, BMNH(E)1420534. Scale bar is equivalent to 2 cm. University of Malaya
70
Figure 3.9: Female of subspecies trogon, deposited at BMNH, BMNH(E)1420526. Scale bar is equivalent to 2 cm.
University of Malaya
3.5 Discussion
Females of the Peninsular and Sumatran populations were readily and consistently
differentiated by the ground colour, which was paler and browner in Sumatran
specimens. In addition, the extent of the green scaling in space 4 of the forewing
71
underside was almost always a reliable distinguishing character. Several other
characters were useful, though not as reliable in differentiating between the subspecies.
Both discriminant function analysis and correspondence analysis showed that the
populations differed morphologically in these character traits.
Males from of the two populations could not be reliably distinguished. However,
there were some morphological differences; in particular, the extent of the green
markings on the upperside of the hindwing and the shape of the discal margin of this
green hindwing patch. The lack of consistency in the morphological differences could
be because of the high discrepancy for the number of specimens with higher sample size
(182 males) for Sumatran population, in contrast to only 22 males for Peninsular
Malaysia population. It is likely that larger sample size has the potential to uncover
variation.
Differences between females of the various subspecies of brookiana are generally
slight, the most extreme being subspecies albescens, in which the female has the most
extensive white markings of all the subspecies. Ground colour differences are also an
important distinguishing character between subspecies albescens and brookiana, which
occur in the central west of the peninsula and on the island of Borneo, respectively. The
latter is a subspecies somewhat intermediate between trogon and albescens. When it
comes to differentiating males of subspecies trogon, albescens and brookiana, the Universitydifficulties are similar to that of separating malesof of trogonMalaya and mollumar. Ecological and behavioural characters, as well as characteristics of the early stages,
can be useful in differentiating taxa (Parsons, 1996). The males of subspecies trogon in
Sumatra have been reported to regularly visit a hot sulphur spring (de Nicéville &
Martin, 1895). However, they do not congregate in numbers (Haugum & Low, 1982).
The behaviour described of trogon by de Nicéville and Martin has not been reported in
72
mollumar and has not been seen in the extensive observations. However, mollumar was
observed briefly landing on wet ground occasionally. The life history of the subspecies
trogon was conducted by Straatman and Nieuwenhuis (1961) and Dahelmi et al. (2008).
The sex ratio recorded was slightly different in both studies, 2 males to 1 female was
found in Straatman and Nieuwenhuis (1961) and 5 males to 4 females was recorded in
Dahelmi et al. (2008). The life history of subspecies mollumar was not reported so far
but from some observation by Eliot (1941), he stated that the sexes are in about equal
proportions, although at first more females than males were taken. If the life history of
the subspecies mollumar is recorded in future, comparison can be made with subspecies
trogon and it can probably contribute some information to the taxonomy of both
subspecies.
Considering the morphological differences between both sexes of the two
populations, and the comparative differences between other populations already
recognised as subspecies, we agree that the eastern peninsular population is a valid
subspecies, mollumar D’Abrera, distinct from the Sumatran subspecies trogon. This
conclusion is also supported by differences in behaviour and a geographical separation
of the two subspecies.
The occurrence of two subspecies butterflies in the Peninsular Malaysia is unusual.
Subspecies trogon and mollumar were separated by the Strait of Malacca and mollumar
was separated from a highly territorial albescens by the main range in Peninsular UniversityMalaysia. The subspecies mollumar was of only occupied Malaya comparatively small area in southeastern part of Johor and Pahang and eastern part of Terengganu. Possibility raised
by Eliot (1972) was that the trogon has only recently reached the Peninsular Malaysia
by immigration across the Straits of Malacca and has not entered the territory of
albescens in the central mountains. This idea was also agreed by Haugum and Low
73
(1982). However, the subspecies were not reported in western part of Johor which is
nearer to Straits of Malacca and Sumatra. One possibility would be the lack of host
plant in that area.
The subspecies brookiana in Borneo, albescens and mollumar in Peninsular
Malaysia and trogon in Sumatra lay on the Sundaland comprises all the areas above and
also Java. Sundaland is a terrestrial extension of Southeast Asia mainland which the
seabed was exposed during low sea levels (Lohman et al., 2011). The falling and rising
sea levels happened in the Pleistocene geological epoch (0.0117–2.58 Ma), it caused the
Sunda Shelf to fuse and split repeatedly and therefore the sea floor was then exposed
and submerged (Lohman et al., 2011). This was the time where the terrestrial disperse
between islands when the sea floor exposed and then isolated when the sea floor
submerged again (Lohman et al., 2011). Perhaps, the species brookiana in Sundaland
during the sea floor exposed was then isolated into different areas (Borneo, Peninsular
Malaysia and Sumatra now) when the sea floor submerged. Each brookiana population
in different areas evolved and become different phenotype that we found today. The
divergence between the Trogonoptera brookiana and trojana happened in the
Pleistocene and probably due to sea level changes (Condamine et al., 2015). Therefore
the subspecies divergence maybe happens shortly after that.
Haugum and Low (1982) described two male’s forms under subspecies trogon
namely walshi and walshoides. The male’s form walshi has semi-translucent and dull Universitybrown areas which replace the normal green of and black Malaya scaling neatly. The brown scales are broadly bordering all veins of the forewing cell and the bases of the adjacent veins
but not up to cell’s apex. The brown areas have an outer marginal border of a weakly
iridescent blue (Haugum & Low, 1982). The walshoides has similar scales which
spreading from the base of the forewings to approximately two-thirds of the cell. The
74
base of the hindwings is also dominated by the brown scales. The two specimens were
figured in Figure 3.10 and Figure 3.11. The details of the type walshi are “J. Menzel,
Loeboe, Limbata, Padang, Sumatra. Mei 1904. Type”, while type walshoides are “J.
Menzel, Loeboe, Padang, Sumatra. Mei 1904. Type”. The specimens were deposited in
Naturalis, Leiden. There are another five specimens have similar scales with different
however the details of each specimens were not recorded. After examination under light
microscope, it seems that the brown scales were curled up and folded at the edge of
each piece of scale resulted a sharp point on the edge of each scale (Figure 3.12). These
scales might be caused by reactions with chemicals (De Vos, personal communication,
circa October, 2013). Therefore, form names given may not appropriate for the
specimens.
Figure 3.10: The Type specimens of Trogonoptera brookiana trogon ♂ form walshi. UniversityDeposited in Naturalis. of Malaya
75
Figure 3.11: The Type specimens of Trogonoptera brookiana trogon ♂ form walshoides. Deposited in Naturalis.
Curled scales
Normal scales
University of Malaya Figure 3.12: The normal scales and curled scales on the type specimens of Trogonoptera brookiana trogon ♂ form walshoides.
76
3.6 Conclusion
Based to the consistency of certain characters and the specimens available in this
study, we agree with D’Abrera (2003) that the eastern peninsular population is a valid
subspecies, mollumar D’Abrera. The geographical barrier between mollumar and
trogon are also quite far and it is doubt that they will able to meet each other and
interbreed in nature. A total of 10 subspecies were recorded in Sundaland so far and it is
uncommon that such a high number of subspecies occurs in a region, it is ideal to make
a thorough comparison of all the subspecies in details.
University of Malaya
77
CHAPTER 4: BIOGEOGRAPHICAL DISTRIBUTION OF THE
TROGONOPTERA BROOKIANA IN PENINSULAR MALAYSIA
4.1 Introduction
There are two distinct populations of Trogonoptera brookiana (Wallace, 1855) in
Peninsular Malaysia and are considered to be separate subspecies, i.e., T. b. albescens
Rothschild, 1895 and T. b mollumar D’Abrera, 1976. A number of literatures have
described the distribution of these species (e.g. Haugum & Low, 1982; Corbet &
Pendlebury, 1992; D’Abrera, 2003; Ek-Amnuay, 2012; Kirton, 2014) however, these
information were very general. Moreover, for the reported distribution, the species
range may have changed due to the rapid development that has converted large forested
areas into agriculture plantation and human habitation. There is also a general
perception that numbers of these birdwings have declined over the years (Phon &
Kirton, 2009). But yet the changes are unknown. Mapping of locations where these
butterflies were found in the past and current distribution is critical in conservation
work. Thus this study objective is to study the distribution changes from historical and
present records and to identify the range limits for the present distribution.
4.2 Literature Review
The general distribution of Trogonoptera brookiana albescens and mollumar is well
documented although there are some differences to the range mentioned in a few
literatures. The albescens confines to the central states of Perak, Pahang and Selangor Universityand subspecies mollumar occurs in swampy of forest landMalaya in Johor, north-eastern Pahang and Terengganu (Corbet & Pendlebury, 1992). D’Abrera (2003) mentioned that
albescens ranges from the eastern central of Peninsular Malaysia to the Thai border. He
stated that the birdwing is absent from Malaysian islands except for Penang Island. The
northern most record was reported to be Mawlamyine in Burma (Haugum & Low,
1982). Both D’Abrera (2003) and Haugum and Low (1982) agreed that mollumar was
78
only found in the Johor state. Ek-Amnuay (2012) stated the distribution of albescens
was confined to the central states of Perak, Pahang, and Selangor to Yala while Kirton
(2014) mentioned the subspecies albescens occured in the western central foothills as
well as mountain ranges from Perak to Negeri Sembilan. The mollumar can be found in
the southeast Johor to southeast Pahang and parts of Terengganu (Kirton, 2014).
Occurrence of mollumar in Terengganu was reported by Eliot (1959) and Whitmore,
1972. Most of the albescens distribution mentioned is confined to Titiwangsa Range
only while mollumar confined to Pantai Timur Range in Terengganu and swampy
lowland forest in southeast Pahang and northeast to southeast Johor. There are a total of
eight mountain ranges in Peninsular Malaysia (Figure 4.1). Some of the reported
distributions coincided with the presence of their host plant Aristolochia foveolata Merr.
The natural habitats of this host plant were reported to be in coastal areas of Terengganu
which is slightly towards the south, central east of Johor and around Cameron
Highlands in Pahang (Yao, 2015).
University of Malaya
79
Figure 4.1: A map illustrating the mountain ranges in Peninsular Malaysia. The map was modified from geogtingsatu.blogspot.my/2012/12/unit-6-bentuk-muka-bumi.html.
4.3 Methodology
4.3.1 Collection Records
Specimens preserved in museums, universities, institutes and private collections were
carefully examined to obtain the secondary data for this work. Collections visited were
the Natural History Museum London (BMNH), Naturalis Biodiversity Centre
(Naturalis) which comprises collections from the Dutch Rijksmuseum van Natuurlijke
Historie (RMNH) and Zoological Museum Amsterdam (ZMA), the Entomological
UniversityReference Collection of Forest Research Instituteof Malaysia Malaya (FRIM), Malaysia National
Museum (MNM), Museum of Zoology in University of Malaya (MZUM), Center for
Insect Systematics in University Kebangsaan Malaysia (CIS) and Chong-Arshad
collection. Specimens from some other collections were not examined in hand but the
detailed information was compiled from data labels taken by some contributors, such as
80
personal collections of Dr. Laurence Kirton and Lee Kong Chian Natural History
Museum Singapore (LKCNHM). Data labels of each specimen and field diaries in the
collections mentioned above were examined and the information was entered in a
database. Locality, collector and date of collection were also recorded.
4.3.2 Literature Surveys and Interviews
Surveys of the literature were carried out to compile the occurrence of the birdwings.
Some major Malaysian publications were surveyed but not all volumes and issues could
be traced in libraries. Six libraries were visited namely the Za’ba Library which housed
published and unpublished materials on Malaysiana and the Central Library in
University of Malaya, Library of Malaysian Nature Society (MNS, the publisher of
Malayan Nature Journal), Library of Charles Cowan and Dato’ Henry Barlow, Library
of Forest Research Institute Malaysia (FRIM) and the library located at the Herbarium
of FRIM. The Malaysian publications that were surveyed are the whole series of
Malayan Nature Journal and Malayan Naturalist or Malaysian Naturalist, Nature
Malaysiana, Journal of BioScience or Tropical Life Sciences Research, Journal of
Tropical Biology and Conservation, Journal of Wildlife and Parks, Serangga, Malayan
Scientist, Malaysian Journal of Science and Sains Malaysiana. Expedition reports
published by Forestry Department of Peninsular Malaysia were scrutinised as well. A
literature search in The Biodiversity Heritage Library (www.biodiversitylibrary.org/)
was carried out too. The localities reporting on the presence and absence of birdwings Universityfor all aforementioned references were of recorded Malaya for this data compilation. The information extracted from the literature surveys was locality, date of collection or
sighting, the quantity of the birdwings and behaviour.
81
Interviews through emails were conducted towards identified butterfly enthusiasts
either in Malaysia or neighbouring countries such as Singapore and Thailand. A set of
questions were asked, such as:
Q1: Where the birdwings were sighted. Please provide the location as specific as you
can, you may provide description of the location and name the nearer township.
Q2: When the birdwings were sighted. Please provide at least the date or the year of
sighting.
Q3: The number of birdwings sighted. Please provide an approximate number.
Q4: The behaviour of the birdwings sighted, puddling, or nectaring, or in flight, or
courting.
Besides that, emails were sent to all staff of FRIM who conducted fieldwork, to
acquire their sightings of the birdwings. Most of these staffs were not working on
butterflies or not entomologist therefore, additional information were needed to verify
their sighted subjects. They were to describe the characters and behaviour of the
butterflies they sighted or to send photographs of the butterflies they saw. The
information collected from the interviews was similar to the literature survey, i.e.
comprised of locality, date of collection or sighting, the quantity of the birdwings and
behaviour.
University4.3.3 Active Site Surveys of Malaya
Active site surveys were conducted in different localities to determine the recent
distribution of the butterfly and it covered areas where the birdwing has been reported
and areas that have not been surveyed before. The surveys were conducted from May
2012 to May 2017. In order to select survey sites, all the forest reserves and recreational
82
forests listed in Hamidah et al. (2011) were compiled and plotted on map using DIVA-
GIS. In addition, the website that listed all the waterfalls in Malaysia
(www.waterfallsofmalaysia.com/) and the recreational forests guide book (Jabatan
Perhutanan Semenanjung Malaysia, 2003) were referred in selecting survey sites. The
map was carefully studied to identify reported distribution and the gaps that have no
information. Once the survey sites were identified, Google Earth map was referred to
find out accessible road and the coordinates were transferred to GPS to facilitate site
surveys. During site surveys, if any other accessible road was spotted, it will be used as
well. The reported distribution was visited to investigate the presence of the birdwing
whenever possible. However, the sites might not be exactly the same as with the
reported sites.
Each identified locality was surveyed from March to August each year in at least one
whole sunny day from 0900–1600 hrs. The survey months and hours were chosen base
to the higher population index in those months from monitoring data collected in Ulu
Geroh village (Chapter 6). Additionally, April to July is the peak butterfly season in
Malaysia (Kirton, 2014). Pre-existing trails, logging tracks, waterfalls and river banks
in the forest were visited to check the presence of the birdwing. Whenever going to
targeted site by trunk road, the trunk road surrounded by forest was included in the
surveys as well. Pre-existing trails, waterfalls and river banks were surveyed by normal
walking speed while logging tracks and trunk road were surveyed by a four-wheel Universityvehicle at less than 50km/h speed. When birdwingof wasMalaya sighted, the location was logged by a handheld GPS. Revisits to the specific sites were required for occasions of bad
weather not permitting spotting of birdwings.
83
4.3.4 Georeferencing
Georeference was carried out by translating the textual locality information recorded
from collection records, literature surveys and interviews into latitude and longitude
coordinates. Some specific localities were visited during site surveys and the localities
were logged by a calibrated GPS. Therefore, the coordinates recorded from the GPS
were used instead of georeference the locality. The resources used for georeference the
locality are, the Botanical Gazetteer for Peninsular Malaysia (Hamidah et al., 2011),
Google Earth, Google Maps, GeoNames and topography maps.
Some criteria were followed to georeference the localities and are listed below:
(a) Labels that only bears the state name were excluded, e.g. Central Perak, Malaka.
(b) Data that has specific locality with wrong state, the specific locality was used,
e.g. Perak, Malacca, Goping. Thus Goping was taken as the correct data since the state
should be Perak and not Malacca state.
(c) Some localities had undergone rapid development in recent years but they are
usually next to a forest reserve. In such cases, the coordinate of forest reserve was used.
For example: Klang Gates was georeferenced as Ulu Gombak Forest Reserve.
(d) When road is mentioned, e.g. Tapah–Tanah Rata Road, the coordinate of
approximate half way of the road is used except when the distance was mentioned. UniversityMeasurement was made on Google Earth programmeof Malayaand the coordinates determined.
4.3.5 Mapping
Data from collections, literature surveys and interviews, and active site surveys were
plotted on map using Diva-GIS. Mapping was also undertaken by compiling the
historical and current distribution records from collections, literature surveys, interviews
84
and site surveys. The birdwings occurrence recorded before year 2000 was categorised
as historical while from 2000 onwards to present was categorised as current distribution.
Any records that have no date or year of collection or sighted were excluded from maps
for historical and current distribution comparison. Spatial geographical data such as
states in Malaysia was collected from Conservation and Biodiversity Informatics
Branch of FRIM while elevation and inland water body were downloaded from Diva-
GIS website. The method of mapping was based on Hijmans et al. (2012).
4.4 Results
4.4.1 Collection Records, Literature Surveys and Interviews, and Active Site
Surveys
Data collected from collections, literature surveys and interviews, and active site
surveys is shown in Figure 4.2. Most of the Trogonoptera brookiana albescens
specimens kept in collections was collected from Titiwangsa Range in Perak, Selangor,
Pahang and Negeri Sembilan except one male specimen from Merapoh (MZUM), one
male from Taman Negara (CIS), one female from Jengka Forest Reserve (CIS) and 2
males in Ulu Sepri and Bukit Putus located around Gunung Angsi (Figure 4.2A). The
northernmost limit was located at Tanjung Rambutan in Perak while the southernmost
limit was located at Ulu Sepri in Negeri Sembilan. Kenaboi Forest Reserve was the
southernmost locality on Titiwangsa Range. There is only one specimen was collected
from Botanical Garden University of Malaya (MZUM). There are three specimens in UniversityMZUM labelled as Melaka and collected of around Malaya August and September 1992 by Shamsul and Yus however it cannot be georeferenced because the locality written is too
broad. Most of the T. b. mollumar specimens were collected from southeastern part of
Johor and only four specimens were collected from Endau Rompin State Park located
on the border of Johor and Pahang. The only specimen collected in Terengganu was
collected by Chong around Kampung Jongok Batu near Jengai Forest Reserve (Chong-
85
Arshad collection). In southeastern part of peninsular, the southern most limit of T. b.
mollumar was located in Panti Forest Reserve in Johor and distributed northwards up to
Endau-Rompin State Park in Pahang through Ulu Sedili, Jemaluang and Mersing Forest
Reserves.
Data collected from literature surveys and interviews were more spread out (Figure
4.2B). Appendix E summarises the literatures surveyed in the study. All the interviews
conducted and cited as personal communication here are reliable sources because they
are lepidopterists, butterfly enthusiasts, owners of butterfly farms or researchers. There
are two butterfly enthusiasts’ sighted a number of Trogonoptera brookiana albescens
males flying at Penang Hill (Cheng, personal communication, May 27, 2014; Khew,
personal communication, October 30, 2011) and Penang Botanical Garden (Cheng,
personal communication, May 27, 2014) in 2000s and D’Abrera (2003) listed Penang as
one of the localities of albescens. However, Chin from Penang Butterfly Farm (Entopia)
(personal communication, January 23, 2015) and Raymond Lim from Butterfly Garden
of National Zoo (personal communication, June 16, 2016) mentioned that the birdwing
is not found on that island. The northernmost limit recorded on Jerai Mount located in
Kedah in 2–5 Jun 2005 and 22–24 Oct 2005 (Norela et al., 2006). The team recorded
one specimen in each occasion. The uppermost limit in Perak extended to Sungai Lata
Puteh (Haugum & Low, 1982) while westernmost limit extended to Terung Forest
Reserve in 1990s (Cheng, personal communication, May 27, 2014) on Bintang Range. UniversityTung (2003) mentioned that the birdwing of occurred Malaya in Maxwell Hill however, Cheng (personal communication, May 27, 2014) did not find any birdwing in that area in
1970s and 1980s. Chong also did not find the birdwing in that area in 1980s and 1990s
(personal communication, April 17, 2017). In between the Bintang and Titiwangsa
ranges, a few individual birdwing was observed in Ulu Kenas around 1936–1937
(Berwick, 1952). The occurrence of the birdwing in Selangor and Kuala Lumpur was
86
similar to collections data except more sightings were recorded in Kuala Lumpur
(Haugum & Low, 1982; Gary Ruben, personal communication, September 19, 2011).
Two sightings were recorded by Gary Ruben at Damansara Town Centre and Taman
Tun Dr. Ismail between 2005 and 2011. The southernmost limit occurred at Sungai
Terip in Negeri Sembilan in which a female was sighted (De Worms, 1973). Bowden
(2000) mentioned that the birdwing occurred in Taman Negara however, Batten and
Batten (1970a, b), Tharmalingam (2009), Kirton et al. (1990) and Goh and Day
(personal communication, April 20, 2014) did not find the birdwing in Taman Negara.
More than 40 puddling males are considered as mass puddling. A total of six sites were
reported mass puddling, namely Ulu Geroh Village (more than 100 birdwings; Goh,
personal communication, September 2, 2013), Tapah-Tanah Rata Road (hundreds to
5000 birdwings; Regan, 1989) and Kuala Woh Recreational Forest (more than 100
birdwings; Goh, personal communication, September 2, 2013) in Perak, Sungai Kerling
(about 1000 birdwings; Liew, personal communication, January 14, 2016) and Sungai
Congkak Recretional Forest in Selangor (dozens; Gary Ruben, personal communication,
September 19, 2011) and Kenaboi Forest Reserve (about 40 birdwings; Pan, personal
communication, October 27, 2011).
University of Malaya
87
Figure 4.2: Distribution records of Trogonoptera brookiana albescens and T. b. mollumar compiled from (A) collections, (B) literature surveys and interviews and (C) Universityactive site surveys. For all maps, the name ofof states is Malayadenoted in (A).
Trogonoptera brookiana mollumar was reported to occur in Terengganu from
literature surveys and interviews (Figure 4.2B). The northernmost limit occurred in
Sungai Kerbat at Kenyir Lake (Whitmore, 1972). It was also recorded from Rimba
88
Bandar Bukit Bauk which appears to be the easternmost limit of the birdwing in
Terengganu (Norela et al., 2008). They collected two specimens in three-day sampling.
While another four records were from Pengkalan Kajang, Ulu Kemaman and a site
between Sungai Putih and Sungai Kemaman (Whitmore, 1972) and Jerangau-Jabor
Road (Berwick, 1952). Endau Rompin State Park located between Pahang and Johor
(Tan et al., 1990; Goh, personal communication, September 2, 2013; Kirton & Kirton,
1987; Malaysian Nature Society, 1994; Sofian-Azirun et al., 2005), Panti Forest
Reserve together with Lombong Road (Khew, personal communication, October 30,
2011; Haugum & Low, 1982), Ulu Sedili Forest Reserve (Stubbs, 1950; Haugum &
Low, 1982) were the usual localities of this birdwing. The birdwing was also reported to
occur in Kluang Forest Reserve (Stubbs, 1950), around Padang Pakloh in Mersing
Forest Reserve (Haugum & Low, 1982) and along Sungai Lenggor crosses Mersing
Road (Stubbs, 1950).
The southernmost limit of Trogonoptera brookiana albescens in active site surveys
appeared to be at Ulu Bendul Forest Reserve while the northernmost limit appeared at
Ulu Kinta Forest Reserve (Figure 4.2C). Most of the albescens were sighted from usual
localities and the number of birdwings sighted varied from one individual to more than
200 individual. The sites with more than 5 sightings were Kenaboi Forest Reserve in
Negeri Sembilan, Sungai Congkak Recreational Forest, Sungai Tua Recreational Forest,
Kedondong Waterfall, Sungai Kerling Waterfall, Kalumpang and Lubuk Lawah Fall at UniversitySungai Bernam in Selangor and Tapah TNB of Tower atMalaya 7th mile, Kuala Woh Recreational Forest, Ulu Geroh Village, Ulu Geruntum Village, Chellick Fall and Ulu Kinta
Recreational Forest in Perak. However, the birdwing was not sighted from some
localities reported from literature surveys and interviews, such as Jerai Mount in Kedah,
Terung Forest Reserve, Sungai Lata Putih and Ulu Kenas in Perak and Taman Negara in
89
Pahang. There are two sites have mass puddling activities of the birdwing, namely Ulu
Geroh Village and Kuala Woh Recreational Forest in Perak.
Only one individual of Trogonoptera brookiana mollumar (Figure 4.2C) was
sighted at Sungai Nipah Forest Reserve in Terengganu. The birdwing was recorded
from Gunung Arong Waterfall which is near to the coastal area, Panti Forest Reserve,
Ulu Sedili Forest Reserve, Endau Rompin State Park, and Mersing Forest Reserve in
Johor. Most of the sites had more than 3 sightings except in Gunung Arong Waterfall
had only one sighting. More birdwings were sighted at Endau Rompin State Park
compared to other sites.
4.4.2 Comparison between Historical and Present Distributions
The occurrence of Trogonoptera brookiana albescens on Bintang Range and
Keledang Range (Figure 4.1) before 2000 (Figure 4.3A )was not found in present
distribution (Figure 4.3B). However, presently the birdwing was found on Penang
Island and Mount Jerai. The birdwing distribution on Titiwangsa Range was very
similar between historical and present distributions and most of the records were
concentrated on the western side of the range. On the other hand, not many specimens
were collected from Taman Negara, there were only two specimens collected in 1990s
(MZUM and CIS). A female was collected from Jengka Forest Reserve in 2004 (CIS)
which is a new locality and further from the Titiwangsa Range in Pahang. In Perak, the
albescens was not found in southern side. During active site surveys, the number of UniversityTrogonoptera brookiana observed at each of locality onMalaya the western side of Titiwangsa Range was greatly more than the eastern side. The southernmost limit of the birdwing
was very close from the past to present. The mass puddling at Sungai Kerling reported
by Liew (personal communication, January 14, 2016.) and Sungai Congkak
Recreational Forest reported by Gary Ruben (personal communication, September 19,
90
2011) reported around 1979–1980 and 1998 respectively was not found during the
active site survey. However, there are two new mass puddling spots identified namely
Ulu Geroh Village (from active site survey; Goh, personal communication, September
2, 2013) and Kenaboi Forest Reserve (Pan, personal communication, October 27, 2011).
Trogonoptera brookiana mollumar was not found in reported Sungai Kerbat and
Sekayu Recreational Forest during active site surveys but it was recently found in Pasir
Raja Forest Reserve (Zaidi et al., 2006) and Bukit Bauk Urban Forest (Norela et al.,
2008). During active site survey, the birdwing was found in Sungai Nipah Forest
Reserve only in which it is very near to the historical distribution in that area. The
birdwing found in central Johor in the past was not found recently however it was found
flying singly near the northeastern coastal site at Mount Arong Waterfall.
University of Malaya
91
Figure 4.3: (A) Historical and (B) Present Distributions of the Trogonoptera brookiana albescens and T.b. mollumar. For both maps, the name of states is denoted in (A).
4.5 Discussion
Most of the sightings of Trogonoptera brookiana albescens and a few of mollumar
were associated with riverbanks. At Ulu Geroh village, where the mass puddling is Universitysituated, a few males were seen wandering ofin the nearby Malaya oil palm plantation. Based on interviews and active surveys, the number of birdwing sightings has greatly
reduced. For example, Gary Ruben (personal communication, September 19, 2011)
mentioned that he saw dozens of T. b. albescens in Sungai Congkak Recreational Forest
in 1998 but only 1 or 2 in 2011. Liew (personal communication, January 14, 2016) saw
about 1,000 birdwings puddling on the old timber track behind Kerling town in
92
Selangor around 1979 or 1980. Some tracks behind Kerling town were surveyed during
active site surveys twice but did not find large puddling groups in the area. The highest
number seen was only about 20 birdwings puddling. The scenario of seeing 5,000
birdwings flying at a waterfall around Tapah-Tanah Rata road by Regan (1989), is a
historical and unique case whereby it was never reported again. The puddling activity
reported by Pan (personal communication, October 27, 2011) is not permanent in
Kenaboi Forest Reserve because while the monitoring work was conducted in that area
(refer Chapter 6), the puddling occurred two times between July 2012 and December
2014.
Despite an extensive survey, T. b. albescens was not recorded from the Bintang and
Keledang ranges, or mollumar from central Johor and northern Terengganu, where they
had previously been recorded. Although there are a few new localities recorded from
literatures and active surveys (e.g. Mount Jerai, Mount Arong Waterfall, Bukit Bauk
Urban Forest) but the number of specimens or sightings were very low and it could be a
wandering individual and might not able to survive long in the wild.
The lower number of sightings could be due to forest loss, forest fragmentation and
collecting pressure. Forest area in Peninsular Malaysia reduced from 8,012,000 ha in
1970 to 5,864,000 ha in 2010 (Miyamoto et al., 2014). The sightings of mollumar
around Kenyir Lake were recorded before the construction of the lake (Whitmore, 1972)
and 1990s, a few years after construction (Cheng, personal communication, May 27, University2014). The construction started between 1978of and 1985Malaya (Kenyir Lake, 2016). Between 1987 and 2005, the total reported export trade for this species in Malaysia was 22,091
individuals, with most specimens collected from the wild (UNEP-WCMC, 2007).
Between 2001 and 2010, the total reported export trade of wild specimens was 5,060
individuals (UNEP-WCMC, 2012).
