Lim Hong Chiun

GENETIC DIVERSITY AND PHYLOGEOGRAPHY OF THE FRESHWATER HALFBEAK, GENUS Hemirhamphodon (TELEOSTEI: HEMIRAMPHIDAE: ZENARCHOPTERINEA) IN SUNDALAND RIVER BASINS

LIM HONG CHIUN

UNIVERSITI SAINS MALAYSIA 2017 GENETIC DIVERSITY AND PHYLOGEOGRAPHY OF THE FRESHWATER HALFBEAK, GENUS Hemirhamphodon (TELEOSTEI: HEMIRAMPHIDAE: ZENARCHOPTERINEA) IN SUNDALAND RIVER BASINS

LIM HONG CHIUN

Thesis submitted in fulfillment of the requirements for the degree of Doctor of Philosophy

August 2017 ACKNOWLEDGEMENT

My most sincere gratitude and appreciation to my great supervisor Prof. Siti

Azizah Mohd Nor who was always behind me in giving endless support, guidance, patience, encouragement throughout my research and believing in me for the ‘flexi working time’. Without her big helping hand, I think I would not get this far. Thanks again to Mummy Sazzy. Besides, I would like to say thank you to Prof. Geoffrey

Chambers and Dr. Eleanor for sharing their knowledge and expertise during their visit in USM. My appreciation also goes to Dr. Mark de Bruyn for providing very helpful ideal, information and technique to make this research better.

Special thanks to Pisceslim (Lim Teow Yeong) for unconditional assistance in field works and the knowledge shared with me. Not forgetting also thanks to Mr.

Michael Lo and the late Mr. Heng Wei Ann for great assistance during field work in

Sarawak, Prof. Muchlisin from Syiah Kuala Universiti, Aceh and Prof. Chaidir from

Riau University in spending their precious time and helping hand during field work in

Sumatra.

My gratitude also goes to members and friends of Lab 308, Kak Adel, Jamsari,

Danial, Adib, Lia, Nurul, Wani, Fong, Nazia, Min Pau as well as the new lab members and everyone that might have cross my path. Thank you for helping me in lab work, accompanying me for lunches, trips (field and holiday) and sometimes need to bear with me (nagging).

My profound gratitude to my parents who is loving and supporting me all the time in my life. To my dear beloved wife, ‘Missi’, a simple word of ‘Thank you’ is not enough for your patience in waiting me to complete this ‘slow moving’ PhD journey, your sacrifice in building our lovely little family and your greatest love. All of these will be in my heart for ever and I love you.

Last but not least, my appreciations to Ministry of Higher Education for financial support through MyBrian15 Scholarship and Universiti Sains Malaysia for funding my research under iReC Project.

iii

TABLE OF CONTENTS

ACKNOWLEDGEMENT ii

TABLE OF CONTENTS iv

LIST OF TABLES ix

LIST OF FIGURES xiv

LIST OF PLATES xxiv

LIST OF ABBREVIATIONS xxv

ABSTRAK xxvi

ABTRACT xxviii

CHAPTER ONE: GENERAL INTRODUCTION

1.1 Introduction 1

1.2 Problem statement 6

1.3 Objectives 8

CHAPTER TWO: LITERATURE REVIEW

2.1 Sundaland 10

2.2 Palaeogeography of Southeast Asia focusing on Sundaland 12

2.3 The cyclical glaciations and the impact on SE Asian biota 18

2.4 Biodiversity and biogeography of Southeast Asia (Sundaland) 21

2.5 Phylogeography 24

2.6 The genus Hemirhamphodon 27

2.7 Molecular markers in phylogenetics and phylogeographic studies 32

2.8 Mitochondrial DNA markers- DNA Barcoding and cytochrome b 33

2.9 Nuclear DNA markers 38

2.10 Biodiversity and conservation in Southeast Asia focusing on Sundaland 40

CHAPTER THREE: SPECIES DIVERSITY ASSESSMENT OF THE GENUS Hemirhamphodon THROUGH DNA BARCODING

3.1 Introduction 43

3.1.1 Current status/ problems 45

3.2 Objectives 46

3.3 Materials and methods 47

3.3.1 Sample collection, preservation and DNA extraction 47

3.3.2 Gene amplification and sequencing 52

3.3.3 Data Analysis 52

3.4 Results 56

3.4.1 Genetic diversity, intraspecific and interspecific divergences 56

3.4.2 Gene tree and OTU counts based on DNA Barcoding 58

3.4.3 Genetic divergence based on the newly assigned groups 64

3.5 Discussion 68

3.5.1 High levels of intraspecific divergences 68

3.5.2 Influence of Paleo-drainage systems 69

3.5.3 Evidence of cryptic species 70

3.5.4 Conservation and management 73

3.6 Conclusions 74

CHAPTER FOUR: POPULATION STUDY AND PHYLOGEOGRAPHY OF Hemirhamphodon pogonognathus IN SUNDALAND RIVER BASINS

4.1 Introduction 75

4.2 Objectives 81

4.3 Materials and methods 82

4.3.1 Sample collection, preservation and DNA extraction 82

4.3.2 Gene amplification and sequencing 82

4.3.3 Data sorting and haplotype generation 85

4.3.3(a) Mitochondrial DNA analysis 85

4.3.3(b) Nuclear DNA analysis 85

4.3.3(c) Genetic diversity 86

4.3.3(d) Gene tree construction 87

4.3.3(e) Haplotype network 88

4.3.3(f) Population structure 88

4.3.3(g) Historical demographic analysis 90

4.4 Results 93

4.4.1 Mitochondrial DNA Analysis 93

4.4.1(a) Nucleotide composition and genetic diversity 93

4.4.1(b) Haplotype distribution 95

4.4.1(c) Phylogeography and evolutionary relationships 102

among haplotypes

4.4.1(d) Population genetic divergence and population 107

structure

4.4.1(e) Population history and demographic changes 112

4.4.2 Nuclear DNA (SCNP: Hp5 and Hp54) analysis 118

4.4.2(a) Nucleotide composition and genetic diversity 118

4.4.2(b) Haplotype distribution 121

4.4.2(c) Phylogeography and evolutionary relationships 132

among haplotypes

4.4.2(d) Population structure 141

4.4.2(e) Population history and demographic changes 150

4.5 Discussion 159

4.5.1 Genetic Diversity and haplotype distribution 159

4.5.2 Population structure and phylogeography 164

4.5.3 Historical demography 167

4.5.4 Evidence for Cryptic Species 169

4.5.5 Conservation 171

4.6 Conclusion 173

CHAPTER FIVE: GENETIC DIVERSITY AND PHYLOGEOGRAPHY OF H. byssus AND H. kuekenthali IN SARAWAK RIVER BASINS (NORTHWESTBORNEO)

5.1 Introduction 174

5.2 Objectives 182

5.3 Materials and methods 183

5.3.1 Sample collection, preservation and DNA extraction 183

5.3.2 Gene amplification and sequencing 184

5.3.3 Data analyses 185

5.3.3(a) Data sorting and haplotype generation 185

5.3.3(b) Genetic diversity 186

5.3.3(c) Gene tree and haplotype network construction 186

vii

5.3.3(d) Population structure 187

5.3.3(e) Historical demographic analysis 188

5.4 Results 190

5.4.1 Nucleotide composition and genetic diversity 190

5.4.2 Haplotype distribution 196

5.4.3 Phylogeography and evolutionary relationships among haplotypes 204

5.4.4 Population structure 216

5.4.5 