93
Zonneveld et al. (2003) noted that more than one survey year is needed to determine
significant extirpation especially for species that has high dispersal ability. Some
butterfly enthusiasts in Peninsular Malaysia repeatedly visit the reported localities from
time to time and noticed the shrunk in number of birdwings. In order to capture the
changes of distribution or the number of occurrence, such information is very important.
The active surveys conducted were extensive and it is good if it can be repeated in
future.
Biases that influence perceived distribution of individual species should be taken
care of when interpreting distribution map (Dennis & Thomas, 2000) such as the
number of visits or recorder efforts influenced species’ richness and species abundance
(Dennis et al., 1999). In this study, collections were found mainly from central Perak,
Selangor, small part of Negeri Sembilan, Johor and a few sites in Pahang only. Whether
the birdwing occurred in other parts of peninsular in the past is still unknown. However,
a few data collected from literature surveys and interviews from those other parts were
found to be in absence of birdwing. Although the data collected from collections,
literature surveys and interviews is sporadic in terms of temporally and spatially but it
showed the changes especially in the mountain range. The national and regional status
and the changes in the range of Eurodryas aurinia (marsh fritillary) was captured from
collections data too (Warren, 1990 in Harding et al., 1995).
There are discrepancies on the presence and absence of the birdwings in some Universitylocalities such as Maxwell Hill, Penang of Island, TamanMalaya Negara and Melaka. The presence of the birdwings on Maxwell Hill was reported by Tung (2003), on the
contrary the birdwing was not found by Stubbs (1950), Cheng (personal
communication, May 27, 2014) and Chong (personal communication, April 17, 2017).
The birdwing was not found during active site surveys. While Khew (personal
94
communication, April 22, 2014), Cheng (personal communication, May 27, 2014) and
D’Abrera (2003) reported the occurrence of albescens on Penang Island. Raymond Lim
(personal communication, June 16, 2016) and Chin (personal communication, January
23, 2015) did not find the birdwing in nature on the island. Raymond Lim gave his
opinion that if there are escapees from Penang Butterfly Farm, they may not able to
survive in the wild because the host plant Aristolochia foveolata is not found on the
island.
The specimens collected at Merapoh in 1992 (MZUM) and Taman Negara in 1999
(CIS), the sighting by Nor Azian (personal communication, March 28, 2012) and
Bowden (2000) are tentatively identified as subspecies albescens. Firstly, it is because
only males were found and it is not possible to differentiate male of albescens and
mollumar. Due to the albescens population is greatly more than mollumar in peninsula,
therefore it is appropriate to put tentatively as albescens. Secondly, based on description
of Nor Azian (personal communication, March 28, 2012), she saw many birdwings
puddling on riverbank of Lata Berkoh. Based on this behaviour, most likely the
birdwing is belongs to albescens. However, there is also a possibility to be a mollumar.
The Taman Negara extended to southern Kenyir Lake and Pantai Timur Range where
mollumar was previously reported occurred in those areas, either historically or
presently. This population in Taman Negara was separated further from Titiwangsa
Range compared to Pantai Timur Range. Therefore, sighting or a female specimen is Universityneeded to confirm to identification. Nonetheless, of thereMalaya were also many other surveys including the active site survey found no birdwing in that area (Batten & Batten, 1970a,
b; Tharmalingam, 2009; Kirton et al., 1990; and Goh & Day, personal communication,
April 20, 2014). The results are very contrast from Nor Azian sighting. Thus, it is worth
to pay more surveys and longer surveys in search of this elusive population.
95
Trogonoptera brookiana was not found in Melaka in surveys conducted by Batchelor
(1959), Norela et al. (2011) and active site surveys from this project. This could be
because, Melaka is located further away from Titiwangsa Range causing the birdwing
unable to currently penetrate that area. There are not many forest land in the area and
most likely do not contain host plant. However, it is puzzling that there were three male
specimens from Melaka found in MZUM’s collection. Therefore, more surveys should
be conducted to confirm the existence of a small population in the area.
The record of albescens in Mount Jerai (Norela et al., 2006) and a female albescens
collected from Jengka in 2004 (CIS) are rather unexpected when contrasting with
historical records. The active site surveys at Mount Jerai and nearby Jengka resulted in
no sighting or record. It could be a vagrant and its sustainability is doubtful since
Jengka is separated from Titiwangsa Range by Benom Range and has many plantations.
Whereas, the nearest historical locality to Mount Jerai is Sungai Lata Puteh on Bintang
Range which is about 80 km apart and the sites are separated by developments,
plantations and small patches of forest.
Kelantan received very less attention from butterfly enthusiasts and collectors thus
not many information can be found on that state. The only survey conducted was at
Gunung Stong Tengah Forest Reserve in 24–26 May and 23–25 October 2003 by Zaidi
et al. (2005a). During active site surveys, due to many areas were difficult to access and Universitytherefore cannot be reached. Many logging ofactivities wereMalaya observed during the surveys. Conservation status of the host plant Aristolochia foveolata in Peninsular Malaysia
Malaysia is categorised as near threatened (Yao, 2015). It occurs in primary lowland
forest and hill forest up to 500 m altitude. It has scattered distribution and only occurs
in, central east of Johor, around Cameron Highlands in Pahang and in coastal area of
Terengganu which is slightly towards the south (Yao, 2015). In a plant database
96
provided by Yao (July 21, 2016), four records of foveolata were found in Peninsular
Malaysia and they were collected from Endau-Rompin State Park (between Johor and
Pahang), Mersing (Johor), Bukit Bauk (Terengganu) and Cameron Highlands Road,
Mile 16 (Pahang). The Cameron Highlands Road, Mile 16 record is actually Tapah–
Tanah Rata Road located in Perak after georeferenced the locality. Although the
birdwings are more concentrated in Perak and Selangor however, there is no record of
the host plant in those areas. Identify the location and distribution of the host plant in
these two states is important for conserving both plants and birdwings. And, searching
host plants in the areas with vagrant individual might able to investigate the
survivability of the birdwing.
4.6 Conclusion
In conclusion, distribution data from collections, literatures and interviews of the
Trogonoptera brookiana albescens and mollumar were compiled and mapped for the
first time. Most of the data was georeferenced and the locality was databased.
Additionally, systematic active surveys were conducted and mapped in this study. And
for the first time the historical and present distributions were compared to examine the
distribution changes.
Subspecies albescens occurred mainly on the Titiwangsa Range on the westward side
from Ulu Kinta (central Perak) southwards through Selangor to Ulu Bendul (Negeri
Sembilan) in the hills south of the range. It had a scattered distribution on the eastward Universityside of the range in Pahang where it wasof commoner Malaya in the foothills of the south. Hotspots for mass puddling behaviour of male albescens were identified. Subspecies
mollumar occurred in the lowlands from south eastern Johor to south eastern Pahang
and more rarely in the southern half of eastern Terengganu. Trogonoptera brookiana
albescens was not recorded from the Bintang and Keledang ranges, or mollumar from
97
central Johor and northern Terengganu, where they had previously been recorded.
However, a few new localities were recorded but need further investigation. Based on
interviews and active surveys, the number of birdwing sightings have reduced, possibly
due to forest loss, forest fragmentation and collecting pressure. The study did not
include islands as active site survey, therefore it is suggested this should be included in
future studies. In order to understand their distribution vertically, a systematic study can
be carried out at different elevation in a few localities based to the maps generated from
this study. Besides their occurrence and puddling distribution, their host plants and
preferred nectar plants surveys are important in order to identify their resources
distribution. Additionally, identifying their courting and mating habitat or location are
also essential. This additional information can be overlaid on the distribution map and
would greatly improve conservation programme. Based to the map, conservation can be
initiated and concentrated on the marked areas. Although Trogonoptera brookiana
mollumar was the rarer subspecies and had a narrower range, parts of its range are
within state parks. In contrast, there are no permanently protected forests within the
range of albescens.
University of Malaya
98
CHAPTER 5: BEHAVIOURAL STUDIES OF THE TROGONOPTERA
BROOKIANA ALBESCENS
5.1 Introduction
Behavioural study of the Trogonoptera brookiana is scarce compared to its life
history study. Most of the information related to its behaviour was gathered by
naturalists through general observations or by ad libitum manner except for the study
conducted by Orr (1982) on flight rhythm. Casual descriptions on flight periods, mating
repertoires, nectaring and puddling periods were documented by naturalists. Plant food
resources for the adult were identified whenever the naturalists observed birdwings
feeding on them. In Peninsular Malaysia, the subspecies albescens received more
attention from naturalists, probably because it is easier to be found and the population is
larger compared to another subspecies, mollumar.
Understanding the birdwing’s behaviour is essential to enhance the conservation
programme and also to promote eco-tourism of this iconic insect. This chapter describes
the behaviour of the birdwing focussing on the subspecies albescens which has a good
population size and distribution patterns. The specific objectives for this work are:
1. to determine the flight, nectaring and puddling activity rhythms;
2. to study the plants exploited as food resource by the adult birdwings; and
3. to document the courting and mating behaviours of the birdwing. University5.2 Literature Review of Malaya The males of certain subspecies of Trogonoptera brookiana, such as subspecies
brookiana and albescens are well known for their puddling behaviour which often
occurred in hot springs, on paths or river banks or seepages and urinated spots
(Panchen, 1980; Corbet & Pendlebury, 1992; Igarashi & Fukuda, 1997). Subspecies
albescens started puddling around 0800 hours and dispersed during noon (Igarashi &
99
Fukuda, 1997). They usually avoid direct sunlight and are sensitive to interruption.
Trogonoptera brookiana brookiana only puddle on natural puddling sites and they
never visited site enhanced by artificial nutrients such as albumin and sodium chloride
(Beck et al., 1999).
A systematic study conducted by Orr (1982) found that both males and females of
the T. b. albescens in Tapah and Tanah Rata started flying about 0800 hrs and ended at
1700 hrs with the highest flight activities between 1000–1400 hrs. On the other hand,
Panchen (1980) observed that the males of subspecies brookiana were seen from mid-
morning until about 4.30 pm on almost every sunny day and they were more active near
noon. Subspecies trogon started flying about 0700 hrs early in a day, they soon
disappeared during the hottest hours and appeared again in the late evening around 1700
or 1800 hrs (de Nicéville & Martin, 1895). However, comparisons cannot be made
because the time periods were not consistent.
Nectar plants exploited by subspecies albescens were Bauhinia (Fabaceae),
Crossandra infundibuliformi (Acanthaceae), Duranta lorentzii (Verbenaceae),
Mussaenda (Rubiaceae) and Lantana (Verbenaceae) (Corbet & Pendlebury, 1992; Goh,
1994; Cole, 1998; Hoskins, 2015). The subspecies mollumar on the other hand seeked
for Mussaenda (Rubiaceae) (Corbet & Pendlebury, 1992) while subspecies brookiana
seeked for Ixora (Rubiaceae) (Panchen, 1980). The birdwing imbibed nectar while it
was on wing and fluttered without landing (de Nicéville & Martin, 1895; Cole, 1998). UniversityThe subspecies trogon exploited nectar fromof Lantana Malaya when the weather is misty and wet or even when it rains (Straatman, 1955). The nectaring hours were between 0600 to
0800 hrs and 1700 hrs. Another species, Trogonoptera trojana fed on Bauhinia between
0600 to 1500 hrs and the birdwing was picky because it fed on certain vines only
(Jumalon, 1967).
100
Courting behaviour was observed on subspecies brookiana only at the mountain
region of the headwaters of the River Segama in Sabah (Skertchley, 1889) and Gunung
Mulu National Park in Sarawak (Panchen, 1980). In their studies at River Segama, they
reported that the courting behaviours succeeded with successful matings. What was
more interesting, they reported that females initiated the courting behaviour. This
behaviour was not reported for other subspecies. Trogonoptera brookiana is territorial,
such as in subspecies albescens (Cole, 1998; D’Abrera, 2003) and brookiana (Panchen,
1980). The relationship between territoriality, securing and defending resources such as
food and mates remains unknown.
5.3 Methodology
5.3.1 Puddling Rhythm
The study was conducted in Ulu Geroh village for two to three consecutive days in
three high population months (January to March 2010) and three low population months
(September, November and December 2010) in order to enable comparison can be made
between puddling and birdwings in flight for monitoring study in Section 6.3.1.2. The
details of the study site were given in Section 6.3.1.1. In order to enable counting big
groups of T. b. albescens puddling, photography method was applied in the study. The
puddling sites, equipments and image analysis software used for counting birdwings
from images, were similar to the method described in Sections 6.3.1.1. and 6.3.1.2. In
order to ensure coverage of early episodes for puddling activities, data recording started Universityas early as 0800 and ended at 1900 hrs in ofall months Malaya except in December 2010 which started at 0900 hrs and ended at 1700 hrs due to bad weather. Three photographs were
taken every minute at the start of every hour. There were two permanent puddling spots
in the village and these spots were included in the study and given names as Puddle 1 (a
larger puddle) and Puddle 2 (a smaller puddle). The counts of puddling birdwings in the
three images at each hour were averaged to obtain hourly averages. The averages at
101
each hour in day-1, day-2 and day-3 were averaged to obtain monthly averages for each
hour. In order to standardise birdwing counts across months and to counter the high
standard errors resulted from high and low populations, standardisation based on
standardised exam scores method was applied (Kirton, personal communication, circa
2012). The count at each hour in each month was standardised by subtract from each
month average (0800 to 1900 hrs) and added group average (all months average from
0800 to 1900 hrs). A constant value was added to each standardised average to avoid
negative value, if needed. After that, a rhythmic graph which includes actual counts and
standardised counts with standard error for Puddle 1, Puddle 2 and the total at each hour
was plotted. Temperature and relative humidity readings were taken using a calibrated
data logger, Blue Gizmo BG-DL-01 at each hour during the sampling periods. Soil
temperature and pH were measured using a calibrated thermometer and pH meter at the
two puddles and a seapage site beside a river on 25 February 2010. Measurements were
taken at 0900, 1200 and 1500 hrs. Due to the villagers (public guardians for the sites)
objections and are sensitive towards conducting a proper soil study at the puddles,
therefore, only three readings of temperature and pH on a day were recorded.
5.3.2 Flight Rhythm
5.3.2.1 Ulu Geroh Village
Data collections on the flight and puddling rhythms were done concurrently in Ulu
Geroh. For the flight activity study, a 180x20 m belt transect was made near the Universitypuddling sites and the birdwings flying inof the belt transectMalaya were counted by the same observer walking through once at a slow and steady pace for the first 15–20 minutes of
each hour from 0800–1900 hrs for 2–3 days. Back and forth flights of what was
recognisably the same individual birdwing within or across the belt transect were
counted as a single sighting. The calculations of averages and standardisation, the
rhythmic graph presentation and the analysis were similar to description given for the
102
puddling rhythm in Section 5.3.1. Puddling rhythm and flight rhythm studies in Ulu
Geroh Village were conducted at the same time. Therefore, both studies shared the same
environmental data.
5.3.2.2 Kenaboi Forest Reserve
Transect walk and single spot observation methods were applied to study the flight
rhythm of the T. b. albescens in Kenaboi Forest Reserve at two different areas which
was about 300 meter apart. The study was carried out one day in each month from July
2012 to December 2013, excluding September 2012 due to serious logistic constraint.
Transect belt of 400 x 20 meters along flying path of the birdwings were made beside
the Kenaboi River. The same observer walked within the transect belt twice
representing 2 replicates to record the number of birdwings in flight of for each hour
from 0800 to 1800 hours. As in the Ulu Geroh Village, back and forth flights of what
was recognisably the same individual birdwing circling within or across the belt transect
were counted as a single sighting. The time for the observer to finish recording for each
hour was in the range of 20 to 30 minutes. The birdwing counts in 2 replicates at each
hour were averaged to represent each month count and the data was subjected to the
standardisation as described in Section 5.3.1.
At the same time with simplified transect walk, the total number of birdwings in
flight was observed at one observation point about 300 meters apart from the transect Universitybelt for 30 minutes duration and the number of of birdwings Malaya passed by was recorded. The station was beside an unnamed stream which is joined to the Kenaboi River next to the
belt transect. Similar to the Section 5.3.1, the standardisation was used to avoid high
standard error.
103
Temperature, relative humidity and light intensity were measured using calibrated
datalogger Blue Gizmo BG-DL-01 and lux meter Lutron LX-101. Both devices were
placed at the brightest spot near the transect walk area where there was no obstruction
by trees. The parameters were measured after the transect walk was conducted at each
hour.
5.3.3 Nectaring Rhythm
5.3.3.1 Ulu Geroh Village
The study was conducted during the flowering season from September to December
2011. Before the start of observation for each month, the plants in the village were
surveyed for the occurrence of flowers. If flowers were present then the plant was
included for observation. During those four months, there were three individual plants
of Bauhinia audax (given plant code as SSDSS, ASS and SSDJB), two individual plants
of Bauhinia bidentata (JMPW and TJDD) and five individual plants of Hibiscus rosa-
sinensis (BR1, BR2, BR3, BR4 and BR5) flowered. The identifications of the Bauhinia
plants were confirmed by Botanists in FRIM, Botany and Herbarium Branch from
samples brought back for their expert verification.
The flowering plants were observed by two groups of observers in order to record
observations on all plants in each hour. The observation on bauhinia plants and hibiscus
plants were conducted at the same time. In contrast to hibiscus plants which flowered at
low levels and could be counted directly, the bauhinias flowered at high levels (height Universitybetween 10 to 25 metres), thus observations of were made Malaya using a high powered binocular with 10 times magnification. Observations were made hourly, for 10 minutes per plant
during the day from 0800 to 1800 hours for two to three days about one month apart.
Observations were repeated every hour. The plant sequence observed was always the
same each hour and each month.
104
The observation methods used here were scan and focal samplings. At the start of
observation, the whole plant was scanned to record the number of birdwings found
nectaring on the plants. For the focal sampling, an individual was randomly chosen for
detail observation from the start of observation until it flew away from the plant. If both
sexes appeared on the same plant at the same time, the female was chosen for focal
observation because it is rarely seen in nature. Whenever a birdwing went into the scene
while focal observation was conducted, the birdwing was also concurrently scored for
the total number of visitation population. Therefore, the number of birdwings nectaring
at each hour was the sum of the number of birdwings seen at the start of observation
(scan method) and the number of birdwings seen throughout 10 minutes of observation.
The other two parameters recorded were the number of flowers or flower clumps visited
by each birdwing and the total time spent on each plant by each birdwing. Bauhinia’s
flowers are in clumps, Figure 5.1 describes a flower clump of the plant. During rare
occasions, when the flowers were found to be very dense, it is a challenge to
differentiate between clumps, for such instances, any change of movement by the
birdwing was considered as visiting a different clump. Movement caused by wind or
disturbances by other insects was later confirmed by the flying out behaviour from that
floral visit.
University of Malaya
105
Figure 5.1: Example of the series of floral clumps scored for the studies on birdwing floral visitation behaviour.
In order to determine nectaring rhythm of total number birdwings nectaring at each
hour, the total number of birdwings found in all plants of same species at each hour on
each day was averaged in a month. Then, the daily average of each hour for each day
was averaged to get the monthly average. Monthly average of each month was again
averaged to get the final figure. This procedure was repeated for other plant species.
Similar calculations were applied to determine nectaring rhythm using the number of
flowers or flower clumps visited and total time spent by each birdwing. However,
averages will not include zero which indicate no birdwing was nectaring within the Universityobservation period. Standardisation following of Section Malaya 5.3.1 was applied on the total
number of birdwings only.
5.3.3.2 Kenaboi Forest Reserve
The study was conducted in Kenaboi Forest Reserve from July 2012 to February
2013 excluding September 2012 due to logistic constraint. Due to the wide area which
106
is about 18.4 km long, the long distance between flowering plants and the observers’
safety constraints, the method used in Ulu Geroh Village was slightly modified.
Most of the flowering plants flowered was Bauhinia bidentata var. breviflora. Before
each observation was conducted in each month, the number of flower clumps on each
bauhinia was counted and the plant with the most flowers was chosen for observation.
Two days observation was conducted in each month. If the distance between two
chosen plants is very far which the observers cannot communicate with walkie-talkie
and therefore due to safety constraint, observation was conducted on one plant each day.
On the contrary, observation was conducted on two plants each day. Different plant was
observed each day to include more flowering plants. During 7 months observation, a
total of 14 Bauhinia plants were observed, 6 Bauhinia plants were observed repeatedly
in two different months and 1 Bauhinia plant were observed in three different months.
Besides bauhinia, two Saraca declinata plants were flowering in July 2012 and
February 2013 and five Hibiscus rosa-sinensis plants were flowering in July 2012, and
they were observed as well. The hibiscus plants were severely trimmed after that. Each
plant was observed for 20 minutes and the total number of birdwings nectaring within
that time was recorded. The methods of observation and quantifying bauhinia flower
clumps were similar as in Ulu Geroh Village. Quantifying flower clumps on Saraca
declinata was easier because the clumps were very distinct and well spaced from each
other. All calculations are similar to Section 5.3.3.1.
University5.3.4 Preferred Nectar Plant Studies inof Ulu Geroh Malaya Village and Kenaboi Forest Reserve
The same datasets collected for rhythmic study was used to determine the preferred
nectar plants (Sections 5.3.3.1 to 5.3.3.2) in both studies sites Ulu Geroh Village and
Kenaboi Forest Reserve. However, in Kenaboi Forest Reserve, the study was only
107
conducted in July 2012 and February 2013 in the presence of different flowering plant
species in order to compare birdwings flower visitations. For the study month of July
2012, the flowering plants comprised of: Saraca declinata, Hibiscus rosa-sinensis and
Bauhinia bidentata var. breviflora whereby in February 2013, Saraca declinata and
Bauhinia bidentata var. breviflora plants were flowered.
The number of birdwings nectaring on each plant within a day was summed up and
the sum was divided by the number of flowers or flower clumps on each plant to get the
number of birdwings on each flower. This value of each plant was then totalled up. The
number of birdwings on each flower for each plant was then expressed as a percentage.
The daily percentage for each day in a month was averaged to get monthly average and
then each monthly average was again averaged resulting in the final value.
The total flower visited by a birdwings on each plant for each day was averaged.
Then daily averages for each day were averaged to get the monthly average of each
month and then monthly averages were again averaged. This calculation was applied to
the total time spent on a plant by each birdwing. Zero reading means no birdwing was
nectaring and therefore zero was excluded in the average. The number of flowers or
flower clumps in each month was standardised using the method described in Section
5.3.1. The data sets obtained were presented in graphs. The differences between three
plant species on three parameters were tested using Kruskal-Wallis analysis. University5.3.5 Courting and Mating Incidences of Malaya The focal and ad libitum sampling methods were used to gather the courting and
mating behaviour (Martin & Bateson, 1993). Whenever behavioural interactions
between male and female birdwings were encountered, they were observed via focal
samplings until the pairs were no longer sighted. The sampling durations were non-
specific because these behavioural events were rare in occurrence and thus considered
108
as opportunistic recordings. The date and the time of the incidences were noted and all
behaviours were recorded and described in detail.
5.4 Results
5.4.1 Puddling and Flight Rhythms
5.4.1.1 Ulu Geroh Village
Throughout the study, a total of 4313 puddling males were sighted and recorded.
Figure 5.2 shows the puddling rhythms at Puddle 1, Puddle 2 and the total rhythms for
both. The puddling birdwings count at Puddle 1 increased rapidly from 0800 to 1100 hrs
and gradually reached a peak at 1500 hrs. The number was then decreased gradually in
the early evening and then rapidly from 1700 to 1900 hrs. A slight different rhythm was
observed at Puddle 2 where the birdwing counts gradually increased from the start of
day and reached a peak at 1600 hrs then decreased gradually. When the counts from
Puddle 1 and Puddle 2 were combined, two peaks were observed between 1400 and
1600 hrs. Therefore, the counts from 1400 and 1600 hrs were chosen for monitoring
study in Chapter 6. The temperature on the puddles was higher compared to the river
temperature, which is 29.90±0.06°C at Puddle 1, 28.67±0.32°C at Puddle 2 and
21.87±0.52°C in river (Table 5.1). While the pH at the puddles also slightly alkaline,
8.47±0.17 at Puddle 1 and 8.19±0.17 at Puddle 2. The pH at the river was only
7.03±0.12.
Figure 5.3 shows the flight rhythms of male, female and the total for both. Data Universitystandardisation was conducted for all but,of despite Malaya the standard errors were greatly reduced, the differences between the actual count and standardized count were very
slight in male and the total. Therefore, the graphs of actual count for male and total were
not included. The males in flight increased rapidly from 0800 to 1000 hrs. Between
0800 and 1700 hrs, the number decreased and increased several times and decreased
109
rapidly after 1700 hrs. The females increased rapidly and reached a peak at 1000 hrs,
and then the number was rather stable and slightly increased and reached a peak at 1500
hrs. After that, the number decreased. Overall, the rhythm followed the male’s rhythm
as the number of females in flight was very low. The flight activity from 0900 to 1700
was considered high and constant, therefore the counts from 0900 to 1700 during the
study was used to correlate with the puddling birdwings in Section 6.3.1.2.
The temperature increased gradually reaching the highest point around 1300 hrs and
then decreased gradually to the end of sampling hour (Figure 5.4). However, the relative
humidity dropped from early morning and reached its lowest point at 1300 and 1400 hrs
and increased gradually until evening. The puddling rhythm was positively correlated to
temperature but negatively to relative humidity, the correlations were significant (Table
5.2). The highest puddling counts occurred after three hours of the highest temperature
recorded at 1300 hrs. When the relative humidity decreased during afternoon hour, the
puddling counts increased. The peak puddling count occurred about two hours after the
lowest relative humidity recorded at 1400 hrs. The flight activity did not seem to
correlate with temperature and relative humidity (Table 5.2).
University of Malaya
110
160 Total (Standardised) 140 Total (Actual count) 120 100 80 60 40 20 0 60 Puddle 2 (Standardised) Puddle 2 (Actual count) 40
20
0 120 Puddle 1 (Standardised) 100 Puddle 1 (Actual count)
Monthlycount puddlingbirdwing averageof 80
60
40
20
0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900
Time (hrs) Figure 5.2: Daily rhythms of puddling activities at Puddle 1, Puddle 2 and the total from both puddles at Ulu Geroh village. University of Malaya Table 5.1: Averages temperature and pH readings at Ulu Geroh village, measured on 25 February 2010 at 0900, 1200 and 1500 hrs.
Parameters Puddle 1 Puddle 2 River
Temperature (°C) 29.90±0.06 28.67±0.32 21.87±0.52
pH 8.47±0.17 8.19±0.17 7.03±0.12
111
14 Total (Standardised) 12 10 8 6 4 2 0 3 Female (Actual count) Female (Standardised) 2
1
0 14
Monthly average count of birdwing in flightin countof birdwing Monthly average Male (Standardised) 12 10 8 6 4 2 0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 Time (hrs)
Figure 5.3: Daily flight rhythms of the male, female and the total of birdwings at UniversityUlu Geroh village. of Malaya
112
30.0 28.0 26.0 24.0 22.0 Temperature (°C) Temperature Monthlynaverage ofMonthlynaverage 20.0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 100.0 95.0 90.0 85.0
RH (%) RH 80.0 75.0 Monthly average of Monthly 70.0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900
200 14 Puddling (Actual count) Puddling (Standardised) 180 Flight (Standardised) 12 160
140 10
120 8 100 6 80
60 4 40 2 20 Monthly average count of birdwings in flight birdwingscount of average Monthly Monthly average of puddling birdwings count birdwings puddlingof average Monthly 0 0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 Time (hrs)
Figure 5.4: Daily puddling and flight rhythms and the relation with temperature and relative humidity at Ulu Geroh village. University of Malaya
113
Table 5.2: Correlation matrix of puddling and flight rhythms with relative humidity and temperature at Ulu Geroh village (Pearson correlation value, P value).
Puddling Flight RH
RH (-0.839, 0.001) (-0.561, 0.058)
Temperature (0.939, 0.000) (0.526, 0.079) (-0.905, 0.000)
5.4.1.2 Kenaboi Forest reserve
The female occurrence in Kenaboi was very low therefore the data was summed with
the male occurrence. In the transect walk method, females were present for short
periods: in November 2012, between 1300 and 1400 hrs and in October 2013 at 1200
hrs. While in single spot observation method, females were recorded in the months of
April, July and September 2013 from 1200 to 1500 hrs.
By using single spot observation method, the birdwing counts increased rapidly from
0800 to 1000 hrs and then started to decrease gradually until 1800 hrs (Figure 5.5). The
peak count happened one hour after the highest relative humidity. Despite the peak
difference, the birdwing rhythm followed a similar pattern with relative humidity. A
slight difference was observed in the transect walk method (Figure 5.5). Birdwing
counts increased rapidly from 0800 to 1000 hrs and then slowly reaching a peak at 1200
hrs and subsequently decreased. The rhythm recorded from transect walk method
followed a similar trend with light intensity and the peak occurred at 1200 hrs coincided Universitywith the light density peak. Temperature increasedof slowlyMalaya and reached the peak at 1300 hrs and then decreased slowly until 1800 hrs while light intensity increased rapidly to
reach a peak at 1200 and decreased rapidly too after that. At the beginning, relative
humidity increased rapidly from 0800 to 0900 hrs, peak at 0900 hrs. Then it decreased
gradually from 0900 to 1100 hrs. After that it decreased rapidly until 1300 hrs. Then it
increased again and reached a second peak at 1700 hrs then decreased (Figure 5.5). By
114
averaged the peak hours in transect walk and single spot methods (i.e. from 1000 to
1200 hrs) in each month, the birdwing count using the methods was significantly
correlated (Pearson correlation = 0.836, p = 0.001). The flight rhythm recorded from
transect walk method was correlated with temperature and light intensity while the
rhythm recorded from single spot observation was correlated with light intensity only
(Table 5.3).
University of Malaya
115
35.0
30.0
25.0
20.0 Temperature (°C) Temperature 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 85.0
80.0
75.0
70.0
Relative humidity (%) humidity Relative 65.0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 50000 40000 30000 20000 10000
Light intensity (lux)intensityLight 0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 5 Transect walk (Actual count) 4 Transect walk (Standardised)
3
count 2
1
Monthly average birdwings averageMonthly 0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 40 Single spot (Actual count) 35 Single spot (Standardised) 30 25 20 count 15 University10 of Malaya 5
Monthly average birdwings averageMonthly 0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time (hrs)
Figure 5.5: Daily flight rhythms from single spot observation and transect walk methods at Kenaboi Forest Reserve and the relationship with monthly averages of temperature, relative humidity and light intensity.
116
Table 5.3: Correlation matrix of flight rhythms recorded from transect walk and single spot observation methods at Kenaboi Forest Reserve with temperature, relative humidity and light intensity (Pearson correlation value, P value).
Transect walk Single spot Temperature RH
Temperature (0.749, 0.008) (0.376, 0.254)
RH (-0.332, 0.319) (0.098, 0.774) (-0.795, 0.003)
Light intensity (0.881, 0.000) (0.762, 0.006) (0.57, 0.067) (-0.22, 0.516)
5.4.2 Nectaring Rhythms
5.4.2.1 Ulu Geroh Village
By looking at the two period moving average, the males’s nectaring activity on each
plant species was higher in the morning and afternoon compared to evening (Figure
5.6). The male’s nectaring rhythms on each plant species and on the average of all plant
species were similar where the activity increased in the morning and then gradually
decreased towards the evening. The activity on Bauhinia audax was higher compared to
Bauhinia bidentata var. breviflora and Hibiscus rosa-sinensis.
Females were found nectaring on Bauhinia audax and Bauhinia bidentata var.
breviflora only and the count was higher on Bauhinia audax (Figure 5.7). In contrast to
the male, the female nectaring activity on Bauhinia bidentata var. breviflora was only
recorded at 1400, 1600 and 1700 hrs. The activity on Bauhinia audax increased in the
morning then the activity was rather constant until 1700 hrs and after that the activity
Universitydropped until 1800 hrs. The combined activitiesof onMalaya both plant species, revealed that
nectaring activity increased slowly from morning and reached a peak at 1700 hrs.
The total time spent by male birdwing on Bauhinia audax increased from 0800 hrs
and reached a peak at 1200 hrs and then decreased slowly until evening (Figure 5.8).
Whereas, the time spent on Bauhinia bidentata var. breviflora increased from 0800 hrs
117
and remained unchanged until 1100 hrs. After that, the time spent decreased until 1300
hrs and increased again and reached a peak at 1500 hrs. Thereafter, the time spent
decreased. The total time spent on Hibiscus rosa-sinensis was quite long at the
beginning of the day and then decreased and increased gradually within the study hours.
After averaged all the plant species, the total time spent decreased gradually from the
start of the experiment. The male spent more time on Hibiscus rosa-sinensis.
The female only spent their time on Bauhinia bidentata var. breviflora after 1300 hrs
(Figure 5.9). The time spent increased from 1400 to 1700 hrs and then decreased. The
total time spent on Bauhinia audax increased from 0800 hrs to about 1100 hrs and then
gradually decreased until 1800 hrs. The total time spent after average of all plant species
followed the rhythm on Bauhinia audax.
The rhythms of total flowers or flower clumps visited by a male birdwing were
similar to the total time spent by a male birdwing mentioned above except the rhythm
recorded from Hibiscus rosa-sinensis (Figure 5.10). The total flowers or flowers clumps
visited was decreased from 0800 hrs until 1000 hrs and then increased and reached a
peak at 1100 hrs. Then the number decreased until 1500 hrs. After that, the number
increased again and reached a smaller peak at 1600 hrs. Then the number decreased
until 1800 hrs. The rhythms of female also similar to the total time spent (Figure 5.11). University of Malaya
118
6.0 Average of all plant species 5.0 4.0 3.0 2.0 1.0 0.0 2.0 Bauhinia audax 1.6
1.2
0.8
0.4
0.0 0.6 Bauhinia bidentata var. breviflora 0.5 0.4 0.3
Monthly average count of birdwing nectaring count of birdwing averageMonthly 0.2 0.1 0.0 0.4 Hibiscus rosa-sinensis 0.3 0.2 0.1 0.0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time (hrs) Standardised count Actual count University2 per. Mov. Avg.of (Standardised Malaya count)
Figure 5.6: The number of male birdwings nectaring on each plant species and the average of all plant species at Ulu Geroh village.
119
2.0 Average of all plant species 1.6 1.2 0.8 0.4 0.0 0.5 Bauhiniaaudax 0.4 0.3 0.2 0.1 0.0 0.2 Bauhinia bidentata var. breviflora
Monthly average count of birdwing nectaring birdwingof count averageMonthly 0.1
0.0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time (hrs) Standardised count Actual count 2 per. Mov. Avg. (Standardised count)
Figure 5.7: The number of female birdwings nectaring on each plant species and the average of all plant species at Ulu Geroh village.
University of Malaya
120
50.0 Average of all plant species 40.0 30.0 20.0 10.0 0.0 20.0 Bauhinia audax 15.0 10.0 5.0 0.0 10.0 Bauhinia bidentata var. breviflora 5.0
0.0 70.0 Hibiscus rosa-sinensis 60.0 50.0 40.0 30.0 Monthly average of total time spent by a birdwing on a plant (sec) plant a onbirdwingbya spent time total ofaverage Monthly 20.0 10.0 0.0 0800090010001100 1200130014001500160017001800 Time (hrs) Total time spent 2 per. Mov. Avg. (Total time spent) University of Malaya Figure 5.8: Total time spent by a male birdwing at each hour on a plant of each plant species and the average of all plant species at Ulu Geroh village.
121
40.0 Average of all plant species 30.0
20.0
10.0
0.0 40.0 Bauhinia audax 30.0
20.0
10.0
0.0 20.0 Bauhinia bidentata var. breviflora
10.0
0.0 Monthly average of total time spent by a birdwing on a plant (sec) plant a onbybirdwinga spent time total of averageMonthly 080009001000 11001200130014001500160017001800 Time (hrs) Total time spent 2 per. Mov. Avg. (Total time spent)
Figure 5.9: Total time spent by a female birdwing at each hour on a plant of each plant species and the average of all plant species at Ulu Geroh village.
University of Malaya
122
12 Average of all plant species 10 8 6 4 2 0 8 Bauhinia audax 6 4 2 0 6 Bauhinia bidentata var. breviflora 4 2 0 16 14 Hibiscus rosa-sinensis 12 10 8 6 4 2 Monthly average of the total flowers or flower clumps visited by a birdwing on a planta onbirdwingbya visited clumps or flower flowers total the ofaverage Monthly 0 080009001000 11001200130014001500160017001800 Time (hrs) UniversityTotal flowers / flower of clumps Malaya 2 per. Mov. Avg. (Total flowers / flower clumps)
Figure 5.10: Total flowers or flower clumps visited by a male birdwing at each hour on a plant of each plant species and the average of all plant species at Ulu Geroh village.
123
6 Average of all plant species 4
2
0 6 Bauhinia audax 4
2
0 6 Bauhinia bidentata var. breviflora
birdwing on a plant a onbirdwing 4
2
0 080009001000 11001200130014001500160017001800 Time (hrs) Total flowers / flower clumps
Monthly average of the total flowers or flower clumps visited bya visited clumps or flower flowers total the of average Monthly 2 per. Mov. Avg. (Total flowers / flower clumps)
Figure 5.11: Total flowers or flower clumps visited by a female birdwing at each hour on a plant of each plant species and the average of all plant species at Ulu Geroh village.
5.4.2.2 Kenaboi Forest Reserve
The birdwing nectaring count was low for both sexes in Kenaboi Forest Reserve. The Universitynumber of male birdwing nectaring on Bauhiniaof bidentata Malaya var. breviflora was almost constant throughout a day (Figure 5.12). This is the same for average of all plant
species. However, the number of male birdwing on Saraca declinata was higher from
1400 to about 1500 hrs compared to morning hours. It peaked at 1600 hrs. The number
of male birdwing on Hibiscus rosa-sinensis peaked at 1400 hrs and a smaller peak
occurred at 1000 hrs. The number of female birdwing on Saraca declinata peaked at the
124
same time of the male, but there was another small peak occurred at 1200 hrs (Figure
5.13). The number of female birdwings on Bauhinia bidentata var. breviflora and the
average of all plant species was constant throughout a day.
6 5 Average of all plants species 4 3 2 1 0 7 6 Saraca declinata 5 4 3 2 1 0 2 Hibiscus rosa-sinensis 1
0 5 4 Bauhinia bidentata var. breviflora Monthly average count of birdwing nectaring birdwingof count averageMonthly 3 2 1 0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time (hrs) Standardised count Actual count 2 per. Mov. Avg. (Standardised count)
UniversityFigure 5.12: The number of male birdwings of nectaring Malaya on each plant species and the average of all plant species at Kenaboi Forest Reserve.
125
5 Average of all plant species 4 3 2 1 0 2 Saraca declinata 1
0 4 Bauhinia bidentata var. breviflora 3 2 1
Monthly average count of birdwing nectaring birdwingof count averageMonthly 0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time (hrs) Standardised count Actual count 2 per. Mov. Avg. (Standardised count)
Figure 5.13: The number of female birdwings nectaring on each plant species and the average of all plant species at Kenaboi Forest Reserve.
Overall, the total time spent by a male birdwing decreased gradually from 0800 to
about 1200 hrs and then increased gradually and reached a peak between 1400 and 1500
hrs and thereafter it decreased (Figure 5.14). The total time spent on Bauhinia bidentata
var. breviflora followed a similar trend with the average of all plant species. However, Universitymale spent more time on Hibiscus rosa-sinensis of Malaya in the evening compared to the morning. The total time spent on Saraca declinata increased slowly from 0800 hrs and
reached a peak around 1000 to 1100 hrs then it decreased. It increased again after 1300
hrs to about 1500 hrs and then decreased until 1800 hrs.
126
60.0 50.0 Average of all plants species 40.0 30.0 20.0 10.0 0.0 200.0 Saraca declinata 150.0 100.0 50.0 0.0 12.0 8.0 Hibiscus rosa-sinensis 4.0 0.0 60.0 50.0 Bauhinia bidentata var. breviflora 40.0 30.0 20.0
Monthly average of total time spent planttime a on total birdwing byaof averageMonthly 10.0 0.0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time (hrs) Total time spent 2 per. Mov. Avg. (Total time spent)
Figure 5.14: Total time spent by a male birdwing at each hour on a plant of each plant species and the average of all plant species at Kenaboi Forest Reserve.
Whereas, the total time spent by a female birdwing on Bauhinia bidentata var.
breviflora increased gradually from 0900 to 1800 hrs (Figure 5.15). On Saraca
declinata, the female spent more time between 1400 and 1500 hrs. A smaller peak Universityoccurred at 1200 hrs. In overall, the total of time spent Malaya by a female birdwing increased gradually from 0900 to about 1500 hrs and then decreased slightly until 1700 hrs and
increased again until 1800 hrs. The total flowers or flower clumps visited by a male and
female birdwings (Figure 5.16 and Figure 5.17) followed similar rhythm patterns of the
total time spent by both birdwing sexes.
127
100.0 Average of all plants species 80.0 60.0 40.0 20.0 0.0 250.0 Saraca declinata 200.0 150.0 100.0 50.0 0.0 100.0 Bauhinia bidentata var. breviflora 80.0 60.0 40.0 20.0 0.0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time (hrs) Monthly average of total time spent(sec) plant time a on total birdwing byofa average Monthly Total time spent 2 per. Mov. Avg. (Total time spent)
Figure 5.15: Total time spent by a female birdwing at each hour on a plant of each plant species and the average of all plant species at Kenaboi Forest Reserve.
University of Malaya
128
12.0 10.0 Average of all plants species 8.0 6.0 4.0 2.0 0.0 12.0 10.0 Saraca declinata 8.0 6.0 4.0 2.0 0.0 5.0 4.0 Hibiscus rosa-sinensis 3.0
birdwing on a planta onbirdwing 2.0 1.0 0.0 12.0 10.0 Bauhinia bidentata var. breviflora 8.0 6.0 4.0 2.0
Monthly average of the total flowers or flower clumps visited bya visited clumps or flower flowers total the ofaverage Monthly 0.0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time (hrs) Number of flowers/flower clumps 2 per. Mov. Avg. (Number of flowers/flower clumps)
Figure 5.16: Total flowers or flower clumps visited by a male birdwing at each hour on a plant of each plant species and the average of all plant species at Kenaboi Forest Reserve.
University of Malaya
129
12.0 10.0 Average of all plants species 8.0 6.0 4.0 2.0 0.0 10.0 Saraca declinata 8.0 6.0 4.0 2.0 0.0 12.0 10.0 Bauhinia bidentata var. breviflora 8.0 6.0 4.0 2.0 0.0 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time (hrs)
Monthly average of total time spent(sec) plant time a on total birdwing byofa average Monthly Number of flowers/flower clumps 2 per. Mov. Avg. (Number of flowers/flower clumps)
Figure 5.17: Total flowers or flower clumps visited by a female birdwing at each hour on a plant of each plant species and the average of all plant species at Kenaboi Forest Reserve.
5.4.3 Preferred Nectar Plants
5.4.3.1 Ulu Geroh Village
Table 5.4 shows that not all plants were flowering in each month especially the UniversityBauhinia audax and Bauhinia bidentata var.of breviflora Malaya. Among Bauhinia audax plants, plant SSDJB was flowering in all months and the number of flowers also at the most.
The SSDSS only flowered in September 2011 while the ASS flowered in the first three
months but the number of flowers was very less. The JMPW of Bauhinia bidentata var.
breviflora flowered in September while TJDD flowered at the most in September and
very less in November and December. Most of the Hibiscus rosa-sinensis plants
130
flowered every month except BR5 which did not flower in September. The BR4 had the
most flowers among the Hibiscus rosa-sinensis plants.
Table 5.4: The number of flowers or flower clumps on each plant in each month in Ulu Geroh Village.
Month Bauhinia audax (clump) Bauhinia Hibiscus rosa-sinensis bidentata var. (individual flower) breviflora (clump) SSDSS ASS SSDJB JMPW TJDD BR1 BR2 BR3 BR4 BR5 Sep 2011 43 3 40 46 70 9 11 5 68 0 Oct 2011 0 2 61 0 0 9 11 10 70 8 Nov 2011 0 1 47 0 2 10 12 10 80 6 Dec 2011 0 0 32 0 4 9 11 7 79 8
All five Hibiscus rosa-sinensis plants attracted very less birdwings even on the plant
that had many flowers (BR4) (Figure 5.18). The percentages of male birdwing found on
Hibiscus rosa-sinensis were less than 5% for BR2 (2.79%), BR3 (0.85%), BR4 (3.59%)
and BR5 (1.41%). The BR1 had slightly higher percentage, about 8.78%. Female
birdwing was not found on Hisbiscus rosa-sinensis. The plant TJDD attracted more
males (15.47%) compared to JMPW (15.47 versus 7.76%) of Bauhinia bidentata var.
breviflora. But TJDD attracted less female compared to JMPW (0.40% versus 33.33%).
Two plants of Bauhinia audax (ASS and SSDJB) attracted more male birdwings
compared to other plants of the same species or different species such as in Hibiscus
rosa-sinensis, with 34.11% on ASS and 43.45% on SSDJB. There were more female
birdwings attracted to SSDJB compared to other plants (e.g. 60.57% on SSDJB and University24.07% on ASS. The SSDSS attracted fewer of males (1.46%)Malaya and females (0.98%). The significant results indicated that at least one plant species attracted significantly
different number of birdwings among all the plant species as shown in Table 5.5
(Kruskal Wallis: H=7.92, df=2, p<0.05 for males; H=13.75, df=2, p<0.005 for females).
131
100.0 Bauhinia audax Bauhinia bidentata Hibiscus rosa-sinensis 80.0 var. breviflora
60.0
40.0 clumps
20.0 Monthly average of the ofthe average Monthly number of flowers/flower of flowers/flower number 0.0 SSDSS ASS SSDJB JMPW TJDD BR1 BR2 BR3 BR4 BR5 Standardised count Actual count 100.0 Bauhinia audax Bauhinia bidentata Hibiscus rosa-sinensis 80.0 var. breviflora
60.0
40.0
20.0
of birdwing nectaring on nectaringon ofbirdwing 0.0 each flower / flower clump flower/ eachflower Monthly average percentage Monthlypercentage average SSDSS ASS SSDJB JMPW TJDD BR1 BR2 BR3 BR4 BR5
Male Female 7 Bauhinia audax Bauhinia bidentata Hibiscus rosa-sinensis 6 var. breviflora 5 4 3 2
Monthly average of of Monthlyaverage 1 birdwing on eachplant on birdwing total flowers visited by a a totalby flowersvisited 0 SSDSS ASS SSDJB JMPW TJDD BR1 BR2 BR3 BR4 BR5
Male Female 35 Bauhinia audax Bauhinia bidentata Hibiscus rosa-sinensis 30 var. breviflora 25 20 15 10 each each plant 5 0 Monthly average oftotalaverage Monthly
time spent by a birdwing on on birdwing by time spent a SSDSS ASS SSDJB JMPW TJDD BR1 BR2 BR3 BR4 BR5 Plants code Male Female
Figure 5.18: Monthly averages of the number of flowers or flower clumps, Universitypercentage of birdwing nectaring on eachof flower Malaya or flower clump, total flowers visited and total time spent by a birdwing on each plant in second in Ulu Geroh Village.
132
Table 5.5: Comparisons between three flowering plant species on the percentage of birdwing nectaring on each flower or flower clump, total flowers visited and total time spent by a birdwing using Kruskal-Wallis analysis in Ulu Geroh Village.
Plant Kruskal- Male Female species Wallis Percentage Total Total Percentage Total Total analysis flower time flower time values spent spent Bauhinia Median 33.914 2.925 8.200 40.343 2.000 10.875 audax Average 23.3 10.6 9.4 25.6 5.6 5.7 rank Z value 2.62 0.04 -0.59 3.45 1.17 1.46 Bauhinia Median 6.445 1.650 5.583 0.600 1.250 6.750 bidentata Average 17.8 4.3 3.7 18.3 3.0 2.5 var. rank breviflora Z value 0.41 -1.96 -2.17 0.53 -1.17 -1.46 Hibiscus Median 3.020 3.333 14.750 0.000 - - rosa- Average 12.6 12.3 13.3 11.5 - - sinensis rank Z value -2.64 1.36 2.12 -3.47 H value 7.92 4.19 6.47 13.75 1.37 2.14 df 2 2 2 2 1 1 p value 0.019* 0.123 0.039* 0.001* 0.242 0.143
When the birdwing was found nectaring on flowers, the total flowers of flower
clumps visited was recorded together with the total time spent on each plant. Figure
5.18 shows the male birdwings visited more flowers on BR4 (5.17 flowers) compared to
other plants. The number of flowers visited by a male on other plants was 3.29
(SSDJB), 3.00 (BR1), 2.50 (BR2), 2.00 (SSDSS), 2.00 (BR5), 1.88 (ASS), 1.65
(JMPW), 1.33 (TJDD) and 1.00 (BR3). Whereas for female, they visited 3.00 flowers
(SSDSS), 2.69 flowers (SSDJB), 1.50 flowers (ASS), 1.50 flowers (TJDD) and 1.00
flower (JMPW). There was not a significant different between all plant species in both Universitymale and female birdwings (Kruskal Wallis: of H=4.19, Malaya df=2, p>0.05 for males; H=1.37, df=1, p>0.05 for females; Table 5.5).
The male birdwings also spent more time on BR4 (26.48 sec) compared to BR1
(16.17 sec), BR2 (13.00 sec), BR5 (10.00 sec), SSDSS (9.50 sec), SSDJB (9.26 sec),
ASS (8.13 sec), JMPW (5.80 sec), BR3 (5.00 sec) and TJDD (4.79 sec) (Figure 5.18).
133
There was a significant different between all three plant species (Kruskal Wallis:
H=6.47, df=2, p<0.05; Table 5.5). The female birdwings spent longer time on SSDSS
(16.00 sec), followed by SSDJB (11.20 sec), ASS (8.75 sec), TJDD (8.50 sec) and
JMPW (5.00 sec) (Figure 5.18). The total time spent by a female birdwing was not
different significantly among all three plant species (Kruskal Wallis: H=2.14, df=1,
p>0.05; Table 5.5).
5.4.3.2 Kenaboi Forest Reserve
In 25 and 26 July 2012, only two Bauhinia bidentata var. breviflora plants flowered,
BBB1 has 3 flower clumps and BBB2 has 15 flower clumps (Figure 5.19). At the same
time, a Saraca declinata (SD1) flowered with 14 flower clumps. A total of five
Hibiscus rosa-sinensis plants flowered which is about 100 m away from the SD1, that is
HRS1 has 42 flowers, HRS2 has 45 flowers, HRS3 has 23 flowers, HRS4 has 25
flowers and HRS5 has 32 flowers. There was about 56.83% of males found nectaring on
SD1, 20.40% on BBB1, 8.16% on BBB2, 2.91% on HRS1, 6.80% on HRS2 and 4.90%
on HRS4 (Figure 5.19). No male was found on HRS3 and HRS5. Females only attracted
to SD1 and BBB1 with 39.13% on SD1 and 60.87% on BBB1. The number of flowers
or flower clumps visited by a male on SD1 was about 4.10 flower clumps, on BBB1
was 1.00 flower clump, on BBB2 was about 3 flower clumps, on HRS1 was about 2.50
flowers, on HRS2 was about 1.17 flowers and HRS2 was about 3.50 flowers (Figure
5.19). Whereas, female birdwing visited 6.67 flower clumps of SD1 and 3.00 flower Universityclumps on BBB1. Figure 5.19 also shows thatof a male Malayabirdwing spent 70.47 sec on SD1, 7.00 sec on BBB1, 104.50 sec on BBB2, 4.00 sec on HRS1, 2.83 sec on HRS2 and 8.50
sec on HRS4. Female spent 102.33 sec on SD1 and 33.00 sec on BBB1.
134
60 Saraca Bauhinia Hibiscus rosa-sinensis 45 50 declinata bidentata var. 42 breviflora 40 32 30 23 25 plant 20 14 15 10 3 / flower clump on each clump / flower Total number of flower of number Total 0 SD 1 BBB 1 BBB 2 HRS 1 HRS 2 HRS 3 HRS 4 HRS 5 100 Saraca Bauhinia Hibiscus rosa-sinensis 80 declinata bidentata var. breviflora 60
40
flower clump clump flower 20 nectaring on a flower / on flower a nectaring Percentage of birdwingof Percentage 0 SD 1 BBB 1 BBB 2 HRS 1 HRS 2 HRS 3 HRS 4 HRS 5 10 Saraca Bauhinia Hibiscus rosa-sinensis 8 declinata bidentata var. breviflora 6
4
on each plant on each 2 visited byabirdwing visited Average of totalofflowers Average 0 SD 1 BBB 1 BBB 2 HRS 1 HRS 2 HRS 3 HRS 4 HRS 5 200 Saraca Bauhinia Hibiscus rosa-sinensis 150 declinata bidentata var. breviflora 100
50 plant(sec)
0 by a birdwing on each abirdwing by SD 1 BBB 1 BBB 2 HRS 1 HRS 2 HRS 3 HRS 4 HRS 5 Average of total time spent totalof Average Plant's code University ofMale FemaleMalaya
Figure 5.19: The total number of flowers or flower clumps on each plant, percentage of birdwing nectaring on each flower or each flower clump, average of total flowers visited by a birdwing on each plant and average of total time spent by a birdwing on each plant in Kenaboi Forest Reserve. Data was summarised from 25 and 26 July 2012, 20 minutes observation at each hour from 0800 to 1800 hrs.
135
During observations on 27 and 28 February 2013, there were 13 flower clumps found
on Saraca declinata (SD2), 318 flower clumps on Bauhinia bidentata var. breviflora
(BBB3), 129 flower clumps on BBB4 and 110 on BBB5 (Figure 5.20). In the same
graph shows that the percentage of male birdwing nectaring on a flower clump of SD2
was 96.21% and only 3.79% on BBB5 however, there was about 83.34% on SD2 for
female birdwing, 6.81% on BBB3 and 9.85% on BBB5. The number of flower clumps
visited by a male birdwing was 5.33 on SD2 and 0.50 on BBB5 while female visited
2.00 on SD2, 4.50 on BBB3 and 12.00 on BBB5. The male birdwing also spent more
time on SD2 (44.33 sec) but only 3.00 sec on BBB5. However, the female birdwing
spent more time on BBB5 (45.00 sec) compared to 23.50 sec on BBB3 and 11.00 sec on
SD2 (Figure 5.20). The differences among all plant species on the percentage of
birdwing nectaring (Kruskal Wallis: H=4.81, df=2, p>0.05 for males; H=5.71, df=2,
p>0.05 for females), total flower clumps visited (Kruskal Wallis: H=4.69, df=2, p>0.05
for males; H=0.33, df=1, p>0.05 for females), and total time spent (Kruskal Wallis:
H=2.47, df=2, p>0.05 for males; H=0.00, df=1, p>0.05 for females), were not
significant (Table 5.6).
5.4.4 Courting and Mating Incidences
A total of seven male and female interactions were observed during this study which
may lead to successful mating out of sight from observer. Another nine observations
had no indication of successful mating because the males and the females were Universityseparated in the end. A successful mating ofpair was notMalaya seen during the study duration. Behavioural descriptions of these episodes are described in Table 5.7.
136
350 Saraca Bauhinia bidentata 300 318 declinata var. breviflora 250 200 150 129 100 110 50 clumps on each planteach on clumps Total number of flower ofnumber Total 0 13 SD 2 BBB 3 BBB 4 BBB 5 100 Saraca Bauhinia bidentata 80 declinata var. breviflora
60
clump 40
20 nectaring on a flowera on nectaring Percentage of birdwingof Percentage 0 SD 2 BBB 3 BBB 4 BBB 5 15 Saraca Bauhinia bidentata 12 declinata var. breviflora
9
6
on each plant each on 3 visited by a birdwinga by visited Average of total flowers flowers total of Average 0 SD 2 BBB 3 BBB 4 BBB 5 60 Saraca Bauhinia bidentata declinata var. breviflora 40
20 (sec)
0 SD 2 BBB 3 BBB 4 BBB 5 Average of total time spent time total of Average by a birdwing on each plant each onbirdwing aby Plant's code University ofMale Female Malaya
Figure 5.20: The total number of flower clumps on each plant, percentage of birdwing nectaring on each flower clump, average of total flowers visited by a birdwing on each plant and average of total time spent by a birdwing on each plant in Kenaboi Forest Reserve. Data was summarised from 27 and 28 February 2013, 20 minutes observation at each hour from 0800 to 1800 hrs.
137
Table 5.6: Comparisons between three flowering plant species on the percentage of birdwing nectaring on each flower or flower clump, total flowers visited and total time spent by a birdwing using Kruskal-Wallis analysis in Kenaboi Forest Reserve.
Plant Kruskal- Male Female species Wallis Percentage Total Total Percentage Total Total analysis flower time flower time values spent spent Saraca Median 76.52 4.72 57.40 61.23 4.33 56.67 declinata Average 11.50 7.50 6.50 11.00 2.50 3.00 rank Z value 2.15 2.00 1.33 1.93 -0.58 0.00 Bauhinia Median 3.79 1.00 7.00 6.81 4.50 33.00 bidentata Average 6.00 2.70 4.70 7.20 3.30 3.00 var. rank breviflora Z value -0.41 -1.64 0.15 0.57 0.58 0.00 Hibiscus Median 2.914 2.50 4.00 0.00 - - rosa- Average 5.00 4.30 3.00 4.00 - - sinensis rank Z value -1.22 -0.15 -1.34 -2.03 - - H value 4.81 4.69 2.47 5.71 0.33 0.00 df 2 2 2 2 1 1 p value 0.090 0.096 0.291 0.058 0.564 1.00
University of Malaya
138
Table 5.7: Interactions between females and males in Ulu Geroh Village and Kenaboi Forest Reserve.
Date Time (hrs) Location Behavioural Descriptions A female was flying along the pathway Ulu Geroh besides Puddle 2 at eye level and two males 16-Mar-10 0924 Village followed her, they later disappeared into the forest behind the puddling spot. A male chased a female at the pathway besides Puddle 1 at eye level. Both birdwings Ulu Geroh displayed circling flight with female 18-Mar-10 0927 Village constantly flying above the male of 1 meter apart and then flew out of sight. The female's abdomen curled down when she was flying. A female was sighted at Puddle 2 and a male Ulu Geroh from the puddle approached and followed her 23-Jun-10 0955 Village into the forest behind. They flew at about 5 meter above eye level. A pair of birdwings was spotted at Puddle 2, Ulu Geroh the male approached the female and then flew 23-Dec-10 1530 Village into the forest behind. They were flying at about 5 meter above eye level. A female was flying around Puddle 1 at about 10 meter above ground. A male was trying to Ulu Geroh 24-Aug-11 1010 approach her for about 1 to 2 minutes. The Village male was flying together with the female but after that he flew away from the female. A female was flying beside the camera at Puddle 1, she met with a male flying from the Ulu Geroh opposite direction. There was no interaction 25-Aug-11 0910 Village between them. The male was flying to puddle while the female to the smaller puddling spot then into the forest. A male was chasing a female, both floating in the air and in less than a few seconds, the Ulu Geroh 18-Oct-11 1320 female flapped her wings quickly. Then the Village male flew to other direction while the female disappeared into the forest. A male was chasing a female and both of them landed on a same leaf of a tree at about 5 meter above eye level. The female landed above and the male below her. The male's Ulu Geroh University19-Oct-11 0850 wingsof were Malayashivering. After about 15 minutes, Village the male flew close to the female and touch the female right wing slightly with his left wing. After this action, the female flew into the forest and the male followed.
139
Table 5.7, continued.
Date Time (hrs) Location Observation Two males were trying to approach a nectaring female around Saraca declinata however, the Kenaboi female flew away after few seconds. After 25-Jul-12 1508 Forest that, the two males were chasing each other Reserve and disappeared from that area after few seconds. A male was trying to approach a female who was nectaring on Saraca declinata. The male Kenaboi followed the female from one flower to 25-Jul-12 1650 Forest another flower but the male did not take nectar Reserve from the plant. The female always above the male. Both disappeared together after 5 minutes on the plant. Kenaboi A male was trying to go near a nectaring 7-Nov-12 1500 Forest female on Bauhinia bidentata but about 10 Reserve seconds, he left. A male was approaching a nectaring female but about few seconds, he left the area. Two males were trying to approach a female Kenaboi on Bauhinia bidentata plant while she was 7-Nov-12 1700 Forest nectaring. After about 20 seconds, she Reserve abandoned the plant and the two males were chasing each other and flew away from the plant.
A male was following a female to take nectar Kenaboi on Bauhinia bidentata. The female ignored 7-Nov-12 1800 Forest him and continued to feed. Then the male flew Reserve away after about 30 seconds.
Kenaboi Two males disturbed a female taking nectar on 8-Nov-12 1000 Forest Bauhinia bidentata. Reserve Kenaboi 8-Nov-12 1200 Forest A male followed a female to take nectar. Reserve
A male followed a female at the small stream Kenaboi University whereof the Malaya monitoring using single spot 25-Feb-13 1540 Forest observation was conducted. The female flew Reserve above and together disappeared in the forest.
140
5.5 Discussion
5.5.1 Puddling and Flight Rhythms
The males of T. b. albescens exhibits communal puddling behaviour along river
banks that sometimes reached hundreds of individuals (Corbet & Pendlebury, 1992).
The precise reason for this behaviour had not been investigated, but there were reports
on male butterflies that physiologically concentrated salts and transfer them to females
in the spermatophores as a nuptial gift which could enhance fecundity and longevity
(Boggs & Gilbert, 1979; Adler & Pearson, 1982; Pivnick & McNeil, 1987; Molleman et
al., 2005). In Trogonoptera brookiana, stable aggregations appeared to be associated
with hot springs or geothermal seepages (Corbet & Pendlebury, 1992), while males
might briefly alight to puddle at other locations along rivers and forest roads, or for
longer periods in areas with transient source of salts such as urine (Corbet &
Pendlebury, 1992). In Malaysia, as in other non-volcanic regions, hot springs resulting
from the heating of water by granitic intrusions of cooling magma or from the heating
of cold water that seeped deep into the earth’s mantle through faults and rises to the
surface by convectional pressure (Baioumy et al., 2014). The reasons for the attraction
of this birdwing to certain hot springs had not been investigated, but could be related to
thermoregulation or mineral and organic content in the water, or a combination of these
reasons.
The puddling activity in Ulu Geroh Village sometimes started as early as 0720 hrs Universityand sometimes several birdwings were observed of puddling Malaya at night. Usually in the early morning, birdwings were observed flying from the forest near the puddles but high up in
the canopy or resting on leaves before coming down to puddle. In the early morning,
behaviour of a birdwing flying and a few were following him was observed several
times during the study.
141
High puddling activity happened when the temperature was higher and the relative
humidity was lower throughout a day. The water temperature at two puddling sites was
higher and slightly alkaline compared to the river temperature. However, a proper
comparison of both parameters on the puddles and soil nearby cannot be done because
of villager’s issue. At Kuala Woh Recreational Forest in Perak state, another area well
known of birdwing puddling, the water temperature at river bank was also high and
reached about 41°C. Pola and García-París (2005) hypothesised that Papilio polytes
puddled on reef shelf could be due to two reasons and one of these was the higher water
temperature at sea reef platform.
Early morning puddling (i.e. around 0700 hrs) observed during the study probably is
a thermoregulatory strategy of the birdwings despite taking nutrients from the puddles
as reported by Adler and Pearson (1982) and Arms et al. (1974). Because the puddles
are naturally warm all the time, it probably encourages the birdwings to warmth their
bodies and wings by landing on it with their wings open horizontally. This posture is
similar to conduction basking which is a variation of dorsal basking described by
Kingsolver (1985a) or ground-contact described by Clench (1966). This basking method
was observed in alpine Parnassius on rocky and bare-ground areas which the papilionid
open its wings and together with the body oppressed to the ground to trap warm surface
air and conducting heat from substrate to the body. The night puddling observed during
the study period could be also due to the heat from the puddles. Barton et al. (2014) Universityshowed that Heteronympha merope, a common of brown Malaya in Australia with the wings held open is able to reduce heat lost through convection and maximise heat gained through
radiation.
During the hottest hours in a day, the puddling birdwings were observed avoiding
direct sunlight. They moved towards a shady area at the puddle and according to
142
Rawlins (1980), this is to avoid overheating that can reduce survivorship. Avoiding
overheating by decreasing activity or seeking for shade was reported in small
ectothermic animals (Sunday et al, 2014). Butterflies also seek shade when the
temperature rises above a limit (Clench, 1966). This behaviour is similar to the
description by Igarashi and Fukuda (1997). However, in the present study, the
birdwings did not disperse in the afternoon as mentioned by Igarashi and Fukuda but
continued to puddle with a few left or sometimes none after 1900 hrs. The birdwings
were sensitive to disturbance, a gust of strong wing or a fallen leaf will disrupt them
from puddle and land on trees nearby. However, they will puddle again after about 10 to
15 minutes.
A slight different puddling pattern between Puddle 1 and Puddle 2 could be because
of the location of the two puddles. The area at Puddle 1 is rather opened compared to
Puddle 2. Puddle 2 is surrounded by plants while Puddle 1 is not. Therefore, Puddle 2 is
more protected by wind and other disturbance. Puddle 1 is partially located under direct
sunlight while Puddle 2 was under a canopy, thus shaded. The rapid increased followed
by a gradual increased at Puddle 1 after 1100 hrs could be due to avoiding overheating
by afternoon sunlight. While overheating was not happened at Puddle 2, therefore the
number of puddling birdwings increased gradually all the time until reaching the peak.
Puddle 1 also has more birdwings because the area is bigger than Puddle 2.
A unique behaviour which has not been documented before were observed in several Universityepisodes during this current work: When theof birdwings Malaya fly off from the puddle due to disturbances, the birdwings will land on trees nearby. After a while, a few birdwings
will start flying around and approach other landing birdwings by getting very close to
them. When the birdwing approaching landing ones, he will flap his wing at high speed
and remain nearly stationary in the air. Sometimes, he will also fly up and down when
143
approaching the landing birdwing. After a few seconds, the landing birdwing will leave
the leaf and fly either to the puddle or follow the approaching birdwing or fly around
and behave like the birdwing approached him before. However, sometimes the landing
birdwing will not bother the calling from approaching birdwing. This behaviour was
also observed in the early morning when they start to puddle and in the evening when
they start to leave the puddle. Birdwings usually start to fly off from the puddle spot
when the time reaching 1700–1800 hrs. This novel behaviour can be interpreted to be
‘micro-territoriality’ where invading individuals were forcing attempts to displace
puddling birdwings to leave the area by flying above and very close to them. The
invaders wing- flapped quickly and stationary in the air for a few seconds until they
succeed to chase away the puddling birdwings to leave the area and flew up and land on
the trees nearby in the forest. This territorial acquisition behaviour is very pronounced
and never been reported for this birdwing. Other puddling butterfly species were rarely
seen puddling together with the birdwings. There were about two occasions that a
Papilio memnon was able to intrude to their territory for a short while. There were a few
occasions where butterflies Eurema from Pieridae family puddling about 2 meters away
from the birdwing territory.
Flight activity in Ulu Geroh Village was high just before and just after peak puddling
and remained constant during peak puddling. It is likely that flight activity was high in
the early morning as the birdwings flew from the forest to start the day’s puddling, and Universityagain in the evening as the birdwings left of the puddles Malaya to return to the forest. During afternoon, the birdwings usually flying around the village and landed on trees for a
while and fly again either to the puddle or patrolling around the village. The constant
high flying activity in from 0900–1700 did not seem to be affected by temperature and
relative humidity. The flight activities recorded from the study was different from
Trogonoptera brookiana trogon as reported in de Nicéville and Martin (1895) which
144
trogon was rarely seen during hottest hours. Slight differences between the present
findings and Orr (1982) were found. The starting flight hours of males and females in
the study were before 0800 hrs and 0900 hrs respectively while Orr (1982) reported that
the flight started at 0800 hrs for both males and females. While the last flight of male
and female in this study was 1900 hrs and 1700 hrs respectively. Orr’s findings were
slightly earlier. The female occurrence in Ulu Geroh Village was very low.
The female occurrence in Kenaboi Forest Reserve was also very low and only
occurred in the afternoon hours. The flight activity in Kenaboi Forest Reserve started at
0900 hrs which was about 2 hours later than in Ulu Geroh Village. This could be due to
the lack of warm puddle in Kenaboi Forest Reserve and therefore, they can’t apply the
thermoregulatory strategy as in the birdwings in Ulu Geroh Village. The birdwings in
Kenaboi Forest Reserve have to delay their flight activity until the air temperature
increase at later hours in the morning to warm their body. The peaks occurred at
different hours between single spot observation (i.e. 1000 hrs) and transect walk method
(i.e. 1200 hrs). Although the two study areas were not far apart, but the microhabitat
could caused the slight different on the flight activity and the number of birdwing in
flight. The sunlight penetrates better at single spot observation area compared to
transect walk area. The flight patterns seem to be followed the light intensity pattern in
which the activities were decreased when the light intensity was lower. On the other
hand, when the temperature remained high in the afternoon, the flight activities also Universitydecreased. of Malaya 5.5.2 Nectaring Behaviour
The nectaring rhythm differences in term of the number of birdwings and total time
spent between males and females, and between plant species are probably due to the
different daily activity budget of the birdwing sexes (Bąkowski et al., 2010) and maybe
145
also due to the nectar production of the flowering plants (Cruden et al., 1983). Both
quality and quantity of nectar play a role on the attractiveness of a given plant. Different
plant species may produce different nectar volume and therefore affecting the nectaring
behaviour of birdwings on different plant species. Several studies showed that nectar
production rates vary within and among plants, changing with time of day, genetic
differences, age, nutritional status of the plant, location of flower on the plant, and other
factors (Shuel, 1957; Percival, 1965; Cruden et al., 1983; Pleasants & Chaplin, 1983).
The flowering plants in Ulu Geroh Village were not flowering consistently and the
number of flowers on each plant also varies greatly during the observation months. In
months where more plants were in bloom, the birdwings shared more plants for
nectaring. On the other hand, in months where fewer plants were flowering, the
birdwings concentrated on the plants in bloom and therefore the percentages increased
tremendously (Appendix F). However, Hibiscus rosa-sinensis always has very low
percentage. When most of the plants were flowering, i.e. in September, the birdwings
attracted more to two out of three plants of Bauhinia audax, one plant of Bauhinia
bidentata and two out of four plants of Hibiscus rosa-sinensis. The birdwings attracted
to certain plants of the same plant species only and this observation was coincided with
Jumalon (1967) on Trogonoptera trojana. These observations can probably be
explained by different sugar production on different plants as supported by a study
conducted on Aralia hispida (Thomson et al., 1989). Throughout the study period, Universityfemale did not feed on Hibiscus rosa-sinensis of at all.Malaya The significant different in the birdwing percentage nectaring indicate at least one plant species attracted more
birdwings compared with other plant species. The birdwings seem to prefer on Bauhinia
species over hibiscus. The flower structure of Bauhinia audax, Bauhinia bidentata var.
breflora and Hibiscus rosa-sinensis are very different which the both Bauhinia flowers
are in clumps and Hibiscus is not. Bauhinia bidentata flowers are denser packed
146
compared to the other one. However, Bauhinia audax has stronger odour then Bauhinia
bidentata.
The non-significant results in Kenaboi Forest Reserve were probably due to small
sample size. This is happened because the lack of different flowering plant species
flowered at the same time, logistic and safety issues. The limited data presented
indicates more birdwings were found nectaring on Saraca declinata in both occasions.
Fewer birdwings were found on Bauhinia bidentata var. breviflora and Hibiscus rosa-
sinensis even the plants were blooms heavily. Similar to Ulu Geroh Village, females did
not occur on hibiscus. Clumps of Saraca declinata are greatly bigger than the Bauhinia.
The birdwings perhaps preferred flowers in clumps and this corresponds with most
of the nectaring sightings by the naturalists and other literature. They reported
Trogonoptera spp. was found nectaring on Bauhinia sp. (Jumalon, 1967; Corbet &
Pendlebury, 1992), Mussaenda sp. (Corbet & Pendlebury, 1992; Cole, 1998), Lantana
sp. (Straatman, 1955; Hoskins, 2015), Ixora sp. (Panchen, 1980; Momose et al., 1998),
Crossandra infundibuliformi and Duranta lorentzii (Goh, 1994). Birdwings were
observed taking nectar from Bougainvilla sp. in one of the field trips. Most of the
flowers of the plants are in clumps. Tiple et al. (2009) finding concluded that butterflies
with high wing load preferred densely packed flowers and the same found in present
study. All the nectaring plants recorded for Parnassius apollo also have massed flowers, Universitya flower character for large butterfly (Baz, 2002).of Malaya Birdwings visited more flowers and spent longer time on Hibiscus rosa-sinensis in
Ulu Geroh Village were probably due to the insufficient resources achieved in a short
visit. Some of the Bauhinia species visited by butterflies were sucrose dominant
(Hokche & Ramirez, 1990; Lau et al., 2005). While Hibiscus has very less sucrose and
near to zero but has more fructose and glucose (Freeman et al., 1991). Among the 62
147
plant species studied in south and southeast Asia, they found that, Ixora and Mussaenda
(reported nectar plants of Trogonoptera brookiana) have higher sucrose than fructose
and glucose (Freeman et al., 1991). This may explain why the birdwings spent more
time and visited more flowers on Hibiscus. They would like to exploit more sucrose
which is not dominant in Hibiscus. However, different species in same genera may have
different sugar contents (Chalcoff et al., 2006). The presence of other chemical
properties may affect their nectaring preference, such as amino acid that increase
butterflies longevity (Murphy et al., 1983) and fecundity (Mevi-Schütz & Erhardt,
2005) in general and perhaps alkaloid helps in defending predators (Brown, 1984;
Masters, 1990). Among the six Bauhinia species examined by Hokche and Ramirez
(1990), very less alkaloid detected and the quantity of amino acid was quite a lot.
The birdwings spent more time and visited more flowers on Saraca declinata in
Kenaboi Forest Reserve maybe because of two reasons. The first reason is same as in
Hibiscus, the resources needed by the birdwings are not satisfactory, and therefore, they
spent longer time on Saraca. The second reason would be there are many flowers in a
clump, consequently, they spent more time on exploiting each of the flowers.
Unfortunately, the literature of chemical properties in Saraca nectar could not be found.
Considering there were more birdwings visited Saraca, then the nectar contents should
favoured by the birdwings.
In Kenaboi Forest Reserve, males of Trogonoptera brookiana albescens and Troides Universitysp. were always chasing each other on Bauhiniaof plants. Malaya The Bauhinia plants received direct sunlight whereby the Saraca was not. When the hot sunlight shined on the
Bauhinia plants, birdwings stopped nectaring but flying around or just rest on plants.
The influence of these, males chasing each other and avoiding direct sunlight on the
nectaring result was not quantified. During the study, whether or not the same birdwing
148
repeatedly visited the same plant at different time within same observation hour was not
able to be determined.
5.5.3 Courting and Mating Behaviour
Males and females interacted throughout the day with no preference for reproductive
seasons. In Ulu Geroh Village, the interactions between males and females confined to
the area nearby the puddles and flowering plants. The interactions in Kenaboi Forest
Reserve usually occur at flowering plants. These happened because more time was
spent on observations at puddles and flowering plants. The males in Ulu Geroh Village
seemed to use perching and patrolling strategies to locate mates while in Kenaboi Forest
Reserve, they confined to patrolling strategy to locate mates nearby nectaring plants.
Butterflies using different strategies at different locations were reported in Pararge
aegeria, a nymphalid butterfly (Merckx & Van Dyck, 2005). The frequency of using
territorial perching behaviour was lower at agricultural landscape compared to
woodland landscape. The frequency of using intermediate perching and patrolling
behaviour was higher in agricultural landscape. They suspected that the behaviour
differences could be due to ambient temperature and the sunlit penetration. In the
present study, the population size of both sexes and the presence of puddles may affect
the strategy application. When the population size is small, i.e. in Kenaboi Forest
Reserve, the chance for the female to locate perching male is low. Patrolling strategy by
the male might increase the chance to meet female.
UniversityMost of the courting was initiated withof the maleMalaya flying towards the female. Triggering of courtship by females as described by Skertchly (1889) and Panchen
(1980) was not witnessed in this study. Wooing started by female would happen when
the older female have not mated since emergence (Scott, 1972; Bergman et al., 2011).
149
Krebs (1988) reported that the first courting flight of Papilio glaucus was normally by
males and rarely by females.
The males always fly below and behind the females when approaching them. There
was only one occasion where the female probably showing her acceptance to the mate
by the curled down abdominal posture. The female flapped quickly and merely fly away
are two of the “rejection dances” to reject a male from approaching further, behaviour
of general butterflies described by Scott (1972). However, other “rejection dances” and
special rejection, such as the female flies vertically until the male returns to the ground,
abdomen put vertically above the horizontal wing, spread her wings, close her wings
above thorax postures mentioned in Scott (1972) were not observed in this study.
Rejection of mates shown by females could be due to her prior mated status. The only
successful mating was reported on mountain region of the head waters of Segama River
which occurred on top of an orange-blossomed tree. Birdwings might be seeked out
mountain region to mate and might not observe easily. Most of the observation in Ulu
Geroh Village ended by the pairs fly into the forest behind, whereas the pairs in
Kenaboi Forest Reserve ended by the female fly away from the courting spot.
Competition for mate is obvious in this species because virgin females are limited
resources with low availability, thus it is a common phenomenon to observe females
being pursued by more than one male. The competitive males proceed with displays of
aggression even after the female left the courting spot. This is supported by Weir et al. University(2011) that when the operational sex ratio ofis biased Malayato males, the rate of aggression, a sign of contest competition is increased. Incidence of two males approaching a female
was observed in Ulu Geroh too but in lower frequency. This could be because of the
males were too tempted on puddles. In some occasions, female passed through the
puddles but none of the puddling males were bothered by her presence. Lesser courting
150
and copulation activities were reported if two males of Bicyclus anynana were caged
with a female compared to a single pair (Westerman et al., 2014). Eliot (1972) noted
that the rarity of albescens female was perhaps due to its different habits and habitat,
they do not puddle and prefer higher elevations between 1500–4000 ft. However, a
systematic study on vertical distribution of both males and females has never been
conducted to conclude the observation. The occurrence of females in Ulu Geroh Village
(about 250 m a.s.l. around study site) and Kenaboi Forest Reserve (about 200 m a.s.l.)
shows that the females do come down to lower elevation to look for nectaring plants
and also searching for mates. However, specific locations where the mating or
copulation events occurred are still unknown.
In an occasion, a female alighted on a leaf after being chased by a male is a
behaviour showing that she is receptive (Scott, 1972). Then, the male landed below her
and starts shivering. Shivering was probably to increase thoracic temperature, as the
event occurred between 0800 to 0900 hrs, in which the air temperature was not high.
Whether or not the shivering increased the abdomen temperature to release pheromone
is not known. The shivering behaviour was observed on crepuscular owl butterflies
where the author claimed that this special behaviour increased thoracic temperature and
prepares the butterflies for reproductive encounters (Srygley, 1994). Before female
leaving the leaf, the male flew very close above her and touched her wing with his wing.
This behaviour is probably to pass scent or pheromone on the female unlike common Universityreports whereby scent is passed to the femaleof antennae Malaya (Scott, 1972). A Papilionid, Papilio glaucus do not have antennae-wing contact but only occasionally wing-wing
contact (Krebs, 1988). The male of another Papilionid, Troides oblongomaculatus
papuensis, do transfer the courtship pheromone to female antennae by open it
androconial pouch and the hair-like androconial scales will drop off and stick on the
151
antennae. The androconial pouch of Trogonoptera brookiana albescens were never
opened throughout the study.
5.6 Conclusion
To date this is the first known systematic behavioural study which includes puddling,
flying and nectaring behaviours of the birdwing in Malaysia. Courting behaviour was
conducted in ad libitum manner. Herein, a number of novel discoveries on the
behaviour of birdwings have been described, such as a micro-territorial behaviour by
exhibit calling other birdwings to puddle or to leave puddle, avoiding direct heating,
early morning and night puddling, adjusting behaviour due to microhabitat differences,
male competition to approach female and shivering during courting. The objectives of
the study were achieved. Here too it is evident that birdwings showed clear preference
for clumped flowers and females non-preference for hibiscus but there are limitations in
the nectaring studies. There are gaps identified which are not able to fill in the present
studies and therefore proposed for future work. Some of which would be studies on the
following:
contents of the puddle that attract birdwings;
a method to quantify the micro-territorial calling behaviour;
activity budget of both male and female birdwings;
flowering phenology, nectar production and its chemical properties to enable us to
understand the birdwing’s food resource requirements;
University observation on courting and mating ofbehaviours Malaya at different elevation is essential
in order to help us understand the needed type of habitats for successful mating.
similar studies should be applied to other subspecies which may behave
differently in different habitat.
152
CHAPTER 6: DEVELOPMENT OF A NEW MONITORING METHOD AND
MONITORING PROGRAMME USING THREE METHODS FOR THE
TROGONOPTERA BROOKIANA ALBESCENS AND ITS RELATIONSHIP
WITH ENVIRONMENTAL PARAMETERS
6.1 Introduction
A monitoring programme is important to monitor the effectiveness of policy or
legislation, to monitor performance for management purposes and to detect incipient
change (Hellawell, 1991). Incipient change includes habitat degradation and restoration,
and also climate change.
Butterfly monitoring is quite well established, it is because butterflies are well-
documented, easy to recognise and popular with the general public (Harding et al.,
1995). Furthermore, short generation time, sensitivity to environmental change and
comparatively well studied taxonomy making them a good indicator for incipient
change (Thomas, 2005).
Butterfly monitoring is well established in European countries (van Swaay, 2014;
IBMN, 2015; PollardBase, 2016). In tropics, monitoring was recently started in
Panama, Thailand and Papua New Guinea (Basset et al., 2011). In Malaysia, from the
information available, there is no long term butterfly monitoring scheme. Perhaps it is
resulted from the lack of entomologists, insufficient and continuity of funding. In order
to minimise funding issues, the selection of bio-indicator and involvement of volunteers Universitywill probably succeed a long term monitoring. of Monitoring Malaya method used should be easy for the volunteers to follow as well.
The Trogonoptera brookiana is probably can be chosen as a bio-indicator among
other butterflies for several reasons, such as it is readily identify in the field, host
specific and only found in forest (Corbet & Pendlebury, 1992), sensitive to habitat
153
changes (Phon et al., 2011), and more importantly it is one of the conservation targets in
Malaysia (Ministry of Science, Technology and the Environment, 1997). Furthermore,
the population size and it distribution are decreased based to results in Chapter 4. Thus,
it is essential to monitor the changes for long term.
Up till today, no research has been conducted to monitor the population of the
Trogonoptera brookiana systematically and try to relate changes in the population to
habitat changes. There is a need to establish a monitoring protocol and to conduct a
monitoring programme that will enable the effects of weather patterns to be understood
and determine population trends over a longer period of time. Therefore, the objectives
of this study were:
1. to develop a new monitoring method on puddling Trogonoptera brookiana
albescens in Ulu Geroh village;
2. to conduct a trial monitoring for 24 months using the new method in Ulu Geroh
village;
3. to conduct monitoring of the Trogonoptera brookiana albescens in Kenaboi
Forest Reserve using simplified transect walk and single-point observation
methods for 30 months;
4. to relate the monitoring data to environmental parameters such as rainfall,
temperature, relative humidity and number of flowering plants flowered. University6.2 Literature Review of Malaya The two established methods to assess butterfly abundance, that is Pollard transect
walk and Mark-Release-Recapture (MRR) sampling (Nowicki et al., 2008) had its own
pros and cons. Pollard transect walk is less accuracy but it is cost efficient while MRR
is the other way round. Moreover, handling butterflies in MRR sampling had caused
change in flight path (Mallet et al., 1987) and lower recapture rate (Singer & Wedlake,
154
1981). However, it is not easy to study the negative effect to the butterflies in the
changes of butterfly behaviour and longevity in MRR sampling (Murphy, 1988). On the
other hand, Pollard transect walk is the most established monitoring protocol especially
in European countries (van Swaay, 2014) and it was also included as a global
monitoring protocol (van Swaay et al., 2015). However, it suffered detectability of
butterflies issue that caused by many factors for examples, diurnal and seasonal
distribution, weather conditions, representatives of transect routes, vegetation
succession, the ability of observers to detect species and species behavioural response to
population density (e.g. Harker & Shreeve, 2008; Pellet, 2008; Wikström et al., 2009;
Dennis et al., 2006).
Transect walk and MRR methods may not always suit the Trogonoptera brookiana
albescens. By using transect counts method, it may not truly represent the whole
population of the T. b. albescens because this birdwing were seldom seen in flight in the
forest except along river banks where the transect walk could be difficult to conduct and
it is not easy to spot the landing of birdwings against the background of evergreen
tropical forest. Preliminary study of using MRR method sampling was conducted in
September 2009 in Kuala Woh Recreational Forest, however, the number of birdwings
marked was very low and the capture activity disturbed the puddling behaviour of the
birdwings, which in turn affect the collection of data for the other objectives in this
study. Captures affected flight behaviour of the Rajah Brooke’s Birdwing (personal Universityobservation). After 3 to 4 catches of the birdwings,of otherMalaya birdwings eventually changed their flight path and avoiding the collector. This behaviour was mention in Mallet et al.
(1987) study.
155
6.3 Methodology
6.3.1 Development of a New Monitoring Method based on Counts of Puddling
Trogonoptera brookiana albescens
6.3.1.1 Study Site and Camera Set Up
The study was conducted in one of the few remaining sites where relatively large
numbers of male Trogonoptera brookiana albescens consistently puddle in single-
species groups. The puddling occurs along a river (Sungai Geroh) at the forest fringe
beside the very small indigenous-community village of Ulu Geroh (4° 26' 24.9" N, 101°
15' 01.8" E) in the Malaysian state of Perak. The village is planted with fruit trees and
flowering bushes interspersed among small wooden village houses, and is closely
flanked by the extensive Bukit Kinta Forest Reserve. Preliminary surveys identified two
main puddling sites about 30 m apart, which formed the basis for the study (Figures 6.1
and 6.2). A small puddling group occurred next to the larger of the two puddling groups
in the first three months of the study and was included. Very small occasional puddling
groups that were scattered much further away were excluded.
Visual field counts are impractical and unreliable when large numbers of butterflies
puddle. Therefore, we used digital photography to aid in obtaining counts. Two
cameras, a Canon EOS 5D Mark II with an EF 70–200 mm f / 2.8 L USM lens, and a
Nikon Coolpix P5100, were mounted on tripods 5–10 m from the two main puddling
sites and used to capture images of puddling birdwings. The EOS 5D Mark II, which Universitycaptures images of higher resolution, was usedof for theMalaya larger group. The distance of the cameras and their orientation were adjusted each day and month to frame all the
birdwings in each large puddling group and to obtain the best view. Two operators
communicated using walkie-talkies and triggered the cameras simultaneously at the
different puddles. The numbers of birdwings on the small transient puddle were small
156
and were counted visually. Counts of birdwings in the images were made with the aid of
the image analysis software, ImageJ 1.44p.
DSLR Camera
Birdwings Puddle
Figure 6.1: Camera setup to capture images on larger puddling group. Inset: An example of puddling birdwings.
Digital Compact Camera
University of Malaya
Figure 6.2: Camera setup to capture images on smaller puddling group. Inset: An example of puddling birdwings.
157
6.3.1.2 Comparison of Transect Counts and Puddling Counts
A comparison of the technique was made with transect counts that were conducted
near the puddling sites. Birdwings in flight were counted along a 180×20 m belt transect
by the same observer walking through once at a slow and steady pace for the first 15–20
minutes of each hour from 0900–1700 hrs. The location and direction of the transect
walk was such that the observer could avoid disturbing the two puddling sites. It began
near the largest puddle and moved downstream parallel and adjacent to the river, which
was up to about 20 m wide. This stretch of the river edge encompassed by the belt
transect was observed to have the most birdwings in flight, and the birdwings were
sufficiently large and distinctive to be recognised easily within the belt. Back and forth
flights of what was recognisably the same individual birdwing circling within or across
the belt transect were counted as a single sighting. Transect and puddling counts were
obtained for three high (January to March 2010) and three low population months
(September, October and December 2010) for three days each month or, due to
logistical limitations, two days on one of the months. Three photographs were taken one
minute apart at the start of every hour from 0900–1700 hrs. The counts of puddling
birdwings in the three images at the start of every hour were averaged to obtain hourly
averages. Hourly averages of birdwings on puddles and hourly counts of birdwings in
flight were averaged to obtain daily averages which were averaged to obtain the
monthly averages. Pearson’s correlation was used to examine the relationship between Universitythe monthly average numbers of birdwings ofpuddling andMalaya in flight. 6.3.1.3 Sample Requirements for Puddling Counts
To determine the sample requirements for counts of puddling birdwings, a few
component studies were undertaken. In the first component study, the optimum number
of consecutive images that was needed to obtain a reliable count of puddling birdwings
at any given hour was determined from two sets of images taken on a single day,
158
beginning 1100 hrs and 1500 hrs. Each set comprised 30 images taken a minute apart,
with no disturbances to the birdwings. The 30-image sets were divided into six
consecutive blocks of five images each. The averages for the first image, the first three
images, and all five images in each block were taken, and the averages of the six blocks
were compared for increasing numbers of images. Their degree of similarity was
examined using 95% confidence intervals. Based on the results, the average of three
images at the start of each hour was used for all subsequent tests. However, if there was
a disruption at the start of an hour, such as a sudden gust of wind that caused the
puddling birdwings to fly, the three images were taken 10 minutes later. The remaining
component studies utilised the same 6-month puddling dataset as the comparison
between transect counts and puddling counts. The second component study examined
the number of monitoring hours needed. To determine whether counts at a single hour
would be representative of counts throughout the day, the numbers of birdwings
puddling at 1600 hrs was correlated against the average numbers of birdwings puddling
from 0900–1600 hrs. Counts over a two- or three-day period each month were first
averaged for these time periods. The choice of 1600 hrs was based on an observed peak
puddling period from 1400–1600 hrs (Section 5.4.1.1). A third study analysed whether
the numbers of birdwings puddling at a chosen time of day in a monitoring programme
can be predicted from counts taken at an alternative time of day. This would be
important if, for example, rain, heavy cloud or circumstances prevent the count being
made at the chosen monitoring period. Correlation was used to analyse the relationship Universitybetween the numbers of birdwings puddling of at 1100Malaya hrs and 1600 hrs. A fourth component study analysed the number of monitoring days required for a representative
count. This was determined by inspecting the correlation matrix for the average counts
obtained on the first, second and third day at 1600 hrs. For the single month with only
159
two monitoring days, the third day’s count was treated as a missing value in the
correlation.
6.3.1.4 Application of the Technique
Monitoring was carried out for a period of two years from January 2010 to December
2011 using a protocol refined from the above-mentioned studies. Three images were
captured a minute apart beginning on the hour from 1400 to 1600 hrs for three days
each month, except for the first month and last three months in which two days were
used due to logistical limitations. The monitoring days were consecutive except on one
month in which the three sampling days occurred over a four-day period due to a day of
rain.
Environmental variables were recorded or scored 30 minutes into each monitoring
hour and averaged for each day. Temperature and relative humidity were recorded with
a single reading at each hour using a calibrated datalogger (Blue Gizmo BG-DL-01).
Brightness and rainfall were scored hourly on a scale of 1–3 (dull, moderately bright,
and very bright) and 0–2 (dry, drizzle, and light rain), respectively. Monitoring was not
carried out if there was heavy rain during the actual period of monitoring. In addition to
hourly scores, a separate score was given each day to generalise brightness and rainfall
from early morning to late evening (i.e. all-day brightness and all-day rainfall), using
the same scales as the hourly scores. A General Linear Model (GLM) in Minitab 15®
was used to determine whether environmental variables affected counts of puddling Universitybirdwings. Daily average counts were theof response Malaya variable, and month a random categorical variable with days nested within months. Daily environmental
measurements and scores were treated as covariates. A full analysis was conducted with
all variables, and those that were not significant at α = 0.05 were omitted sequentially to
obtain a final model.
160
6.3.1.5 Relationship between the Birdwings Population and the Monthly
Environmental Parameters
The effect of monthly weather parameters on the monthly population counts was
studied by using weather data collected from weather station at Ipoh, 4°34’N 101°06’E.
The distance between the weather station and the study site is about 22 km. Average
counts of puddling birdwings from 1400–1600 hrs every day were averaged for 2 to 3
days to get monthly count. The count was transformed using square root to get a better
fit on the normality line. A regression was applied for the analysis. In this analysis,
monthly mean temperature, relative humidity and rainfall were treated as independent
variables while the monthly average counts were treated as dependent variable. The
weather data was acquired from Malaysia Meteorology Department and it consists of
daily average of temperature, relative humidity and daily total rainfall. Daily averages
of temperature and relative humidity and daily total rainfall in each month were
averaged to get monthly mean for each parameter.
6.3.2 Monitoring using other Methods
There are some areas where the birdwings occur in flight but rarely puddle, therefore
the photography method developed cannot be applied. In such cases, transect walk and
single-point observation to monitor the birdwings in flight will be the alternatives.
Transect walk can be established at the area where there is no obstruction and the
birdwings always seen flying around. When there are obstruction, often at the Universityriverbanks, transect walk is hard to implement. of Therefore, Malaya single-point observation or several-point observation can be implemented near the birdwings flight paths. Single-
point observation was used in this study.
161
6.3.2.1 Study Site and Study Period
The study was conducted in Kenaboi Forest Reserve located in Negeri Sembilan
state from July 2012 to December 2014 excluding September 2012 (due to serious
logistic constrains), giving a monitoring total of 29 months.
6.3.2.2 Simplified Transect Walk
Transect belt of 400 x 20 meters along flying path of the birdwings were made beside
a river. The birdwings activity within the transect was considerably low because it has
overgrown forest in surrounding. This transect was chosen because it is long enough
and has no obstruction along the transect. Furthermore, the distance between the
transect and the single-point observation area was save enough to enable
communication using walkie-talkie between the two observers at two different and
isolated areas.
An observer walked within the transect belt twice representing 2 replicates to record
the number of birdwings in flight of each hour from 1000 to 1200 hrs. This period was
chosen because it was the peak flying activity of the birdwings in the study site (Section
5.4.1.2). The method of counting birdwings was similar in Section 6.3.1.2. The time for
the observer to finish recording each hour was about 20 to 25 minutes. Averages of each
hourly count were averaged and that reading represents the population count of
respective month. The study was conducted in one day each month.
University6.3.2.3 Single-point Observation of Malaya At the same time with simplified transect walk, the birdwings in flight was observed
at one observation point about 300 meters apart from the transect belt for 30 minutes
too. The station was beside a smaller river which is joined to the wider river beside the
belt transect. Within 30 minutes observation, the total number of the birdwings flying
162
through the observation point was recorded. Total numbers of the birdwings each hour
were averaged to get monthly population count.
6.3.2.4 Relationship between the Birdwings Population and Environmental
Parameters
In order to study the monthly weather effect on monthly population fluctuation,
monthly weather data from weather station near Hospital Jelebu (2°57’N 102°04’E) was
used. The distance between the weather station and the study site is about 27 km. The
data of each weather parameter was calculated similar to monthly weather data in Ulu
Geroh (Section 6.3.1.5). The number of flowering plants flowered (i.e. Bauhinia
bidentata var. breviflora and Saraca declinata) along the entrance route to the study site
(about 18.4 km long) was counted in each monitoring period and was included in the
analysis as well. Two regression analyses were conducted for simplified transect walk
and single-point observation respectively. The birdwing counts recorded from both
methods were transformed using logarithm based ten (log10) to fulfil the normality
assumption. Only records with full dataset were analysed, which means that dataset
with missing value in environmental parameters were excluded in analysis. Most of the
weather readings on in October and November 2014 were not recorded due to some
technical problems in the weather station (Madam Siti Fariza Mat Tahir from Malaysian
Meteorological Department officer, personal communication, July 15, 2015).
6.4 Results University6.4.1 Development of a New Monitoring of Method Malaya based on Counts of Puddling Trogonoptera brookiana albescens
During the development of new monitoring method, a total of 540 images were taken
each day and 1080–1620 images each month. A total of 9,180 images were generated
163
for the study. An example of image taken on the puddling birdwings is shown in Figure
6.3.
Figure 6.3: Puddling Trogonoptera brookiana albescens in study site.
6.4.1.1 Comparison of Transect Counts and Puddling Counts
Computer-assisted counts of large puddling birdwings in the digital images from
both types of cameras, and manual field counts of small numbers of puddling birdwings,
were all equally manageable and accurate because of the large size of the birdwings,
yielding actual numbers rather than estimates. Many birdwings were observed flying
out from the trees near the puddling sites in the early morning and flying back to rest on
these trees in the evening. Occasionally, when there were disturbances, such as people
passing by, birds landing nearby, a leaf fall, and gusts of wind, puddling birdwings
became unsettled and flew away, but remained nearby and eventually came back to rest
at the puddle. In one such case, when birds flew near to the puddle, all but five of the 20
puddling birdwings flew away and landed on trees nearby, but they returned to their Universityoriginal numbers at the puddle after 10 minutes.of RecoveryMalaya times were similar when other disruptive events occurred. No disruptions occurred during the comparison
between transect counts and puddling counts. There was a significant correlation
between the average number of birdwings in flight and the average number of birdwings
puddling from 0900–1700 hrs during each month’s sampling period (r=0.880, p=0.021;
Figure 6.4). Average counts of birdwings in flight numbered only about 5–15
164
individuals, whereas counts of puddling birdwings were much higher, between 21 and
120 individuals.
16 r = 0.880, p = 0.021 14
12
10
8
6
4
Number of birdwings in in birdwings flight of Number 2
0 0 20 40 60 80 100 120 140 Number of birdwings puddling Figure 6.4: Correlation of the average numbers of birdwings in flight and puddling in six different months.
6.4.1.2 Sample Requirements for Puddling Counts
The first of the four component studies to determine the sample requirements for
counts of puddling birdwings assessed the optimum number of consecutive images
needed to obtain a reliable count. Very little difference was found between the three sets
of six five-minute block averages and their 95% confidence intervals for subsamples of Universityone image or three images, or the full five of images (TableMalaya 6.1). However, there was a very slight increase in average counts with increasing numbers of images. In the second
component study, which aimed at determining the number of monitoring hours needed,
counts at just a single hour, i.e. 1600 hrs, were seen to be strongly correlated with the
average of hourly counts over an eight-hour period from 0900–1600 hrs (r=0.986,
p=0.000; Figure 6.5). And in the third study, which analysed whether the numbers of
165
birdwings puddling at a specific time of day could be predicted from counts made at
alternative time of day, the log transformed numbers of birdwings puddling at 1100 hrs
and 1600 hrs were highly and significantly correlated (r=0.984, p=0.000; Figure 6.6).
The fourth component study analysed the number of monitoring days needed. The
correlation matrix for average counts on the first, second and third day at 1600 hrs
showed that each day was correlated with each of the others, though not always
significantly (Table 6.2).
Table 6.1: Comparison of the number of consecutive images needed to obtain a reliable count of puddling birdwings, showing one-, three- and five-image averages of the counts and their 95% confidence intervals for two different times of the day.
Number of images per Time of day 5-minute block 1100 - 1130 hours 1500 - 1530 hours 1 image 95.3 102.7 (83.7–107.0) (92.1–113.2) 3 images 96.2 104.2 (84.5–107.8) (94.8–113.5) 5 images 97.0 105.1 (86.0–108.0) (96.4–113.9)
140 120 r = 0.986; p = 0.000 100 80 60 University–1600 hours 0900 40 of Malaya 20 0 0 20 40 60 80 100 120 140 160 180
1600 hours
Figure 6.5: Relationship between the monthly average numbers of birdwings puddling at 1600 hours and 0900–1600 hours in six different months.
166
3 r = 0.984; p = 0.000
2 1600 hours hours (Log10) 1600
1 1 2 3 1100 hours (Log10)
Figure 6.6: Correlation between counts at 1100 hrs and 1600 hrs.
Table 6.2: Correlation matrix for the first, second and third day of the monthly birdwing counts at 1600 hrs (Pearson’s correlation coefficient).
First day Second day Second day 0.943 (p = 0.005) Third day 0.722 0.831 (p = 0.168) (p = 0.081)
6.4.1.3 Application of the Technique
During the monitoring period, a total of 18 images were taken each day and 36–54
images each month. A total of 1,224 images were generated for the study. When Universitypuddling birdwings were monitored monthly of for two Malaya years from 1400 to 1600 hrs for two to three consecutive days each month, a high degree of population fluctuation was
observed. Numbers ranged from only several puddling birdwings at low season to 344
at the highest peak. The lowest population levels occurred in September of the first year
and December of the second year. In the first year of population monitoring, two peaks
occurred, the highest in May, followed by a much smaller peak in November (Figure
167
6.7). In the second year, there was a small peak in February followed by two much
higher peaks in May and August.
In the analysis for the effects of environmental variables, birdwing counts were
significantly associated with relative humidity (F=6.84, p<0.05, slope=2.48). In
addition, the association of birdwing counts with brightness was close to significance
(F=3.91, p<0.1, slope=21.41). There was no significant association with temperature
(F=0.70, p>0.4, slope=4.42), rainfall (F=0.23, p>0.6, slope=7.97), all-day rainfall
(F=0.41, p>0.5, slope=-4.91) or all-day brightness (F=0.00, p>0.9, slope=-0.21). When
non-significant terms were removed from the model, both humidity and brightness
showed a highly significant association (p<0.01; Table 6.3).
University of Malaya
168
15.00
10.00
Rainfall 5.00 (mean, mm) (mean, 0.00 90.00
80.00
RH (mean, RH%) (mean, 70.00 30.00
28.00
26.00 (mean, °C)(mean, Temperature Temperature 24.00 400 350 300 250 200 150
Birdwings count Birdwings 100 50 0 JFMAMJJASONDJFMAMJJASOND 2010 2011 Month Figure 6.7: The 24-month birdwing count fluctuation pattern (photography method) and weather fluctuation from weather station data. The error bar shows the 2-day or 3 day variation during the monitoring period. Dotted line in population count indicates moving average of two. Greyish area indicates the first six months of each year.
Table 6.3: General linear model for counts of puddling birdwing in response to monitoring month, relative humidity and brightness after removing non-significant terms (photography method).
Source DF Seq SS Adj SS Adj MS F P Month 22 461043 461368 20971 47.66 0.000 UniversityHumidity 1 1055 of 3514 Malaya 3514 7.98 0.007 Brightness 1 4385 4385 4385 9.96 0.003 Error 40 17601 17601 440 Total 64 484084 S = 20.9767; R-Sq = 96.36%; R-Sq (adj) = 94.18% SE of Regression term Coefficient TP coefficient Constant -92.57 63.38 -1.46 0.152 Humidity 1.8503 0.6548 2.83 0.007 Brightness 23.597 7.475 3.16 0.003
169
6.4.1.4 Relationship between the Birdwings Population and the Monthly
Environmental Parameters
Monthly birdwing counts were significantly associated with mean temperature
(F=5.680, p<0.05, slope=3.37; Table 6.4). There was no significant association with
mean relative humidity (F=0.060, p>0.05, slope=0.119) and mean rainfall (F=0.180,
p>0.05, slope=0.144). The adjusted R2 for the regression model was 20.09% (Table
6.4). After the non-significant variables, i.e. mean relative humidity and total rainfall
were removed, monthly birdwing counts were still significantly associated with mean
temperature (F=8.87, p<0.05, slope=2.896; Table 6.5). The adjusted R2 value increased
from 20.09% to 25.50%. Figure 6.7 shows the population and weather fluctuations.
Generally, when the mean temperature increased, the population increased as well and
vice versa. The mean temperature fluctuated between 26 and 29ºC throughout the two
years. The first six months in the first year did not show increasing pattern as happened
in the second year. A clear decreasing pattern was shown during the last six months in
both years. The mean relative humidity fluctuated between 75% and 85% and most of
the measurements were above 80%. The mean relative humidity recorded in the first
year did not seem to have a clear increase or decrease pattern compared to the second
year. In the second year, a decrease pattern in the first six months and an increase
pattern in the last six months were observed. Despite the very dry months in June and
July of the second year, the mean rainfall fluctuated between 5 mm to about 13 mm
during the monitoring period. Population peaked in certain months coincided with Universityrainfall peaks, i.e. May and November of theof first yearMalaya and August of the second year. The mean rainfall fluctuation pattern was similar to mean relative humidity, but there
wasn’t a clear pattern of increasing or decreasing in the first year. However, mean
rainfall was decreasing in the first six months and increasing in the last six months in
the second year.
170
Table 6.4: Linear regression model of all month physical parameters collected from weather station and monthly count of Trogonoptera brookiana albescens (photography method).
Analysis of Variance Source DF Adj SS Adj MS F-Value P-Value Regression 3 112.413 37.471 2.930 0.059 Temperature(Mean) 1 72.743 72.743 5.680 0.027 RH(Mean) 1 0.767 0.767 0.060 0.809 Avg_Rainfall 1 2.343 2.343 0.180 0.673 Error 20 255.957 12.798 Total 23 368.370
S = 3.58; R-sq = 30.52%; R-sq (adj) = 20.09%; R-sq(pred) = 9.05%
Regression Term Coef SE Coef T-Value P-Value VIF Constant -92.5 70.3 -1.31 0.204 Temperature(Mean) 3.37 1.41 2.38 0.027 1.97 RH(Mean) 0.119 0.487 0.24 0.809 2.76 Avg_Rainfall 0.144 0.335 0.43 0.673 1.62
Table 6.5: Linear regression model of mean temperature collected from weather station and monthly count of Trogonoptera brookiana albescens (photography method).
Analysis of Variance Source DF Adj SS Adj MS F-Value P-Value Regression 1 105.90 105.87 8.87 0.007 Temperature(Mean) 1 105.90 105.87 8.87 0.007 Error 22 262.50 11.93 Total 23 368.40 UniversityS = 3.45; R-sq = 28.74%; R-sq (adj) = 25.50%;of R-sq(pred)Malaya = 18.68% Regression Term Coef SE Coef T-Value P-Value VIF Constant -68.700 26.400 -2.600 0.016 Temperature(Mean) 2.896 0.972 2.980 0.007 1.000
171
6.4.2 Monitoring using Simplified Transect Walk and Single-point Observation
and the Relationship between the Birdwings Population and the
Environmental Parameters
6.4.2.1 Simplified Transect Walk
The birdwings count occurred in the transect was low throughout the whole
monitoring period. Figure 6.8 shows the population trend. During the period of January
to June in 2013 and 2014, the birdwings count was increasing. Whereas, in general, the
counts was decreasing between July and December in all years. The highest peak in
each year occurred in June 2013 and one month delayed in 2014. Other peaks occurred
in March, August and November 2013, and January, May and September 2014. In 2012,
a peak occurred in July and possibly another peak in October.
6.4.2.2 Single-point Observation
The birdwings count recorded at the observation point was higher than the simplified
transect walk method (Figure 6.8). Generally, the fluctuation pattern was not that great
compared to transect walk result. From July 2012 to May 2013, the counts were about
10 individuals and were constant throughout. The counts increased dramatically to
about 55 individuals in June 2013 then decreased sharply in July. The counts then
increased again to reach another smaller peak in August then decreased. There was
another small peak appeared in November 2013, about 10 individuals. In 2014, the
highest peak occurred one month earlier compared to 2013, which was recorded in May. UniversityThen the counts decreased until August thenof the counts Malaya increased very slightly towards the end of the year.
172
Figure 6.8: The 29-month birdwings population fluctuation using simplified transect Universitywalk and single-point observation methods. of The error Malaya bars shows the 3-hour variation of the count during the monitoring period. Dashed line in population count indicates moving average of two. The greyish area indicates period of July to December and the whitish area indicates period of January to June.
173
6.4.2.3 Relationship between the Birdwings Population and the Environmental
Parameters
There was no relationship found between the environmental parameters, the number
of flowering plants and the birdwings counts in both simplified transect walk method
and single-point observation method except the mean rainfall in single-point
observation (F=4.49, p<0.05, slope=-0.0508) (Table 6.6 and 6.7). However, after the
non-significant variables were removed, the relationship on mean rainfall and birdwing
counts was not significant (F=2.26, p>0.05, slope= -0.0296). The last six months of
each year had lower mean temperature compared to the first six months (Figure 6.8). It
fluctuated between 25.5°C to 27.0°C, approximately. The hottest and coolest month
happened in the first six months each year, from about 25.0°C to about 27.5°C. The
decreasing temperature in the last six months each year seems coincided with the
decreasing count pattern of the period while higher temperature and increasing
temperature pattern coincided with the increasing count pattern in the first six months
each year.
The driest months happened in January to March 2014, about 70% to 75%. The mean
relative humidity of other months fluctuated more or less between 75% and 90%. The
mean relative humidity increased then decreased in the first six months of 2014. In the
last six months each year, the mean relative humidity increased gradually. From January
to June 2013, the mean relative humidity did not fluctuate greatly, about 75% to 80%. UniversityHowever, during the same period in 2014, ofdriest months Malaya happened in this period. After March 2014, the mean relative humidity increased from about 75% to about 85%. The
relative humidity appeared not coincide with the birdwings counts.
174
The highest mean rainfall throughout 30 months was in November 2012. The lowest
mean rainfall recorded in July 2012, August 2013, January, February, June and August
2014. A negative correlation was found between mean rainfall and birdwing counts.
Many plants of Bauhinia bidentata var. breviflora flowered in October 2012, January
and December 2013, February and June 2014. While Saraca declinata flowered mostly
in February 2014, during the driest month. Despite of weather effect, the phenology of
the flowering plants had three to five months lag effect on the counts. Flowering season
in October 2012 and January 2013 caused the population peaked in March 2013 in
transect walk method and June 2013 in both methods, respectively. Flowering season in
December 2013 and February 2014, caused the population peaked in May and July
2014 in both methods, respectively. The higher number of flowering plants flowered in
June 2014 caused another peak in September of the same year in transect walk method.
Table 6.6: Linear regression model of all the physical parameters and the number of flowering plants recorded and birdwings count (simplified transect walk method).
Analysis of Variance Source DF Adj SS Adj MS F-Value P-Value Regression 4 1.6218 0.4054 1.62 0.204 No. of Flower 1 0.0013 0.0013 0.01 0.943 Temp_Mean 1 0.3430 0.3430 1.37 0.254 RH_Mean 1 0.0200 0.0200 0.08 0.780 Rainfall_Mean 1 0.7202 0.7202 2.88 0.104 Error 22 5.5016 0.2501 Total 26 7.1234
UniversityModel Summary of Malaya S = 0.5001; R-sq = 22.77%; R-sq(adj) = 8.72%; R-sq(pred) = 0.00%
Term Coefficients SE Coefficients T-Value P-Value VIF Constant -4.8400 6.6200 -0.73 0.473 No. of Flower -0.0012 0.0169 -0.07 0.943 1.05 Temp_Mean 0.2200 0.1880 1.17 0.254 1.28 RH_Mean 0.0101 0.0357 0.28 0.780 1.96 Rainfall_Mean -0.0582 0.0343 -1.70 0.104 1.62
175
Table 6.7: Linear regression model of all the physical parameters and the number of flowering plants recorded and birdwings count (single-point observation method).
Analysis of Variance Source DF Adj SS Adj MS F-Value P-Value Regression 4 0.9387 0.23467 1.91 0.144 No. of Flower 1 0.0378 0.03781 0.31 0.584 Temp_Mean 1 0.3675 0.36754 3.00 0.097 RH_Mean 1 0.3936 0.39355 3.21 0.087 Rainfall_Mean 1 0.5502 0.55017 4.49 0.046 Error 22 2.6983 0.12265 Total 26 3.6370
Model Summary S = 0.3502; R-sq = 25.81%; R-sq(adj) = 12.32%; R-sq(pred) = 0.00%
Term Coefficients SE Coefficients T-Value P-Value VIF Constant -8.3600 4.6400 -1.80 0.085 No. of Flower -0.0066 0.0118 -0.56 0.584 1.05 Temp_Mean 0.2280 0.1320 1.73 0.097 1.28 RH_Mean 0.0448 0.0250 1.79 0.087 1.96 Rainfall_Mean -0.0508 0.0240 -2.12 0.046 1.62
6.5 Discussion
6.5.1 New Monitoring Method
6.5.1.1 Comparison of Transect Counts and Puddling Counts
The significant correlation between the numbers of birdwings puddling and numbers
of birdwings in flight within a transect showed that the counts of puddling birdwings
corresponded with the more traditional method of a transect count. Transect counts are Universitywidely used in monitoring butterfly populations of (e.g. Malaya van Swaay, 2014; PollardBase, 2016) and determining butterfly species diversity (e.g. Basset et al., 2011; Majumder et
al., 2012), sometimes in conjunction with other methods such as fruit bait trapping (e.g.
Nganso et al., 2012). One of the best known transect counts is the Pollard walk (Pollard,
1977) in which the recorder counts butterflies within a moving 5×5×5m box in a belt
transect that may be divided into sections. The sum of the mean weekly count for the
176
whole transect is used as a population index. Since the birdwings are not frequently seen
in flight and are wide ranging in their flight patterns, but can be recognised from a
distance, we used a 20-metre-wide count zone with height reaching the tree canopies
and forward view as far as the eye could see. This maximised counts and at the same
time made it possible to exclude recounts of the same birdwing circling the transect.
Rarely was more than one birdwing seen in flight at any one time. Counts of birdwings
puddling were much higher in magnitude than counts of birdwings in flight. The former
would therefore be expected to be a more sensitive index of population size.
Like all methods of monitoring, puddling counts may be influenced to some degree
by extraneous factors other than population size. Over a longer period of monitoring
that examines trends over many years, these factors may be largely inconsequential to
the interpretation of population trends. Variations in the availability of the resource over
time and variations in attraction are the factors that could influence the counts of
birdwings at the puddle. However, since the puddle is caused by a geothermal spring, it
is stable year-round and unaffected by periods of dry weather. After rain, its geothermal
and chemical properties are restored by underground seepage. The significant, positive
correlation between the number of puddling birdwings counted on the ground and the
number of birdwings coundted in flight suggests that high and low numbers of puddling
birdwings are not a result of temporal changes in puddling behaviour such as changing
levels of resource attraction, but rather proportional to variations in population size.
University6.5.1.2 Sample Requirements for Puddling of Counts Malaya The first component study to determine the sample requirements for counts of
puddling birdwings demonstrated that, in the absence of events that might disrupt
puddling, just one image taken at each hour was sufficient to represent the numbers of
birdwings puddling for the hour, because there was little difference between the
177
averages and 95% confidence intervals when one, three or five images were used. The
very slight increase in average counts with increasing numbers of images is to be
expected, because the averages are centred one minute apart, and numbers of birdwings
were on an increase during the 30-minute time periods on which the analyses are based.
Where there is a disruptive event that causes the birdwings to take to flight, a delay of
about 10 minutes would be required to allow the birdwings to return before images are
captured.
The strong correlation between average birdwing counts at 1600 hrs and the average
of hourly counts over an eight-hour period (i.e., 0900–1600 hrs) in the second
component study showed that just one hour’s sample at this peak puddling period was
sufficiently representative of counts for the entire day. The results of the third study
showed that the numbers of birdwings puddling during this peak hour of 1600 hrs could
be predicted from the number of birdwings puddling at 1100 hours. Ideally, if a single,
standardised hour is chosen for the monitoring of the birdwing at this site, it should be
between 1400–1600 hours, which is what we observed to be the peak period for
puddling activity. However, the flexibility to predict a standardised hour’s count from
an alternative hour’s count is useful during seasons when heavy afternoon rain is
expected. Although in the fourth component study, the correlation between counts on
the first, second and third days was relatively high, it was not always significant,
indicating that there was a moderate amount of day to day variation in numbers of Universitypuddling birdwings within a month. of Malaya
6.5.1.3 Application of the Technique: Two-year Monitoring
In the 24-month monitoring data, brightness and humidity at the time of monitoring
had the most significant influence on day to day differences within a month, with higher
numbers of birdwings when it was bright and humid.
178
However, the population showed great variation over the two-year period, with three
peaks of about 250–350 puddling birdwings, and lows of less than 25 individuals. With
such high monthly variation, the day to day differences due to daily weather patterns
were relatively small. Variations in weather patterns and host plant phenology over the
years could be responsible for the population fluctuation. There is no obvious wet or dry
season in the geographical area in which the site is located, and weather patterns are not
always consistent from one year to another. All-year-round monitoring is therefore
preferable at this site.
One limitation of a population index based on puddling birdwings is that it excludes
females. In subspecies albescens, however, males are more often seen than females due
to their different habits. Sex ratio information is not available for this subspecies. Eliot
(in Corbet & Pendlebury, 1992) considered Wheeler’s (1940) estimate of one female to
twenty males (quoted also by Straatman & Nieuwenhuis, 1961), to be too high. In
subspecies trogon in Indonesia, the sex ratio was two males to one female among
specimens bred by Straatman and Nieuwenhuis (1961), and five males to four females
among specimens bred by Dahelmi et al. (2008). The sex ratio of collected specimens of
subspecies mollumar, which does not puddle in groups, was about equal (Corbet &
Pendlebury, 1992). Since males appear to be the more numerous sex in albescens, and
their numbers can be expected to correspond with that of females (albeit by an unknown
magnitude), monitoring its population based on an index of counts of puddling males Universitycan be expected to be a reliable method of monitoringof Malaya the overall population. 6.5.1.4 Application of the Technique: Developing Monitoring Programme
The success of a long-term monitoring programme often depends on volunteers.
Having a simple, rapid, practical and affordable monitoring protocol that require the
minimum effort to produce significant results encourages long-term volunteer
179
involvement (Pocock et al., 2014; Pocock et al., 2015). Therefore, the current study
focused not only on developing a reliable monitoring technique but also on
simplification of the monitoring protocol. The method described here can be easily
adopted by volunteers because of its simplicity. It is particularly suitable for monitoring
the Rajah Brooke’s Birdwing population in Ulu Geroh because the village has two
puddling sites that are stable and attract relatively large numbers of birdwings. Most of
the puddling birdwings can be photographed at these two sites, and a small number
outside these sites can either be counted visually or omitted. Before monitoring starts,
the puddles need to be observed for a short while to ensure that the birdwings are not
experiencing a disturbance or recovering from one.
One image taken for each puddle at one representative hour would be sufficient as an
indicator of the population. Three images taken at the start of the hour for three peak
hours in a day over a period of two or three days are still very manageable, and should
be considered for better population index accuracy in the monitoring programme.
Having the counts repeated on three different days, or at least two, to average out daily
variation due to weather conditions, is more important than replication within each day.
The method also yields counts that are not excessively affected by small variations in
weather. If it rains the whole day, counts can be conducted on a different day. When
there is a high likelihood of afternoon rains, usually from the months of March to June,
counts can be conducted in the morning and afternoon counts determined by prediction Universityfrom the regression line. of Malaya 6.5.1.5 Adapting the Method for Other Subspecies and Species
The method can also be used to monitor populations of albescens in other areas, and
can be used for another puddling subspecies, brookiana, which occurs in Borneo. A
good understanding of the behaviour of these birdwings is required at each site before
180
implementing or adapting the method. For example, in Peninsular Malaysia’s Kuala
Woh Recreational Forest, the birdwings puddle at scattered locations along the
riverbanks and may move. In this case, all the puddling birdwings should be sampled
rather than a fixed puddling location, but they should be within a fixed belt transect of
sufficient length to cover as many puddles as possible. In areas such as Kenaboi Forest
Reserve where the birdwings rarely puddle but are commonly seen in flight at specific
locations, a transect walk count or single-point observation count should be used. A
similar count strategy should be used for subspecies mollumar in the east of Peninsular
Malaysia (sometimes referred to as ssp. trogon), which is rarely seen puddling, but the
transects may need to be much longer as sightings of this subspecies are much less
frequent.
The method can also be adapted to monitor other puddling butterfly species,
including mixed species groups. While camera trapping is widely used to monitor larger
wildlife (O’Connell et al., 2011), and night photography has even been used to monitor
congregating fireflies in Malaysia (Kirton et al., 2012), photography of puddling
butterflies has yet to be used as a method for monitoring butterflies. In fact, puddles
have never been included in any butterfly monitoring protocol, perhaps partly because
of the difficulty of counting large numbers of puddling butterflies, and partly because
they are viewed as transient aggregations. However, advances in digital camera
technology have made it possible to count large numbers of puddling butterflies with Universityrelatively little cost, and if puddles are sampledof over Malaya a sufficiently long transect, they should provide a stable enough count, provided puddles that unnaturally increase counts
in the short term, for example because of urine or garbage, are excluded. Preliminary
tests similar to those conducted in this study need to be carried out before the method
can be used for monitoring mixed puddling groups.
181
6.5.2 Other Monitoring Methods
In Peninsular Malaysia, not many accessible areas have puddling activities of the
Rajah Brooke’s Birdwing. However, there are some areas where the birdwings in flight
can be easily seen, e.g. Kenaboi Forest Reserve in Negeri Sembilan state and Sungai
Sendat Recreational Forest in Selangor state. In order to monitor their population in
such area, counting the number of birdwings in flight is the only alternative. The study
was not attempt to develop a specific method to monitor the birdwings in flight.
Additionally, the study was not aim to compare method used for puddling birdwings
and in flight because monitoring puddling birdwings study was conducted in different
years and different locations.
By using three hours peak data with one day replicate within a month, a distinct
population fluctuation was resulted in both simplified transect walk and singe-spot
observation methods. Although the fluctuation patterns resulted from the two methods
were not completely similar but some of the major peaks were coincidentally happened
in the same months. However, patterns during the last six months were totally different.
Daily variation due to daily weather influence may occurred during the monitoring
period however the variation should be small when compared to monthly variation that
caused by seasonal variation as shown in monitoring protocol of puddling.
The number of the birdwings recorded was very low in the transect but yet a distinct
fluctuation was resulted. The highest count was around 6 individuals and lowest was Universityaround 1 individuals only. According to Thomasof (1983) Malaya in Spalding (1997), when the transect count below 40, the count are considered to be inaccurate. The low counts
resulted from the study may not inaccurate. It is probably due to the overgrown forest
along the transect that caused the birdwings seldom fly within the transect, although it
was placed beside the river. Low counts should be expected in other areas where the
182
population is low, e.g. in a visit to Sungai Sendat Recreational Forest in Selangor state,
a total of three individuals recorded in the whole afternoon when the day was sunny.
The population fluctuation from single-point observation method were not as great as
simplified transect walk but the magnitude was higher than the transect walk method.
Some degree of similarity on the population fluctuation patterns between two different
methods is expected and was proved in the study. This finding was supported by
Thomas (1991). In his study, when the monitoring sections were close together, which
was less than 600 m between midpoints of each section, the population densities in each
section tended to fluctuate in synchrony. The duplicate count was unlikely happen
during the monitoring period and this is supported by the capture-recapture trials. Two
occasions of capture-recapture were conducted near the study site before the monitoring
hours in two different months by one of the observers. About four to five birdwings
were marked successfully each time. None of the marked birdwings was recorded in the
monitoring study either in transect walk area or the single-spot observation area.
One day data taken during the peak hours should represent monthly population as
proved in photography method however it is ideal to replicate two to three days to
minimise the daily weather variation. The transect walk and single-point observation
methods should be used in areas when there is no puddling activity. The methods also
can be used for other birdwings, such as for subspecies mollumar in the east of
Peninsular Malaysia, which is rarely seen puddling. Harker and Shreeve (2008) Universitysuggested that selecting transects that are oftruly represent Malaya the vegetation structures and topographies for the habitat of butterflies are needed. Furthermore, they suggested that it
is also important to know the apparency of a butterfly species in relation to weather,
resource distribution and time of day before a monitoring programme can be
implemented. Due to the different butterflies will have different habitat especially
183
microhabitat, they suggested restricting individual site monitoring to specific species.
Therefore, selecting suitable transect(s) and observation point(s) are very important to
obtain a higher chance of detectability.
6.5.3 Seasonal Fluctuation of the Trogonoptera brookiana albescens
The population of the Trogonoptera brookiana albescens in Ulu Geroh increased
when the monthly mean temperature increased. Most of the environmental factors did
not influence the population fluctuation of the birdwings in Kenaboi Forest Reserve,
except that monthly mean rainfall was negatively affected population fluctuation in
single-point observation method. However, the low R2 values in both regressions
indicated the relationships were not good. The number of birdwings in Kenaboi Forest
Reserve increased when the flowering plants flowered three to five months before.
In the present study, especially in Ulu Geroh, temperature affected population
fluctuation significantly. An extreme high peaks of population happened in month-20 is
unexplainable. In Kenaboi Forest Reserve, population was higher when the temperature
was higher and rainfall was lower, and the other way round. This showed that the
population probably grew in the hotter and low rain season. However, some population
peaks coincided with rainfall peaks in Ulu Geroh which didn’t explain well with the
findings from Kenaboi Forest Reserve. Higher temperature and low rainfall resulted
more captures of from carrion baits (Checa et al., 2009) and warmer summer and lower
rainfall during the current and previous year resulted higher butterfly abundance too University(Roy et al., 2001). Previous months rainfall,of relativeMalaya humidity and temperature influenced Geometroid moth abundance especially high rainfall 3 months before caused
an increased of abundance (Intachat et al., 2001). A serious drought in a year caused
significant dropped in butterfly abundance in Sabah (Hill et al., 2003). However,
comparatively drier month in month-18 and -19 in Ulu Geroh didn’t seem to affect the
184
birdwing abundance. It was the same as in January and February 2014 in Kenaboi. In
Hill et al. studies, after a high rainfall, the butterfly abundance decreased. The results in
Ulu Geroh and Kenaboi were not consistent with the study.
The three to five months lag effect was observed between the flower plants
phenology and population fluctuation in Kenaboi Forest Reserve. The flower plants
phenology recorded in this study was Bauhinia bidentata var. breviflora and Saraca
declinata. Usually the birdwings were seen taking nectar from these two plants species
(see Chapter 5). Nectar plants play an important role in the life history of butterflies –
enhancing butterfly fecundity (Mevi-Schütz & Erhardt, 2005) and longevity (Murphy et
al., 1983). In a study conducted in Pasoh Forest Reserve in Negeri Sembilan state,
effects of plants phenology on Geometroid moths were examined (Intachat et al., 2001).
High flowering and flushing in the previous month and low flowering in the month
before that caused the abundance of the moths increased. However, lag effect was very
long in the present study. This flowering phenology factor probably has to associate
with host plant flushing season which the information was lacking. Host plants flushing
season has direct impact on the seasonal changes of species richness and abundance
(e.g. Owen et al., 1972; Yamamoto et al., 2007; Kunte, 1997; Valtonen et al., 2013).
The life history of the albescens in captivity recorded approximately 65 days, from eggs
incubation period to adult emergence (Goh, 1994). The lag effect found in the study
indicated that the birdwings took approximate one to three months to find their mates Universityand lay eggs in order to have an increased ofpopulation Malaya after two months. Survival of the birdwings in such a long period is unknown. Beck and Fiedler (2009) studied the traits
affected life span from literatures. They concluded that mud-puddling butterflies
associated with short life span (approximately 45 days, the mean, and all the figures
given here was transformed from log10 graph in their paper). The 95% confidence
interval recorded was about 35–55 days. In general nectar feeding butterflies had shorter
185
life span, 25 days with the 95% confidence interval very close to the mean. Liana
feeders had long life span, 45 to 95 days with the mean 65 days. The life span of
Papilionidae was inconsistent and tropical butterflies can live longer, 65 days with 50 to
100 days of 95% confidence interval (Beck & Fiedler, 2009). Therefore, there is a
possibility for the birdwings lived up to 3 months. However, it should be tested in the
field before any conclusion can be made. In general, leafing peaks of trees in Malaysia
occur between February to May and September to November (Kiew, 1986). This
coincided with most of the population fluctuation peaks in both Ulu Geroh and Kenaboi
Forest Reserve.
6.6 Conclusion
A new monitoring method has been developed from this study for the puddling
Trogonoptera brookiana albescens and this method was tested in Ulu Geroh village for
24 months. The study proved that puddling activity of the birdwings can be used as a
reliable population index. The minimum sample requirements were identified too in
order to be applicable for volunteers. The method has the potential to be applied for
other puddling butterflies. In the study, the daily puddling activity of the birdwings was
found to be more active when the weather is bright and humid. The study also showed
that the birdwings fluctuation had some level of relationship with the monthly mean
temperature (in Ulu Geroh) and mean rainfall (in Kenaboi Forest Reserve). The lag
effect between flowering seasons and birdwings fluctuation was observed too. The Universitystudy should be extended for a longer termof in order Malaya to study the population changes towards the changes on weather, climate and their natural habitat. Furthermore, other
biological factors such as birth dan death rates and migration of the birdwings should be
studied to understand the population dynamics of the birdwings. Nevertheless,
monitoring should not restrict to subspecies albescens but to include other subspecies
such as mollumar in Peninsular Malaysia which the population is much smaller than
186
albescens and too is threatening by rapid development in their habitat. The new
monitoring method developed in the study should be expanded by testing it in different
localities where the birdwings puddle. The potential to apply the method on other
puddling butterflies should be tested because most likely the method is able to generate
population indices for puddling butterflies which is now lacking.
University of Malaya
187
CHAPTER 7: GENERAL DISCUSSION
7.1 Synthesis
The taxonomy studies of Chapter 3 identified key characters to separate the
Peninsular and Sumatra subspecies of Trogonoptera brookiana based on morphology of
females. Consistency is lacking in order to separate the males. Based to the results, we
agree with D’Abrera to treat Peninsular subspecies as a valid subspecies, mollumar. The
diagnoses of subspecies mollumar and trogon given help scientists, researchers and the
stakeholders to identify which subspecies they are working on. The results enabled
studies on ecology and behaviour with the correct identity to be carried out in future.
The unusual high numbers of subspecies in the country should serve as evidence to the
government and stakeholders the need to conserve this birdwing in the country. The
urgency of conservation of this birdwing is also supported by the shrunken distribution
and reduction of number of birdwings sighted as concluded in Chapter 4, although a few
years of surveys are needed to confirm the results.
The behavioural studies in Chapter 5 revealed the daily activity rhythms of
subspecies albescens in two study sites, i.e. Ulu Geroh Village and Kenaboi Forest
Reserve. The correlation between the number of birdwings puddling and number of
birdwings in flight with temperature (positive correlation) and relative humidity
(negative correlation) was found in the study. Light intensity also found to be positively
correlated with the number of birdwings in flight in one of the study sites. Through this Universitystudy, the peak hours of the number of birdwingsof Malaya in flight and puddling have been identified and used in the monitoring studies in Chapter 6. Among the flowering plants
studied, albescens showed preference on Bauhinia bidentata var. breviflora and B.
audax compared to Hibiscus rosa-sinensis in Ulu Geroh Village. The birdwing showed
slight preference on Saraca declinata and Bauhinia bidentata var. breviflora over
hibiscus in Kenaboi Forest Reserve although the result was not significant. The study in
188
Ulu geroh also showed that the males and females of albescens nectared at different
time of the day, with highest activity in the morning to early afternoon for males and in
the evening for females. However, males and females in Kenaboi Forest Reserve did not
show distinct nectaring activity patterns. Several interesting behaviours were described
in this study. The male birdwings in Ulu Geroh Village exhibits micro-territorial
behaviour. The males’ exhibit calling other males to puddle or to leave puddle is an
interesting behaviour however difficult to quantify. Although the males in Ulu Geroh
Village like heat from the puddles but they do avoid direct heating from sunlight during
hotter hours. Besides that, the males exhibit early morning and night puddling in some
occasion’s throughtout the study period. The courting behaviour is interesting and it was
found that the male shivered during courting, although this was based on one time
observation. Competition between males to approach female was observed frequently in
Kenaboi Forest Reserve but less frequent in Ulu Geroh Village.
The last study (Chapter 6) monitored the birdwing populations using three methods,
i.e. a new photography method in Ulu Geroh Village and transect walk and single-point
observation in Kenaboi Forest Reserve. The two-year (in Ulu Geroh Village) and thirty-
month (in Kenaboi Forest Reserve) revealed distinct monthly fluctuations of population
indices. The fluctuations were to some extent influenced by temperature and rainfall.
The monitoring using photography method was not developed with the intention to
replace mark-release-recapture method or other established methods but to monitor Universitybutterflies that puddle in numbers. The method of is simple, Malaya rapid, practical and affordable that requires minimum effort to produce significant results which in turn encourages
long-term volunteer involvement. Distinct monthly fluctuation also suggest that
monthly monitoring over an extended period is needed for short- and long term
monitoring in order to understand the dynamics of the birdwings.
189
7.2 Aspects Requiring Futher Research
7.2.1 Genetics and Molecular
The present taxonomy study did not include genetic component due to the lack of
specimens from Sumatra. However, a total of 6 DNA sequences were found in Gene
Bank (https://www.ncbi.nlm.nih.gov/nuccore/?term=trogonoptera). A comparison was
made from the published sequences. Two specimens from Cameron Highlands
(subspecies albescens, AB044648 and AB084428) and one from Sumatra (subspecies
trogon, AB044647) was analysed using ND5 gene. Two specimens from Philippines
(Trogonoptera trojana, EF514444 and EF514436) and one from Indonesia (could be
subspecies trogon or brookiana, EF514439) were analysed using COI gene. Using ND5
gene and within a same subspecies, it was found that there are 3 base pairs (within 877
base pairs in total) different between Cameron Highlands specimens. However, between
subspecies, there are 2 to 3 base pairs different between Cameron Highlands specimens
and Sumatra specimens (within 813 base pairs in total). Using COI gene, 4 base pairs
(within 596 base pairs in total) are different between Philippines’ specimens and 6 to 8
base pairs are different between Philippines and Indonesia specimens (within 596 base
pairs). The Philippines and Indonesia specimens belong to different species. The
differences between subspecies or species were very small (between 0.2 to 1.3 %) using
ND5 gene and COI gene. Therefore futher research using other genetic markers with
non-destructive methods together with morphological method should be conducted on
all ten reported subspecies in order to have a full comparison. Because this birdwing is a Universityprotected species, therefore non-destructive of methods Malaya should be prioritised to minimise destruction on the specimens (Keyghobadi et al., 2009).
7.2.2 Distributions and Monitoring
The shrunk in distribution and the number of birdwings is noticeable yet continuous
monitoring is needed in order to examine the changes throughout the years as suggested
190
by Zonneveld et al. (2003). The maps generated can be further enhanced with vertical
distribution of the birdwings, host plants, preferred nectar plants, courting and mating
locations in order to mark the important areas for conservation. The host plant
distribution is still based on old collection records (Yao, 2015) and an update version is
very much needed.
The birdwing’s population should be monitored at a few sites continuously to
understand what factors trigger the changes (DeVries & Walla, 2001; Molleman et al.,
2006; DeVries et al., 2011). The distinct monthly fluctuation was found in the present
study however, the factor caused the fluctuation is still unknown. Therefore, the
seasonality study of the birdwings is too important. It is also interesting to study the
climatic changes effect on the birdwing’s population because of the different habitat of
both sexes particularly on subspecies albescens where the females prefer highlands
while males prefer lowlands, although this different habitat concept needs further
confirmation from vertical distribution study. The climatic changes may change the
distribution of butterflies (Forister et al., 2010; Zografou et al., 2014) and thus may
cause two sexes of the birdwing get closer to each other and increase the reproduction
activities, or conversely reduce the survival rate. This requires a study to confirm it.
7.2.3 Resources, Habitat and Behaviour
Although females are sighted more frequently in the highland but it doesn’t imply in
certain that the larval resources are more abundant at higher elevations compared to Universitylowland forest. Therefore, a systematic studyof on theMalaya host plant distribution is needed. Perhaps, surveys should be carried out at areas where the birdwings have been reported
in the study.
The reproductive biology of the birdwing is still not studied in detailed. Although the
life history of some subspecies has been reported (Straatman & Nieuwenhuis, 1961;
191
Goh, 1994; Chey, 1997; Igarashi & Fukuda, 1997; Dahelmi et al., 2008) but other
requirements are still unknown. The courting and mating habitats and habits of the
birdwings remain unexplored. Dispersal of both females and males in order for them to
meet, court and mate should be studied to investigate whether they are following certain
path in the forest.
Understanding the food resources is important in conservation (Ehrlich, 1984). Some
preferred nectar plants were identified in the present study. However, due to the logistic
issues and difficulty in getting flowers from canopy, studies on chemical properties of
the flowers cannot be conducted. Therefore, future study examines nectar quality,
quantity, production schedule and pollination strategy of the favourite plants is very
important. Nevertheless, puddle content, host plant chemical properties and its
pollination strategy should not be forgotten. In order to cultivate host plants and
preferred nectar plants in mass, the correct propagation techniques to yield a large
numbers of healthy plants should be explored.
Some special behaviour described in the study open a new research area for future
studies to explore in detail on the behaviours. The communal puddling of a single
species revealed the strong territorial behaviour of the birdwings. This territorial
behaviour is so prominent when the birdwings exhibit calling each other either to
puddle or to leave the puddle. We do not aware that such behaviour has been reported
before. The number of subspecies albescens that exhibits this territorial is far more Universityoutnumbered subspecies mollumar in the of Peninsular. Malaya This behaviour may cause the albescens establish more successfully than mollumar in the peninsula.
7.2.4 Other Subspecies
Subspecies mollumar in the peninsula and brookiana in Sabah and Sarawak received
less attention compared to albescens. Their presence should not be neglected and
192
studies such in this present project are needed. Furthermore, future studies should
include life history, resources requirement, behaviour, monitoring of both subspecies
and all the further studies suggested here. Once we have this information in hand, then a
complete conservation programme of all subspecies can only be developed by then.
7.3 Conservation
The subspecies albescens depends solely on Aristolochia foveolata during its
immature stages (Corbet & Pendlebury, 1992; Goh, 1994; Chey, 1997; Igarashi &
Fukuda, 1997; Dahelmi et al., 2008). When the host plant of other subspecies have been
identified, those host plant together with A. foveolata should be protected and enhanced
in order to sustain the populations (Fourcade & Öckinger, 2016). The favourite
nectaring plants should be enhanced too. Sufficient resources are the key aspects
towards the maintenance of the populations of the birdwings. If these butterflies need to
move e.g. from emergence site to puddling site, creation of corridors would be
important (Fourcade & Öckinger, 2016). Planting the larval host plants and adult
nectaring plants along the corridors would direct the birdwings in the desired direction.
We understand that the birdwings are exploited to gain income for certain
households especially the indigenous people (Kirton, 1991). The extent of exploitation
should be established in order to sustain the birdwing’s populations in the wild. The
indigenous people should be educated on the importance to conserve the birdwings.
Perhaps, after the populations have been sustained or increased by increasing the Universityresources of the birdwings, the indigenous ofpeople willMalaya not just depend on income from exploitation but also from eco-tourism activities such as butterflies’ watching .
193
CHAPTER 8: CONCLUSION
The present study is to some extent contributing information towards producing a
conservation programme especially for subspecies albescens yet many more studies are
needed in making sure the programme is successful. Both subspecies are endemic to
Malaysia, thus concerted conservation efforts are needed to ensure their survival.
The results in the taxonomy study reveal the distinguishable morphological
characters especially on female and therefore support the re-instatement of mollumar for
subspecific status. The historical and present distributions of both Peninsular Malaysia
subspecies were mapped using data collected from collections, literatures, interviews
and active surveys. The birdwings sighted or collected in localities reported previously,
were not sighted in the active surveys and the numbers of both subspecies sighted were
greatly reduced compared to previous records. The distribution of subspecies mollumar
was very restricted and the numbers were very low whereas subspecies albescens is not
protected under totally protected area. Therefore, immediate attention is needed to
conserve these subspecies. The rhythmic studies on puddling, flying and nectaring
behaviours, preferred nectar plants and courting behaviour of the subspecies albescens
were documented. Several habits that were not reported before were observed in the
study. A photography method was developed and tested to monitor puddling albescens
in Ulu Geroh Village for 24 months and the method was simplified, making it suitable
to be utilised for voluntary monitoring programme. The study proved that, puddling Universityactivity of the albescens can be used as a of reliable populationMalaya index. The photography method may be applicable for this birdwing or other butterflies at other locations with
little or no modification. This method can be used within a long transect with images
taken on multiple puddling spots that occur within a transect. The population in
Kenaboi Forest Reserve was monitored for 29 months using other methods, i.e.
194
simplified transect walk and single-point observation. The albescens in Ulu Geroh was
more active when the weather is bright and humid. And, the monthly temperature
affected positively on the monthly fluctuation of albescens in Ulu Geroh. Three
methods were used to monitor albescens population under different situations.
University of Malaya
195
REFERENCES
Abu Hassan, A., Mohd Fairus, J., Che Hanisah, I., & Che Salmah, M. R. (1998). A list of butterflies (Lepidoptera: Rhopalocera) from Kampung Bongor, a remote village near the Malaysia-Thailand border. Serangga, 3, 243-246.
Adler, L. S., & Irwin, R. E. (2012). Nectar alkaloids decrease pollination and female reproduction in a native plant. Oecologia, 168, 1033-1041.
Adler, P. H., & Pearson, D. L. (1982). Why do male butterflies visit mud puddles? Canadian Journal of Zoology, 60(3), 322-325.
Afik, O., Dag, A., & Shafir, S. (2006). The effect of avocado (Persea americana) nectar composition on its attractiveness to honey bees (Apis mellifera). Apidologie, 37, 317-325.
Alarcón, R., Riffell, J. A., Davidowitz, G., Hildebrand, J. G., & Bronstein, J. L. (2010). Sex-dependent variation in the floral preferences of the hawkmoth Manduca sexta. Animal Behaviour, 80, 289-296.
Alcock, J. (1988). The mating system of three territorial butterflies in Costa Rica. The Journal of Research on the Lepidoptera, 26, 89-97.
Arms, K., Feeny, P., & Lederhouse, R. C. (1974). Sodium, stimulus for puddling behavior by Tiger Swallowtail Butterflies, Papilio glaucus. Science, 185, 372- 374.
Azman, S., & Zaidi, M. I. (2007). Butterfly fauna (Lepidoptera: Rhopalocera) of Bukit Labohan, Ma'Daerah, Kemaman, Terengganu. In D. Sharma, A. W. Ahmad Zafir, B. Nadzirah, & A. A. B. Dayang Norwana (Eds.), Biodiversity expedition in Bukit Labohan, Ma'Daerah, Terengganu (pp. 44-49). Selangor, Malaysia: WWF-Malaysia.
Baioumy, H. M., Nawawi, M., Wagner, K., & Mohd Hariri, A. (2014). Geological setting and origin of non-volcanic hot springs in West Malaysia. In 3rd International Conference on Geological and Earth Sciences (GEOS 2014) (pp. 14-18). doi: 10.5176/2251-3353_GEOS14.37.
Baker, H. G. (1977). Non-sugar chemical constituents of nectar. Apidologie, 8, 349-356.
Baker, R. R. (1983). Insect territoriality. Annual Review of Entomology, 28, 65-89. University of Malaya Bąkowski, M., Filipiak, A., & Fric, Z. (2010). Foraging behaviour and nectar use in adult Large Copper Butterflies, Lycaena dispar (Lepidoptera: Lycaenidae). Entomologica Fennica, 21, 49-57.
Barton, M., Porter, W., & Kearney, M. (2014). Behavioural thermoregulation and the relative roles of convection and radiation in a basking butterfly. Journal of Thermal Biology, 41, 65-71.
196
Basset, Y., Eastwood, R., Sam, L., Lohman, D. J., Novotny, V., Treuer, T., ... Osorio- Arenas, M. A. (2011). Comparison of rainforest butterfly assemblages across three biogeographical regions using standardized protocols. The Journal of Research on the Lepidoptera, 44, 17-28.
Batchelor, D. M. (1959). Asahan butterflies. Malayan Nature Journal, 14, 1-7.
Batten, A. G. M., & Batten, A. M. (1970a). A month in Malaysia in the footsteps of Corbett and Pendlebury February - March, 1968. The Entomologist's Record and Journal of Variation, 82(1), 7-12.
Batten, A. G. M., & Batten, A. M. (1970b). A month in Malaysia in the footsteps of Corbett and Pendlebury February - March, 1968. The Entomologist's Record and Journal of Variation, 82(2), 44-48.
Baz, A. (2002). Nectar plant sources for the threatened Apollo butterfly (Parnassius apollo L. 1758) in populations of central Spain. Biological Conservation, 103, 277-282.
Beck, J., & Fiedler, K. (2009). Adult life spans of butterflies (Lepidoptera: Papilionoidea + Hesperioidea): Broadscale contingencies with adult and larval traits in multi-species comparisons. Biological Journal of the Linnean Society, 96, 166-184.
Beck, J., Mühlenberg, E., & Fiedler, K. (1999). Mud-puddling behavior in tropical butterflies: in search of proteins or minerals? Oecologia, 119, 140-148.
Bergman, M., Gotthard, K., Berger, D., Olofsson, M., Kemp, D. J., & Wiklund, C. (2007). Mating success of resident versus non-resident males in a territorial butterfly. Proceedings of the Royal Society B, 274, 1659-1665.
Bergman, M., Gotthard, K., & Wiklund, C. (2011). Mate acquisition by females in a butterfly: the effects of mating status and age on female mate-locating behaviour. Animal Behaviour, 81, 225-229.
Berwick, E. J. H. (1952). The distribution of Troides brookiana [Letter to the editor]. Malayan Nature Journal, 7, 40.
Blackburn, A. P. (1948). Brookiana swallowtail butterfly [Letter to the editor]. Malayan Nature Journal, 3, 214.
Boggs, C. L., & Dau, B. (2004). Resource specialization in puddling Lepidoptera. UniversityEnvironmental Entomology, 33, 1020-1024. of Malaya
Boggs, C. L., & Gilbert, L. E. (1979). Male contribution to egg production in butterflies: evidence for transfer of nutrients at mating. Science, 206(4414), 83- 84.
Boggs, C. L., & Jackson, L. A. (1991). Mud puddling by butterflies is not a simple matter. Ecological Entomology, 16, 123-127.
197
Bonebrake, T. C., Ponisio, L. C., Boggs, C. L., & Ehrlich, P. R. (2010). More than just indicators: A review of tropical butterfly ecology and conservation. Biological Conservation, 143, 1831-1841.
Bosi, S. G., Hayes, J., Large, M. C. J., & Poladian, L. (2008). Color, iridescence, and thermoregulation in Lepidoptera. Applied Optics, 47(29), 5235-5241.
Bowden, D. (2000). Visitor's guide Taman Negara Malaysia's Premier National Park. United Kingdom, Cape Town, Sydney, Auckland: New Holland Publishers (UK) Ltd.
Braby, M. F., Trueman, J. W. H., & Eastwood, R. (2005). When and where did troidine butterflies (Lepidoptera: Papilionidae) evolve? Phylogenetic and biogeographic evidence suggests an origin in remnant Gondwana in the Late Cretaceous. Invertebrate Systematics, 19, 113-143.
Brown, K. S. Jr. (1984). Adult-obtained pyrrolizidine alkaloids defend ithomiine butterflies against a spider predator. Nature, 309, 707-709.
Butler, A. G. (1879). The butterfllies of Malacca. Transactions of the Linnean Society of London, 1(2), 552.
Chalcoff, V. R., Aizen, M. A., & Galetto, L. (2006). Nectar concentration and composition of 26 species from the temperate forest of South America. Annals of Botany, 97, 413-421.
Chan, B. F. (1981). Nature photoscope: Photographing butterflies. Nature Malaysiana, 6, 10-11.
Chan, L. L. (2010). Habitats may soon vanish - Expert: Stop removal works or risk losing sanctuary. The Star, pp. N22.
Checa, M. F., Barragán, A., Rodríguez, J., & Christman, M. (2009). Temporal abundance patterns of butterfly communities (Lepidoptera: Nymphalidae) in the Ecuadorian Amazonia and their relationship with climate. Annales de la Société entomologique de France (N.S.): International Journal of Entomology, 45, 470- 486.
Chey, V. K. (1997). Rajah Brooke's Birdwing from a caterpillar to a king. Malaysian Naturalist, 5(2), 29-31.
Chiew, H. (2010). Damage to site a setback for conservation efforts. The Star, N22. University of Malaya CITES. (2017). Convention on International Trade in Endangered Species of Wild Flora and Fauna: Appendix II. Retrieved from https://www.cites.org/eng/app/appendices.php. Date of accessed: 22 December 2017.
Clench, H. K. (1966). Behavioral thermoregulation in butterflies. Ecology, 47, 1021- 1034.
198
Cole, S. (1998). A return to Fraser's Hill, Pahang, Malaysia. Bulletin of the Amateur Entomologists' Society, 57, 61-66.
Cole, S. (2000). Insects at Fraser's Hill and the Semangkop Pass, Malaysia. Bulletin of the Amateur Entomologists' Society, 59, 61-67.
Collins, N. M., & Smith, H. M. (1995). Threats and priorities in conserving Swallowtails. In J. M. Scriber, Y. Tsubaki, & R. C. Lederhouse (Eds.), Swallowtail butterflies: Their ecology and evolutionary biology (pp. 345-357). Gainesville, United States: Scientific Publishers.
Condamine, F. L., Toussaint, E. F. A., Clamens, A.-L., Genson, G., Sperling, F. A. H., & Kergoat, G. J. (2015). Deciphering the evolution of birdwing butterflies 150 years after Alfred Russel Wallace. Nature, Scientific Reports, 5(11860), 1-11.
Cook, D., Manson, J. S., Gardner, D. R., Welch, K. D., & Irwin, R. E. (2013). Norditerpene alkaloids concentrations in tissues and floral rewards of larkspurs and impacts on pollinators. Biochemical Systematics and Ecology, 48, 123-131.
Corbet, A. S., & Pendlebury, H. M. (1956). The butterflies of the Malay Peninsula (2nd ed.). Edinburgh and London, United Kingdom: Oliver and Boyd.
Corbet, A. S., & Pendlebury, H. M. (1992). The butterflies of the Malay Peninsula (4th ed., revised by J. N. Eliot). Kuala Lumpur, Malaysia: Malayan Nature Society.
Crane, J. (1955). Imaginal behavior of a Trinidad butterfly, Heliconius erato hydara Hewitson, with special reference to the social use of colour. Zoologica (N.Y.), 40, 167-196.
Cruden, R., Hermann, S., & Peterson, S. (1983). Patterns of nectar production and plant-pollinator coevolution. In B. Bentley, & T. Elias (Eds.), The biology of nectaries (pp. 80-125). New York, United States: Columbia University Press.
D'Abrera, B. (1975). Birdwing butterflies of the world. Melbourne, Australia: Lansdowne.
D'Abrera, B. (1982). Butterflies of the Oriental region (Part 1). Victoria, Australia: Hill House.
D'Abrera, B. (2003). Birdwing butterflies of the world. Victoria, Australia: Hill House Publishers Pty Ltd.
UniversityD'Abrera, B., Doggett, V., & Parker, N. (1976).of A newMalaya race of Trogonoptera brookiana Wallace (Lepidoptera: Papilionidae) from West Malaysia. The Victorian Naturalists, 93, 21-24.
Dahelmi. (2002). Life history and ecology of Papilionid butterflies of province of Sumatra Barat, Indonesia (Rep. No. 12). Indonesia: Pro Natura Foundation Japan and the Nature Conservation Society of Japan.
199
Dahelmi, Salmah, S., Abbas, I., Fitriana, N., Nakano, S., & Nakamura, K. (2008). Duration of immature stages of eleven swallowtail butterflies (Lepidoptera: Papilionidae) in West Sumatra, Indonesia. Far Eastern Entomologist, 182, 1-9.
Davis, A. K., Cope, N., Smith, A., & Solensky, M. J. (2007). Wing color predicts future mating success in male Monarch butterflies. Annals of the Entomological Society of America, 100, 339-344.
de Nicéville, L., & Martin, L. (1895). A list of the butterflies of Sumatra with especial reference to the species occurring in the north-east of the island. Journal of the Asiatic Society of Bengal, 64, 357-555.
De Worms, C. G. M. (1973). A brief collecting tour in Malaya, August 1972. The Entomologist's Record and Journal of Variation, 85, 205-210.
Dennis, R. L. H., Shreeve, T. G., Isaac, N. J. B., Roy, D. B., Hardy, P. B., Fox, R., & Asher, J. (2006). The effects of visual apparency on bias in butterfly recording and monitoring. Biological Conservation, 128, 486-492.
Dennis, R. L. H., Sparks, T. H., & Hardy, P. B. (1999). Bias in butterfly distribution maps: The effects of sampling effort. Journal of Insect Conservation, 3, 33-42.
Dennis, R. L. H., & Thomas, C. D. (2000). Bias in butterfly distribution maps: The influence of hot spots and recorder's home range. Journal of Insect Conservation, 4, 73-77.
Denton, S. W. (1889). Catching butterflies by means of decoys. The Canadian Entomologist, 21(6), 110-113.
DeVries, P. J., Alexander, L. G., Chacon, I. A., & Fordyce, J. A. (2011). Similarity and difference among rainforest fruit-feeding butterfly communities in Central and South America. Journal of Animal Ecology, 81, 472-482.
DeVries, P. J., & Walla, T. R. (2001). Species diversity and community structure in neotropical fruit-feeding butterflies. Biological Journal of the Linnean Society, 74, 1-15.
DeVries, P. J., Walla, T. R., & Greeney, H. F. (1999). Species diversity in spatial and temporal dimensions of fruit-feeding butterflies from two Ecuadorian rainforests. Biological Journal of the Linnean Society, 68, 333-353.
Distant, W. L. (1882–1886). Rhopalocera malayana: A description of the butterfly of Universitythe Malay Peninsula. London, United of Kingdom: Malaya West, Newman and Co.
Douwes, P. (1976). An area census method for estimating butterfly population numbers. Journal of Research on the Lepidoptera, 15, 146-152.
Ehrlich, P. R. (1984). The structure and dynamics of butterfly populations. In R. I. Vane-Wright, & P. R. Ackery (Eds.), The biology of butterflies (pp. 25-40). London, United Kingdom: Academic Press.
200
Eisner, T., & Meinwald, J. (1995). The chemistry of sexual selection. Proceedings of the National Academy of Sciences USA, 92, 50-55.
Ek-Amnuay, P. (2012). Butterflies of Thailand (2nd ed.). (Fascinating Insects volume 2). Bangkok, Thailand: Baan Lae Suan.
Eliot, J. N. (1941). Papilio brookiana sex ratio. The Entomologist, 74, 100.
Eliot, J. N. (1959). New or little known butterflies from Malaya. Bulletin of the British Museum (Natural History) Entomology, 7, 371-391.
Eliot, J. N. (1972). Rajah Brooke's Birdwing in Trengganu [Letter to the editor]. Malayan Nature Journal, 26, 175-176.
Eliot, J. N. (2006). Updating the butterflies of the Malay Peninsula. Edited, enlarged and prepared for publication by H. Barlow, R. Eliot, L. G. Kirton, & R. I. Vane- Wright. Malayan Nature Journal, 59, 1-49.
Felder, C., & Felder, R. (1864). Species Lepidopterorum hucusque descriptae vel iconibus expressae in seriem systematicam digestae. Verhandlungen Zoologisch- Botanischen Gesellschaft, 14, 289-378.
Fickert, C. (1889). Ueber die zeichnungsverhältnisse der gattung Ornithoptera. Zoologische Jahrbücher Abtheilung Fur Systematik, Geographie und Biologie Der Thiere, 4, 692-770.
Fleming, W. A. (1975). Butterflies of West Malaysia and Singapore (Vol. 1). Kuala Lumpur, Malaysia: Longman Malaysia Sdn. Berhad.
Forister, M. L., McCall, A. C., Sanders, N. J., Fordyce, J. A., Thorne, J. H., O’Brien, J., ... Shapiro, A. M. (2010). Compounded effects of climate change and habitat alteration shift patterns of butterfly diversity. PNAS, 107(5), 2088-2092.
Fourcade, Y., & Öckinger, E. (2016). Host plant density and patch isolation drive occupancy and abundance at a butterfly’s northern range margin. Ecology and Evolution, 7(1), 331-345.
Freeman, C. E., Worthington, R. D., & Jackson, M. S. (1991). Floral nectar sugar compositions of some South and Southeast Asian species. Biotropica, 23, 568- 574.
Freitas, A. V. L., Benson, W. W., Marini-Filho, O. J., & de Carvalho, R. M. (1995). UniversityTerritoriality by the dawn's early light:of The Malaya Neotropical owl butterfly Caligo idomenaeus (Nymphalidae: Brassolinae). Journal of Research on the Lepidoptera, 34, 14-20.
Gall, L. F. (1984). The effects of capturing and marking on subsequent activity in Boloria acrocnema (Lepidoptera: Nymphalidae), with a comparison of different numerical models that estimate population size. Biological Conservation, 28, 139-154.
201
Gall, L. F. (1985). Measuring the size of Lepidopteran populations. Journal of Research on the Lepidoptera, 24, 97-116.
Gilbert, L. E., & Singer, M. C. (1975). Butterfly ecology. Annual Review of Ecology, Evolution, and Systematics, 6, 365-395.
Goh, D. (1994). Life history of Trogonoptera brookiana albescens in Malaysia (Lepidoptera: Papilionidae). Tropical Lepidoptera, 5, 1-5.
González, A., Rossini, C., Eisner, M., & Eisner, T. (1999). Sexually transmitted chemical defense in a moth (Utetheisa ornatrix). Proceedings of the National Academy of Sciences USA, 96, 5570-5574.
Gosse, P. H. (1881). Ornithoptera Brookeana, Wall. (Description of female). The Entomologist, 14, 156-157.
Haddad, N. M., Hudgen, B., Damiani, C., Gross, K., Kuefler, D., & Pollock, K. (2008). Determining optimal population monitoring for rare butterflies. Conservation Biology, 22, 929-940.
Hagen, B. (1894). Verzeichniss der von mir auf Sumatra gefangenen Rhopalocera. Deutsche Entomologische Zeitschrift, 7, 1-41.
Hamer, K. C., Hill, J. K., Benedick, S., Mustaffa, N., Chey, V. K., & Maryati, M. (2006). Diversity and ecology of carrion- and fruit-feeding butterflies in Bornean rain forest. Journal of Tropical Ecology, 22, 25-33.
Hamer, K. C., Hill, J. K., Mustaffa, N., Benedick, S., Sherratt, T. N., Chey, V. K., & Maryati, M. (2005). Temporal variation in abundance and diversity of butterflies in Bornean rain forests: opposite impacts of logging recorded in different seasons. Journal of Tropical Ecology, 21, 417-425.
Hamidah, M., Chua, L. S. L., Suhaida, M., Yong, W. S. Y., & Kiew, R. (2011). Botanical gazetteer for Peninsular Malaysia. Kuala Lumpur, Malaysia: Forest Research Institute Malaysia.
Hancock, D. L. (1991). Notes on the phylogeny and biogeography of Ornithoptera Boisduval (Lepidoptera: Papilionidae). The Lepidopterological Society of Japan, 42, 17-36.
Harding, P. T., Asher, J., & Yates, T. J. (1995). Butterfly monitoring 1 — recording the changes. In A. S. Pullin (Ed.), Ecology and conservation of butterflies (pp. 3- University22). London, United Kingdom: Chapman of and MalayaHall.
Harker, R. J., & Shreeve, T. G. (2008). How accurate are single site transect data for monitoring butterfly trends? Spatial and temporal issues identified in monitoring Lasiommata megera. Journal of Insect Conservation, 12, 125-133.
Haugum, J., & Low, A. M. (1982). A monograph of the birdwing butterflies. The systematics of Ornithoptera, Troides and related genera (vol. 2, part 1. The genera Trogonoptera, Ripponia and Troides). Leiden, The Netherlands and Copenhagen, Denmark: E.J. Brill/Scandinavian Science Press Ltd.
202
Hellawell, J. M. (1991). Development of a rationale for monitoring. In B. Goldsmith (Ed.), Monitoring for conservation and ecology (pp. 1-14). London, United Kingdom: Chapman and Hall.
Hijmans, R. J., Guarino, L., & Marthur, P. (2012). DIVA-GIS Version 7.5 Manual. Retrieved from http://www.diva-gis.org/docs/DIVA-GIS_manual_7.pdf. Date of accessed: 31 July 2012.
Hill, C. J., & Pierce, N. E. (1989). The effect of adult diet on the biology of butterflies 1. The common imperial blue, Jalmenus evagoras. Oecologia, 81, 249-257.
Hill, J. K., Hamer, K. C., Dawood, M. M., Tangah, J., & Chey, V. K. (2003). Rainfall but not selective logging affect changes in abundance of a tropical forest butterfly in Sabah, Borneo. Journal of Tropical Ecology, 19, 35-42.
Hislop, J. A. (1950). Butterflies of Ophir. Malayan Nature Journal, 5, 9-16.
Hislop, J. W. (1949). A note on the Geometrid moth, Milionia, in Malaya, and further notes on Troides brookiana. Malayan Nature Journal, 4, 37-39.
Hokche, O., & Ramirez, N. (1990). Pollination ecology of seven species of Bauhinia L. (Leguminosae: Caesalpinioideae). Annals of the Missouri Botanical Garden, 77, 559-572.
Hoskins, A. (2015). Butterflies of Thailand, Malaysia and Borneo - Rajah Brooke's Birdwing - Trogonoptera brookiana Wallace, 1855. Retrieved from http://www.learnaboutbutterflies.com/Malaysia%20- %20Trogonoptera%20brookiana.htm. Date of accessed: 22 August 2015.
Hübner, J. (1819). Verzeichniss bekannter Schmettlinge verfasst, 6, 88.
Hughes, J. B., Daily, G. C., & Ehrlich, P. R. (1998). Use of fruit bait traps for monitoring of butterflies (Lepidoptera: Nymphalidae). Revista de Biologia Tropical, 46, 697-704.
IBMN. (2015). Illinois Butterfly Monitoring Network. Retrieved from http://bfly.org. Date of accessed: 27 November 2015.
Ide, J., & Kondoh, M. (2000). Male-female evolutionary game on mate-locating behaviour and evolution of mating systems in insects. Ecology Letters, 3, 433- 440.
UniversityIdris, A. B., Zaidi, M. I., & Mahadimenakbar, of M. Malaya D. (1999). Butterfly fauna of the Pasoh Forest Reserve, Negeri Sembilan, Malaysia: A preliminary survey. Serangga, 4, 35-51.
Igarashi, S., & Fukuda, H. (1997). The life histories of Asian butterflies (Vol. 1). Tokyo, Japan: Tokai University Press.
Ilse, D., & Vaidya, V. G. (1956). Spontaneous feeding response to colours in Papilio demoleus L. Proceedings of the Indian Academy of Science B, 43, 23-31.
203
Intachat, J., Holloway, J. D., & Staines, H. (2001). Effects of weather and phenology on the abundance and diversity of geometroid moths in natural Malaysian tropical rain forest. Journal of Tropical Ecology, 17(3), 411-429.
Jabatan Perhutanan Semenanjung Malaysia. (2003). Buku panduan hutan lipur Semenanjung Malaysia. Kuala Lumpur, Malaysia: Jabatan Perhutanan Semenanjung Malaysia.
Jordan, K. (1908). Family Papilionidae. In A. Seitz (Ed.), Macrolepidoptera of the world (9th ed.) (pp. 11-118). Paris, France: E. Le Moult.
Jumalon, J. N. (1967). Notes on Trogonoptera trojana of Palawan - Philippines. Journal of the Lepidopterists' Society, 21, 49-52.
Kadlec, T., Tropek, R., & Konvicka, M. (2012). Timed surveys and transect walks as comparable methods for monitoring butterflies in small plots. Journal of Insect Conservation, 16, 275-280.
Kemp, D. J. (2000). Contest behaviour in territorial male butterflies: does size matter? Behavioral Ecology, 11, 591-596.
Kemp, D. J., & Krockenberger, A. K. (2004). Behavioural thermoregulation in butterflies: The interacting effects of body size and basking posture in Hypolimnas bolina (L.) (Lepidoptera: Nymphalidae). Australian Journal of Zoology, 52, 229-236.
Kenyir Lake. (2016). Kenyir Dam. Retrieved from https://www.kenyirlake.com/kenyir_dam.htm. Date of accessed: 27 December 2016.
Kéry, M., & Plattner, M. (2007). Species richness estimation and determinants of species detectability in butterfly monitoring programmes. Ecological Entomology, 32, 53-61.
Keyghobadi, N., Crawford, L. A., & Maxwell, S. A. (2009). Successful analysis of AFLPs from non-lethal sampled wing tissues in butterflies. Conservation Genetics, 10, 2021.
Kiew, R. (1986). Phenological studies of some rain forest herbs in Peninsular Malaysia. Kew Bulletin, 41, 733-746.
Kingsolver, J. G. (1985a). Butterfly thermoregulation: organismic mechanisms and Universitypopulation consequences. Journal ofof Research Malaya on the Lepidoptera, 24, 1-20.
Kingsolver, J. G. (1985b). Thermal ecology of Pieris butterflies (Lepidoptera: Pieridae): a new mechanism of behavioral thermoregulation. Oecologia, 66(4), 540-545.
Kirby, W. F. (1896). A handbook to the order Lepidoptera (Part I ed., Vol. II). London, United Kingdom: Edward Lloyd, Limited.
Kirton, C. G., & Kirton, L. G. (1983). Notes on some butterflies from Peninsular Malaysia. Malayan Nature Journal, 36, 141-152.
204
Kirton, L. G. (1991). Butterflies in Peninsular Malaysia. Malayan Nature Society, 15, 105-110.
Kirton, L. G. (2014). A naturalist's guide to the butterflies of Peninsular Malaysia, Singapore and Thailand. Oxford, United Kingdom: John Beaufoy Publishing Limited.
Kirton, L. G., & Kirton, C. G. (1987). Butterflies of the Kuala Jasin Region, Ulu Endau, Johore, Malaysia. Malayan Nature Journal, 41, 365-377.
Kirton, L. G., Kirton, C. G., & Tan, M. W. (1990). Butterflies of Kuala Tahan Gunung Tahan in Taman Negara. The Journal of Wildlife and Parks, 10, 62-77.
Kirton, L. G., Nada, B., Khoo, V., & Phon, C.-K. (2012). Monitoring populations of bioluminescent organisms using digital night photography and image analysis: A case study of the fireflies of the Selangor River, Malaysia. Insect Conservation and Diversity, 5, 244-250.
Kobayashi, H. (1986). A new subspecies of Trogonoptera brookiana from Tuangku Island, N.W. Sumatra, Indonesia (Lepidoptera, Papilionidae). Transaction Himeji Natural History Association, 12, 10-12.
Kobayashi, H. (1991). A new subspecies of Trogonoptera brookiana Wallace, 1855, (Lepidoptera, Papilionidae) from Mentawei Islands. Transaction Himeji Natural History Association, 18, 48-49.
Kondo, K., Shinkawa, T., & Matsuka, H. (2003). Molecular systematics of birdwing butterflies (Papilionidae) infered from mitochondrial ND5 gene. Journal of the Lepidopterists' Society, 57, 17-24.
Krebs, R. A. (1988). The mating behavior of Papilio glaucus (Papilionidae). Journal of Research on the Lepidoptera, 26, 27-31.
Kunte, K. J. (1997). Seasonal patterns in butterfly abundance and species diversity in four tropical habitats in northern Western Ghats. Journal of Biosciences, 22, 593-603.
Lai-Fook, J. (1991). Absorption of phosphorus from the spermatophore through the cuticle of the bursa copulatrix of the butterfly, Calpodes ethlius. Tissue and Cell, 23(2), 247-259.
Lamarre, G. P. A., Molto, Q., Fine, P. V. A., & Baraloto, C. (2012). A comparison of Universitytwo common flight interception trapsof to survey Malaya tropical arthropods. ZooKeys, 216, 43-55.
Larsen, T. B. (2000). Hazards of butterfly collecting: Visiting the Flemings - Malaysia, 1975. The Entomologist's Record and Journal of Variation, 112, 135-137.
Lau, C. P. Y., Ramsden, L., & Saunders, R. M. K. (2005). Hybrid origin of "Bauhinia blakeana" (Leguminosae: Caesalpinioideae), inferred using morphological, reproductive, and molecular data. American Journal of Botany, 92, 525-533.
205
Launer, A. E., Murphy, D. D., Boggs, C. L., Baughman, J. F., Weiss, S. B., & Ehrlich, P. R. (1993). Puddling behavior by Bay Checkerspot butterflies (Euphydryas editha bayensis). Journal of Research on the Lepidoptera, 32, 45-52.
Laws of Malaysia. (2010). Act 716: Wild Life Conservation Act 2010. Malaysia: Government of Malaysia.
Lehnert, M. S., Emmel, T. C., & Garraway, E. (2013). Male-male interactions in the endangered Homerus Swallowtail, Papilio homerus (Lepidoptera: Papilionidae), in Jamaica. Caribbean Journal of Science, 47, 57-66.
Lindzey, S., & Connor, E. F. (2011). Monitoring the mission blue butterfly using immature stages. Journal of Insect Conservation, 15, 765-773.
Lohman, D. J., De Bruyn, M., Page, T., Von Rintelen, K., Hall, R., Ng, P. K. L., ... Von Rintelen, T. (2011). Biogeography of the Indo-Australian Archipelago. Annual Review of Ecology, Evolution, and Systematics, 42, 205-226.
Lomov, B., Keith, D. A., Britton, D. R., & Hochuli, D. F. (2006). Are butterflies and moths useful indicators for restoration monitoring? A pilot study in Sydney's Cumberland Plain Woodland. Ecological Management and Restoration, 7, 204- 210.
Lye, K. Y. (1979). Some notes on the butterflies of and butterfly collecting in the Pantai Acheh Forest Reserve. Malayan Naturalist, 33, 19-20.
Maimon, A., Norela, S., Idris, A. B., & Ng, Y. F. (2005). Insect diversity and abundance of Sungai Bebar, peat swamp forest in Pahang (Rep. No. PSF Technical Series No.4). Selangor, Malaysia: Peat Swamp Forest Project, UNDP/GEF Funded.
Majumder, J., Lodh, R., & Agarwala, B. K. (2012). Variation in butterfly diversity and unique species richness along different habitats in Trishna Wildlife Sanctuary, Tripura, northeast India. Journal of Species Lists and Distribution, 8, 432-436.
Malayan Nature Society. (1982). Butterfly news - Papillon factory at Batu Caves (Rep. No. 36 (1)). Kuala Lumpur, Malaysia: Author.
Malayan Nature Society. (1990). Pulau Redang State Park: Terrestrial resources, management problems and proposed landuse management guidelines. Kuala Lumpur, Malaysia: Author.
UniversityMalaysian Nature Society. (1994). Endau-Rompinof Malaya (Johor) National Park: A development and management plan. Kuala Lumpur, Malaysia: Author.
Malaysian Nature Society. (2002). Biodiversity audit on the viability of the proposed Tembila conservation area in northern Terengganu. Kuala Lumpur, Malaysia: Author.
Mallet, J., Longino, J. T., Murawski, D., Murawski, A., & Simpson de Gamboa, A. (1987). Handling effects in Heliconius: Where do all the butterflies go? Journal of Animal Ecology, 56, 377-386.
206
Manson, J. S., Rasmann, S., Halitschke, R., Thomson, J. D., & Agrawal, A. A. (2012). Cardenolides in nectar may be more than a consequence of allocation to other plant parts: a phylogenetic study of Asclepias. Functional Ecology, 26, 1100- 1110.
Martin, P., & Bateson, P. (1993). Measuring behaviour: An introductory guide. (2nd ed.). Cambridge, United Kingdom: Cambridge University Press.
Masters, A. R. (1990). Pyrrolizidine alkaloids in artificial nectar protect adult Ithomiine butterflies from a spider predator. Biotropica, 22(3), 298-304.
Matsuka, H. (2001). Natural history of birdwing butterflies. Tokyo, Japan: Matsuka Printing Co. (Checklist of Birdwing Butterfly authored by Ohya, T.).
Medway, L. (1965). The Ulu Gombak Field Studies Centre, University of Malaya. The Malayan Scientist, 2, 7-22.
Merckx, T., & Van Dyck, H. (2005). Mate location behaviour of the butterfly Pararge aegeria in woodland and fragmented landscapes. Animal Behaviour, 70, 411- 416.
Mevi-Schütz, J., & Erhardt, A. (2004). Mating frequency influences nectar amino acid preference of Pieris napi. Proceedings of the Royal Society B, 271, 153-158.
Mevi-Schütz, J., & Erhardt, A. (2005). Amino acids in nectar enhance butterfly fecundity: a long-awaited link. The American Naturalist, 165, 411-419.
Miller, J. S. (1987). Phylogenetic studies in the Papilioninae (Lepidoptera: Papilionidae). Bulletin of the American Museum of Natural History, 186, 356- 512.
Ministry of Science, Technology and the Environment. (1997). Malaysia: Country study on biological diversity. Assessment of biological diversity in Malaysia. Malaysia: Ministry of Science, Technology and the Environment.
Miyamoto, M., Mohd Parid, M., Noor Aini, Z., & Michinaka, T. (2014). Proximate and underlying causes of forest cover change in Peninsular Malaysia. Forest Policy and Economics, 44, 18-25.
Mohamed, S., Norela, S., & Ismail, S. (1990). Fauna kupu-kupu di sekitar kampus Universiti Kebangsaan Malaysia, Bangi. In A. Latiff (Ed.), Ekologi dan biologi Hutan Simpan Bangi (pp. 229-238). Bangi, Malaysia: Penerbit Universiti UniversityKebangsaan Malaysia. of Malaya
Mohd Fairus, J., Abu Hassan, A., & Che Hanisah, I. (1998). A list of butterflies (Lepidoptera: Rhopalocera) from Pulau Kajicuaca, Muda Reservoir, Kedah, Malaysia. Serangga, 3, 239-241.
Molleman, F., Grunsven, R. H. A., Liefting, M., Zwaan, B. J., & Brakefield, P. M. (2005). Is male puddling behaviour of tropical butterflies targeted at sodium for nuptial gifts or activity? Biological Journal of the Linnean Society, 86, 345-361.
207
Molleman, F., Kop, A., Brakefield, P. M., DeVries, P. J., & Zwaan, B. J. (2006). Vertical and temporal patterns of biodiversity of fruit-feeding butterflies in a tropical forest in Uganda. Biodiversity and Conservation, 15, 107-121.
Momose, K., Yumoto, T., Nagamitsu, T., Kato, M., Nagamasu, H., Sakai, S., ... Inoue, T. (1998). Pollination biology in a lowland dipterocarp forest in Sarawak, Malaysia. I. Characteristics of the plant-pollinator community in a lowland dipterocarp forest. American Journal of Botany, 85, 1477-1501.
Morton, A. C. (1984). The effects of marking and handling on recapture frequencies of butterflies. In R. I. Vane-Wright, & P. R. Ackery (Eds.), The biology of butterflies (pp. 55-58). London, United Kingdom: Academic Press.
Murphy, D. D. (1988). Are we studying our endangered butterflies to death? Journal of Research on the Lepidoptera, 26, 236-239.
Murphy, D. D., Launer, A. E., & Ehrlich, P. R. (1983). The role of adult feeding in egg production and population dynamics of the checkerspot butterfly Euphydryas editha. Oecologia, 56, 257-263.
Murphy, S. M. (2008). Notes on Papilio machaon aliaska (Papilionidae) populations near Fairbanks, AK. Journal of the Lepidopterists' Society, 62, 80-83.
Neukirchen, W. M. (1993). Trogonoptera brookiana mariae n. subsp., eine neue Subspecies von den Batu-Inseln, Indonesien (Lepidoptera: Papilionidae). Entomologische Zeitschrift, 103(18), 335-338.
Nganso, B. T., Kyerematen, R., & Obeng-Ofori, D. (2012). Diversity and abundance of butterfly species in the Abiriw and Odumante sacred groves in the eastern region of Ghana. Research in Zoology, 2, 38-46.
Nicolson, S. W., & Thornburg, R. W. (2007). Nectar chemistry. In S. W. Nicolson, M. Nepi, & E. Pacini (Eds.), Nectaries and nectar (pp. 215-264). Netherlands: Springer.
Noack, F. (2002). A checklist of butterflies of Perlis State Perak. Malayan Nature Journal, 56, 175-182.
Nobre, C. E. B., Iannuzzi, L., & Schlindwein, C. (2012). Seasonality of fruit-feeding butterflies (Lepidoptera, Nymphalidae) in a Brazilian semiarid area. International Scholarly Research Network, 2012, 268159.
UniversityNorela, S., Ade, S., & Maimon, A. of (2011). Malaya Fauna kupu-kupu (Lepidoptera: Rhopalocera) di Hutan Lipur Negeri Melaka. In A. R. Abd Rahman, H. L. Koh, A. Muhamad, W. K. Wan Yusof, & A. Latiff (Eds.), Siri kepelbagaian biologi Hutan. Hutan Simpan Melaka: Pengurusan hutan, persekitaran fizikal, kepelbagaian biologi dan sosio-ekonomi. Kuala Lumpur, Malaysia: Jabatan Perhutanan Semenanjung Malaysia.
Norela, S., Azman, S., & Ismanizam, I. (2007). Fauna kupu-kupu (Lepidoptera: Rhopalocera) di Taman Rimba Kenong. In M. Azahar, A. R. Azman, H. Che Hashim, H. L. Koh, A. Norhayati, & A. Latiff (Eds.), Siri kepelbagaian biologi
208
hutan. Taman Rimba Kenong, Pahang: Pengurusan, persekitaran fizikal dan kepelbagaian biologi (pp. 211-226). Kuala Lumpur, Malaysia: Jabatan Perhutanan Semenanjung Malaysia.
Norela, S., Azman, S., Saidah, M. S., Mohd Zabidi, Y., & Farizawati, S. (2008). Butterfly (Lepidoptera: Rhopalocera) fauna of Bukit Bauk Urban Forest. In U. Razani, H. L. Koh, R. Mohd Rahim, J. Na'aman, I. Faridah Hanum, & A. Latiff (Eds.), Siri kepelbagaian biologi hutan. Hutan Rimba Bandar Bukit Bauk, Terengganu: Pengurusan hutan, persekitaran fizikal dan kepelbagaian biologi (pp. 255-263). Kuala Lumpur, Malaysia: Jabatan Perhutanan Semenanjung Malaysia.
Norela, S., Azman, S., & Zaidi, M. I. (2006). Butterflies (Lepidoptera: Rhopalocera) Fauna of Gunung Jerai, Kedah. In M. I. Shaharuddin, W. A. Wan Yusoff, M. S. Jalil, M. Yusoff, & A. Latiff (Eds.), Siri kepelbagaian biologi hutan. Hutan Simpan Gunung Jerai, Kedah: Pengurusan, persekitaran fizikal dan kepelbagaian biologi (pp. 373-377). Kuala Lumpur, Malaysia: Jabatan Perhutanan Semenanjung Malaysia.
Norela, S., Dayang Hayanti, A. A., Maimon, A., & Mohd Zabidi, Y. (2008). Fauna kupu-kupu (Lepidoptera: Rhopalocera) di Hutan Simpan Bukit Belata. In M. Azahar, H. L. Koh, N. M. Nik Mohd Shah, N. Samsul Anuar, & A. Latiff (Eds.), Siri kepelbagaian biologi hutan. Hutan Bukit Belata, Selangor: Pengurusan, persekitaran fizikal, kepelbagaian biologi dan sosio-ekonomi (pp. 199-206). Kuala Lumpur, Malaysia: Jabatan Perhutanan Semenanjung Malaysia.
Norela, S., Nur Diana, M., & Maimon, A. (2004). Butterfly fauna (Lepidoptera: Rhopalocera) of Sungkai Wildlife Conservation Centre in Perak, Malaysia. The Journal of Wildlife and Parks, 22, 119-128.
Norela, S., Risda, F., Mohd Zabidi, Y., Suraya, S., & Maimon, A. (2011). Fauna Kupu- kupu (Lepidoptera: Rhopalocera) di Gunung Jasar dan Air Terjun Parit, Tanah Rata, Cameron Highlands. In A. R. Abd Rahman, H. L. Koh, K. Mohd Paiz, A. Muhamad, & A. Latiff (Eds.), Siri kepelbagaian biologi hutan. Hutan Pergunungan Cameron Highlands: Pengurusan hutan, persekitaran fizikal dan kepelbagaian biologi (pp. 227-239). Kuala Lumpur, Malaysia: Jabatan Perhutanan Semenanjung Malaysia.
Norela, S., Rosni, M., & Mohd Asri, A. B. (2013). Butterfly (Lepidoptera: Rhopalocera) diversity in Sungai Enam Basin, Temengor, Perak. In M. Abdul Latiff, K. Daniel Baskaran, K. Zainon, A. K. Zulfadhlan, I. Nurul Irdayu, N. R. Eza Feizaty, S. Alizah, M. Mohd Syaiful, H. Mohd Najmi, & Y. A. Ahmad Ridzuan University(Eds.), Proceedings of the 2nd Temengor of ScientificMalaya Expedition 2012 (pp. 261- 273). Petaling Jaya, Malaysia: Pulau Banding Foundation.
Nowicki, P., Bonelli, S., Barbero, F., & Balletto, E. (2009). Relative importance of density-dependent regulation and environmental stochasticity for butterfly population dynamics. Oecologia, 161, 227-239.
Nowicki, P., Settele, J., Henry, P.-Y., & Woyciechowski, M. (2008). Butterfly monitoring methods: The ideal and the real world. Israel Journal of Ecology and Evolution, 54, 69-88.
209
O'Connell, A. F., Nichols, J. D., & Karanth, K. U. (2011). Camera traps in animal ecology: Methods and analyses. Tokyo, Japan: Springer.
Orr, A. G. (1982). Observations on the flight periodicity of butterflies in West Malaysia. Journal of the Lepidopterists' Society, 36, 54-60.
Otis, G. W., Locke, B., McKenzie, N. G., Cheung, D., MacLeod, E., Careless, P., & Kwoon, A. (2006). Local enhancement in mud-puddling swallowtail butterflies (Battus philenor and Papilio glaucus). Journal of Insect Behavior, 19(6), 685- 698.
Owen, D. F., Owen, J., & Chanter, D. O. (1972). Seasonal changes in relative abundance and estimates of species diversity in a family of tropical butterflies. OIKOS, 23, 200-205.
Panchen, A. L. (1980). Notes on the behaviour of Rajah Brooke's Birdwing Butterfly, Trogonoptera brookiana brookiana (Wallace), in Sarawak. The Entomologist's Record and Journal of Variation, 92, 98-102.
Parrott, R. E. (1991). New Borneo (Kalimantan) subspecies of Troides and Trogonoptera (Lepidoptera: Papilionidae). Tropical Lepidoptera, 2, 122-136.
Parsons, M. J. (1996). Gondwanan evolution of the Troidine Swallowtails (Lepidoptera: Papilionidae): Cladistic reappraisals using mainly immature stage characters, with focus on the birdwings Ornithoptera Boisduval. Bulletin of the Kitakyushu Museum of Natural History, 15, 43-118.
Pasteur, D. (1972). More notes on butterfly collecting on Langkawi Island. Malayan Nature Journal, 25, 18-21.
Pellet, J. (2008). Seasonal variation in detectability of butterflies surveyed with Pollard walks. Journal of Insect Conservation, 12, 155-162.
Pellet, J., Bried, J. T., Parietti, D., Gander, A., Heer, P. O. C. D., & Arlettaz, R. (2012). Monitoring butterfly abundance: Beyond Pollard walks. PLOS ONE, 7, e41396.
Percival, M. (1965). Floral biology. Oxford, United Kingdom: Pergamon Press.
Phon, C.-K. (2008). Beautiful butterfly. The Rajah Brooke's Birdwing. FRIM in Focus, 6-7.
Phon, C.-K., & Kirton, L. G. (2009). Conservation of the Rajah Brooke's Birdwing, one Universityof Malaysia's natural wonders. Conservation of Malaysia,Malaya 10, 2-3.
Phon, C.-K., Kirton, L. G., & Wong, S. (2011). A glimmer of hope for the Rajah Brooke's Birdwing. Malaysian Naturalist, 65-1, 24-25.
Phuphathanaphong, L. (1987). Aristolochiaceae. In T. Smitinand, K. Larsen, & I. Nielsen (Eds.), Flora of Thailand (pp. 1-31). Bangkok, Indonesia: Suchada Chutima.
210
Pinheiro, C. E. G. (1990). Territorial hilltopping behavior of three Swallowtail butterflies (Lepidoptera, Papilionidae) in Western Brazil. Journal of Research on the Lepidoptera, 29, 134-142.
Pinto, C. F., Troncoso, A. J., Urzúa, A., & Niemeyer, H. M. (2009). Aristolochic acids affect the feeding behaviour and development of Battus polydamas archidamas larvae (Lepidoptera: Papilionidae: Troidini). European Journal of Entomology, 106, 357-361.
Pinto, C. F., Urzúa, A., & Niemeyer, H. M. (2011). Sequestration of aristolochic acids from meridic diets by larvae of Battus polydamas archidamas (Papilionidae: Troidini). European Journal of Entomology, 108, 41-45.
Pivnick, K. A., & McNeil, J. N. (1987). Puddling in butterflies: Sodium affects reproductive success in Thymelicus lineola. Physiological Entomology, 12(4), 461-472.
Pleasants, J. M., & Chaplin, S. J. (1983). Nectar production rates of Asclepias quadrifolia: Causes and consequences of individual variation. Oecologia, 59, 232-238.
Pocock, M. J. O., Chapman, D. S., Sheppard, L. J., & Roy, H. E. (2014). Choosing and using citizen science: a guide to when and how to use citizen science to monitor biodiversity and the environment. Wallingford, Oxfordshire, United Kingdom: Centre for Ecology and Hydrology.
Pocock, M. J. O., Newson, S. E., Henderson, I. G., Peyton, J., Sutherland, W. J., Noble, D. G., ... Roy, D. B. (2015). Developing and enhancing biodiversity monitoring programmes: A collaborative assessment of priorities. Journal of Applied Ecology, 52, 686-695.
Pola, M., & García-París, M. (2005). Marine puddling in Papilio polytes (Lepidoptera: Papilionidae). Florida Entomologist, 88, 211-213.
Pollard, E. (1977). A method for assessing change in the abundance of butterflies. Biological Conservation, 12, 115-132.
PollardBase. (2016). PollardBase. Retrieved from http://www.pollardbase.org/. Date of accessed: 17 December 2016.
Rawlins, J. E. (1980). Thermoregulation by the Black Swallowtail Butterfly, Papilio polyxenes (Lepidoptera: Papilionidae). Ecology, 61, 345-357. University of Malaya Rawlins, J. W. (1948). Brookiana swallowtail butterfly [Letters to the editor]. Malayan Nature Journal, 3, 215-216.
Rawlins, J. W. (1949). A further list of butterflies from Penang [Letters to the editor]. Malayan Nature Journal, 4, 41.
Regan, S. (1989). In search of the elusive "Raja Brooke". Malayan Naturalist, 42(4), 33-34.
211
Riffell, J. A., Alarcón, R., Abrell, L., Davidowitz, G., Bronstein, J. L., & Hildebrand, J. G. (2008). Behavioral consequences of innate preferences and olfactory learning in hawkmoth-flower interactions. Proceedings of the National Academy of Sciences USA, 105, 3404-3409.
Rippon, R. H. F. (1898). A monograph of the Papilionine tribe Troides of Hübner, or Ornithoptera [Bird-wing butterflies] of Boisduval (Vol. 1). London, United Kingdom.
Rippon, R. H. F. (1902). Genera insectorum. Bruxelles, Belgium: Wytsman, M.P.
Rippon, R. H. F. (1906). A monograph of the Papilionine tribe Troides of Hübner, or Ornithoptera [Bird-wing butterflies] of Boisduval (Vol. 2). London, United Kingdom.
Rosenberg, R. H., & Enquist, M. (1991). Contest behaviour in Weidemeyer's admiral butterfly Limenitis weidemeyerii (Nymphalidae): The effect of size and residency. Animal Behaviour 42(5), 805-811.
Rothschild, W. (1895). A revision of the Papilios of the eastern Hemisphere, exclusive of Africa. Novitates Zoologicae, 2, 167-463.
Rothschild, W. L. (1908). New oriental Papilionidae. The Entomologist, 41(536), 1-4.
Roy, D. B., Rothery, P., Moss, D., Pollard, E., & Thomas, J. A. (2001). Butterfly numbers and weather: Predicting historical trends in abundance and the future effects of climate change. Journal of Animal Ecology, 70, 201-217.
Rusterholz, H.-P., & Erhardt, A. (2000). Can nectar properties explain sex-specific flower preferences in the Adonis Blue butterfly Lysandra bellargus? Ecological Entomology, 25, 81-90.
Scott, J. A. (1972). Mating of butterflies. Journal of Research on the Lepidoptera, 11, 99-127.
Shuel, R. W. (1957). Some aspects of the relationship between nectar secretion and nitrogen, phosphorous and potassium nutrition. Canadian Journal of Plant Sciences, 37, 220-236.
Singer, M. C., & Wedlake, P. (1981). Capture does affect probability of recapture in a butterfly species. Ecological Entomology, 6(2), 215-216.
UniversitySkertchly, S. B. J. (1889). On the habits ofof certain BorneanMalaya butterflies. The Annals and Magazine of Natural History, including Zoology, Botany and Geology, 4, 209- 218.
Smedley, S. R., & Eisner, T. (1996). Sodium: A male moth's gift to its offspring. Proceedings of the National Academy of Sciences USA, 93, 809-813.
Snellen, P. C. T., & Snelleman, F. (1892). Lepidoptera. Natuurlijke Historie, 4, 4-92.
212
Sofian-Azirun, M., Khaironizam, M. Z., Norma-Rashid, Y., & Daicus, B. (2005). Butterflies (Insecta: Lepidoptera) of the southwestern Endau-Rompin National Park, Johor, Malaysia. In H. Mohamed, & M. Zakaria-Ismail (Eds.), The forests and biodiversity of Selai, Endau-Rompin (pp. 169-175). Kuala Lumpur, Malaysia: University of Malaya.
Soong, M. K. (1948). Brookiana swallowtail butterfly [Letter to the editor]. Malayan Nature Journal, 3, 214.
Southwood, T. R. E. (1971). Absolute population estimates using marking techniques. In Ecological methods - with particular reference to the study of insect populations (pp. 57-98). London, United Kingdom: Chapman and Hall.
Spalding, A. (1997). The use of the butterfly transect method for the study of the nocturnal moth Luperina nickerlii leechi Goater (Lepidoptera: Noctuidae) and its possible application to other species. Biological Conservation, 80, 147-152.
Srygley, R. B. (1994). Shivering and its cost during reproductive behaviour in Neotropical owl butterflies, Caligo and Opsiphanes (Nymphalidae: Brassolinae). Animal Behaviour, 47(1), 23-32.
Staudinger, O. (1889). Lepidoptera der Insel Palawan. Deutsche entomologische Zeitscrift, 2(1), 3-180.
Straatman, R. (1955). Notes on methods of collecting Indo-Australian Lepidoptera. The Lepidopterists' News, 9, 74-76.
Straatman, R., & Nieuwenhuis, E. J. (1961). Biology of certain Sumatran species of Atrophaneura, Trogonoptera, and Troides (Lepidoptera, Papilionidae). Tijdschrift voor Entomologie, 104, 31-41.
Stubbs, G. (1950). Further note on Troides brookiana. Malayan Nature Journal, 5, 31- 32.
Sunday, J. M., Bates, A. E., Kearney, M. R., Colwell, R. K., Dulvy, N. K., Longino, J. T., & Huey, R. B. (2014). Thermal-safety margins and the necessity of thermoregulatory behavior across latitude and elevation. Proceedings of the National Academy of Sciences, 111, 5610-5615.
Swihart, C. A., & Swihart, S. L. (1970). Colour selection and learned feeding preferences in the butterfly, Heliconius charitonius Linn. Animal Behaviour, 18, 60-64. University of Malaya Takeuchi, T., & Imafuku, M. (1999). Territorial behavior of Favonius taxila (Lycaenidae): territory size and persistency. Journal of Research on the Lepidoptera, 38, 59-66.
Tan, M. W., Kirton, L. G., & Kirton, C. G. (1990). Composition and distribution of butterflies in Rompin-Endau, especially along Sungai Kinchin and its vicinity. In S. K. Yap (Ed.), In harmony with nature: Proceedings of the International Conference on Tropical Biodiversity, Kuala Lumpur (pp. 193-212). Malayan Nature Journal, 45(1-4).
213
Tharmalingam, M. (2009). Trip Report: Malaysia and Borneo: Mammals of Tropical Asia. South Africa: INDRI Ultimate Mammal Voyages.
Thomas, C. D. (1991). Spatial and temporal variability in a butterfly population. Oecologia, 87, 577-580.
Thomas, J. A. (1983). A quick method of estimating butterfly numbers during surveys. Biological Conservation, 27, 195-211.
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.
Thomson, J. D., Draguleasa, M. A., & Tan, M. G. (2015). Flowers with caffeinated nectar receive more pollination. Arthropod-Plant.Interactions, 9, 1-7.
Thomson, J. D., McKenna, M. A., & Cruzan, M. B. (1989). Temporal patterns of nectar and pollen production in Aralia hispida: Implications for reproductive success. Ecology, 70, 1061-1068.
Tiedge, K., & Lohaus, G. (2017). Nectar sugars and amino acids in day- and night- flowering Nicotiana species are more strongly shaped by pollinators' preferences than organic acids and inorganic ions. PLOS ONE, 12, e0176865.
Tilley, R. J. D., & Eliot, J. N. (2002). Scale microstructure and its phylogenetic implications in lycaenid butterflies (Lepidoptera, Lycaenidae). Transactions of the Lepidopterological Society of Japan, 53, 153-180.
Tiple, A. D., Khurad, A. M., & Dennis, R. L. H. (2009). Adult butterfly feeding-nectar flower associations: Constraints of taxonomic affiliation, butterfly, and nectar flower morphology. Journal of Natural History, 43, 855-884.
Tsukada, E., & Nishiyama, Y. (1982). Butterflies of the South East Asian Islands (Vol. 1, Papilionidae). Tokyo, Japan: Plapac Co Ltd.
Tufto, J., Lande, R., Ringsby, T.-H., Engen, S., Sæther, B.-E., Walla, T. R., & DeVries, P. J. (2012). Estimating Brownian motion dispersal rate, longevity and population density from spatially explicit mark-recapture data on tropical butterflies. Journal of Animal Ecology, 81, 756-769.
Tung, A. V. W. Y. (1995). Biodiversity of invertebrates in Belum. Kuala Lumpur, Malaysia: Malaysian Nature Society. University of Malaya Tung, A. V. W. Y. (2003). Two aberrant forms of the Rajah Brooke's Birdwing. Malayan Naturalist, 57(1), 48-49.
Tyler, P. A. (1962). Collecting butterflies in Malaya. The Bulletin of the Amateur Entomologists' Society, 21, 47-50.
UNEP-WCMC (2007). Review of trade in ranched birdwing butterflies. Cambridge, United Kingdom: Author.
214
UNEP-WCMC (2012). Review of butterflies from Asia and Oceania subject to long- standing positive opinions. Cambridge, United Kingdom: Author.
Valtonen, A., Molleman, F., Chapman, C. A., Carey, J. R., Ayres, M. P., & Roininen, H. (2013). Tropical phenology: Bi-annual rhythms and interannual variation in an Afrotropical butterfly assemblage. Ecosphere, 4, 36.
van Swaay, C. A. M. (2014). Tracking butterflies for effective conservation (Doctoral thesis, Wageningen University, The Netherlands. Retrieved from http://www.netwerkecologischemonitoring.nl/downloads/292/proefschrift%20va n%20Swaaij.pdf.
van Swaay, C. A. M., Regan, E., Ling, M., Bozhinovska, E., Fernandez, M., Marini- Filho, O. J., ... Underhill, L. (2015). Guidelines for Standardised Global Butterfly Monitoring (Rep. No. GEO BON Technical Series 1). Leipzig, Germany: Group on Earth Observations and Biodiversity Observation Network.
Vollenhoven, S. C. S. V. 1860. Description d'une nouvelle espèce de Lépidoptère du Genre Papilio, suivie d'une énumération des espèces de ce Genre, qui se trouvent au Musée Royal des Paysbas à Leide. Tijdschrift voor Entomologie, 3, 67-89.
Wallace, A. R. (1855). Description of a new species of Ornithoptera. Transactions of the Entomological Society of London, 2, 104-105.
Walpole, M. J., & Sheldon, I. R. (1999). Sampling butterflies in tropical rainforest: An evaluation of a transect walk method. Biological Conservation, 87, 85-91.
Warren, M. S. (1990). The conservation of Euphydryas aurinia in the United Kingdom, in Colloquy on the Berne Convention on invertebrates and their conservation (Rep. No. 10). Strasbourg: Council of Europe.
Watanabe, M., & Kamikubo, M. (2005). Effects of saline intake on spermatophore and sperm ejaculation in the male swallowtail butterfly Papilio xuthus (Lepidoptera: Papilionidae). Entomological Science, 8(2), 161-166.
Watt, W. B., Hoch, P. C., & Mills, S. G. (1974). Nectar resource use by Colias butterflies: Chemical and visual aspects. Oecologia, 14(4), 353-374.
Wedell, N. (2005). Female receptivity in butterflies and moths. The Journal of Experimental Biology, 208, 3433-3440.
UniversityWeir, L. K., Grant, J. W. A., & Hutchings, of J. A. (2011).Malaya The influence of operational sex ratio on the intensity of competition for mates. The American Naturalist, 177, 167-176.
Weiss, M. R. (1997). Innate colour preferences and flexible colour learning in the pipevine swallowtail. Animal Behaviour, 53, 1043-1052.
Westerman, E. L., Drucker, C. B., & Monteiro, A. (2014). Male and female mating behaviour is dependent on social context in the butterfly Bicyclus anynana. Journal of Insect Behavior, 27(4), 478-495.
215
Wheeler, L. R. (1940). On the alleged rarity of certain Papilio females in Malaya. Entomologist, 73, 269-274.
Whitmore, T. C. (1972). Rajah Brooke's Birdwing in Trengganu [Letter to the editor]. Malayan Nature Journal, 25, 41.
Wiklund, C. (2003). Sexual selection and the evolution of butterfly mating systems. In C. L. Boggs, W. B. Watt, & P. R. Ehrlich (Eds.), Butterflies: Ecology and evolution taking flight (pp. 67-90). Chicago, United States and London, United Kingdom: The University of Chicago Press and The University of Chicago Press, Ltd.
Wikström, L., Milberg, P., & Bergman, K.-O. (2009). Monitoring of butterflies in semi- natural grasslands: diurnal variation and weather effects. Journal of Insect Conservation, 13, 203-211.
Wilcocks, P. H. J. (1969). Butterfly collecting on Pulau Langkawi. Malayan Nature Journal, 23, 15-20.
Yamamoto, N., Yokoyama, J., & Kawata, M. (2007). Relative resource abundance explains butterfly biodiversity in island communities. Proceedings of the National Academy of Sciences USA, 104, 10524-10529.
Yao, T. L. (2015). Aristolochiaceae. In R. Kiew, R. C. K. Chung, L. G. Saw, & E. Soepadmo (Eds.), Flora of Peninsular Malaysia (pp. 5-46). Selangor, Malaysia: Forest Research Institute Malaysia.
Yong, D. L., Lohman, D. J., Gan, C. W., Qie, L., & Lim, S. L-H. (2012). Tropical butterfly communities on land-bridge islands in Peninsular Malaysia. The Raffles Bulletin of Zoology, 25, 161-172.
Zaidi, M. I., & Abin, S. (1995). Butterfly fauna of Krau Wildlife Reserve, Pahang: A preliminary survey and assessment. The Journal of Wildlife and Parks, 14, 71- 82.
Zaidi, M. I., Azman, S., Norela, S., & Poh, L. S. (2005a). Butterfly (Lepidoptera: Rhopalocera) fauna of Gunung Stong Forest Reserve. In M. I. Shaharuddin, H. T. Dahalan, S. Abdullah Sani, M. S. Jalil, I. Faridah-Hanum, & A. Latiff (Eds.), Siri kepelbagaian biologi hutan. Taman Negeri Gunung Stong, Kelantan: Pengurusan, persekitaran fizikal, biologi dan sosio-ekonomi (pp. 168-179). Kuala Lumpur, Malaysia: Jabatan Perhutanan Semenanjung Malaysia.
UniversityZaidi, M. I., Azman, S., Norela, S., & Wong,of C. Y.Malaya (2005b). Butterfly (Lepidoptera: Rhopalocera) fauna of Ulu Muda Forest Reserve. In M. I. Shaharuddin, H. Che Hashim, D. Mohd Puat, M. S. Jalil, Y. Norhaidi, & A. Latiff (Eds.), Siri kepelbagaian biologi hutan. Hutan Simpan Ulu Muda, Kedah: Pengurusan, persekitaran fizikal dan biologi (pp. 115-126). Kuala Lumpur, Malaysia: Jabatan Perhutanan Semenanjung Malaysia.
Zaidi, M. I., Azman, S., & Wilfred, G. M. (2006). Butterflies (Lepidoptera: Rhopalocera) of Hutan Simpan Pasir Raja. In M. Azahar, J. Na'aman, S. Mohd Rizal, M. S. Jalil, M. S. Nizam, & A. Latiff (Eds.), Siri kepelbagaian biologi
216
hutan. Gunung Mandi Angin, Terengganu: Pengurusan, oersekitaran fizikal, kepelbagaian biologi dan pelancongan ekologi (pp. 359-370). Kuala Lumpur, Malaysia: Jabatan Perhutanan Semenanjung Malaysia.
Zaidi, M. I., Idris, A. B., & Hadzureen, H. (1998). Butterfly fauna of the Kuala Lompat sector of Krau Wildlife Reserve, Pahang: An additional survey and assessment. Serangga, 3, 211-226.
Zaidi, M. I., & Soh, P. H. (2000). Butterflies (Lepidoptera: Rhopalocera) from Pinang Islands, Terengganu, Malaysia. Serangga, 5, 221-225.
Zografou, K., Kati, V., Grill, A., Wilson, R. J., Tzirkalli, E., Pamperis, L. N., & Halley, J. M. (2014). Signals of climate change in butterfly communities in a Meditterranean protected area. PLOS ONE, 9, 1, e87245.
Zonneveld, C., Longcore, T., & Mulder, C. (2003). Optimal schemes to detect the presence of insect species. Conservation Biology, 17, 476-487.
University of Malaya
217
LIST OF PUBLICATIONS AND PAPERS PRESENTED
LIST OF PUBLICATIONS
1. Phon, C.-K., Kirton, L.G., & Norma-Rahid, Y. (2017). Monitoring butterflies using counts of puddling males: A case study of the Rajah Brooke's Birdwing (Trogonoptera brookiana albescens). PLOS ONE, 12, 12, e0189450.
2. Phon, C.-K., Kirton, L. G., Joanne Tan, P. C., & Norma-Rashid, Y. (2015, January). Bringing back birdwings: A colourful success story at the Forest Research Institute Malaysia. Malaysian Naturalist, 68-2, 72–75.
LIST OF PAPERS PRESENTED
3. Phon, C.-K., Kirton, L. G., & Norma-Rashid, Y. (2018, Jan). Methods of monitoring populations of the Rajah Brooke’s Birdwing, Trogonoptera brookiana (Lepidoptera, Papilionidae). Oral paper presented at the 3rd International Forum on Sustainable Future in Asia and 3rd NIES International Forum, Seri Pacific Hotel, Kuala Lumpur, Malaysia.
4. Phon, C.-K., Kirton, L. G., & Norma-Rashid, Y. (2016, Nov). Enhancement of the birdwing populations in FRIM by germinating host plants and recording the life history of the birdwings found on the host plants. Oral paper presented at the Project Evaluation Meeting (PEM), FRIM, Malaysia.
5. Phon, C.-K., Kirton, L. G., & Norma-Rashid, Y. (2015, Nov). The potential of enhancing birdwing populations and their visibility as ecotourism icons by planting indigenous host and nectar plants. Oral paper presented at the Seminar Tapak Warisan Negeri Selangor Siri 1: FRIM road to WHS-UNESCO: FRIM campus research 1926–2015, FRIM, Malaysia.
6. Phon, C.-K., Kirton, L. G., & Norma-Rashid, Y. (2015, Sep). Nectaring behaviour of the Rajah Brooke’s Birdwing, Trogonoptera brookiana albescens (Papilionidae) in Ulu Geroh village. Poster presented at the International Science and Nature Congress incorporating of the Conference on Forestry and Forest Product Research, Putra World Trade Centre, Kuala Lumpur, Malaysia.
7. Phon, C.-K., Kirton, L. G., & Norma-Rashid, Y. (2014, Dec). A monitoring technique for puddling butterflies. Oral paper presented at The Expert Workshop to Develop Guidelines for Standardised Global Butterfly Monitoring, German Centre Universityfor Integrative Biodiversity Research, Leipzig,of Germany.Malaya
8. Phon, C.-K., Kirton, L.G., & Norma-Rashid, Y. (2014, Feb). The distribution, behaviour and ecology of the Rajah Brooke’s Birdwing (Trogonoptera brookiana) in Peninsular Malaysia. Oral paper presented at the Candidature Defense Seminar, Institute of Biological Sciences, University of Malaya, Kuala Lumpur, Malaysia.
9. Phon, C.-K., Kirton, L. G., & Norma-Rashid, Y. (2014, Jan). Nectaring behaviour of the Rajah Brooke’s Birdwing, Trogonoptera brookiana albescens (Papilionidae) in Ulu Geroh village. Oral paper presented at the 18th Biological Sciences Graduate
218
Congress, “Nurturing Talents, Stimulating Research”, University of Malaya, Kuala Lumpur, Malaysia. (In collaboration with University of Malaya, National University of Singapore and Chulalongkorn University).
10. Phon, C.-K., Kirton, L. G., & Norma-Rashid, Y. (2013, Nov). The potential of enhancing birdwing populations and their visibility as ecotourism icons by planting indigenous host and nectar plants. Oral paper presented at the Conference on Forestry and Forest Products Research 2013, “Forestry R, D & C: Meeting National and Global Needs”, Sunway Putra Hotel, Kuala Lumpur, Malaysia.
11. Phon, C.-K., Kirton, L. G., & Norma-Rashid, Y. (2012, Dec). Monitoring and behavioural rhythms of the Rajah Brooke’s Birdwing, Trogonoptera brookiana albescens (Papilionidae). Oral paper presented at the Seminar on Zoological and Ecological Research in Progress 2012 (ZERP 2012), Institute of Biological Sciences, Faculty of Science, University of Malaya, Kuala Lumpur, Malaysia.
University of Malaya
219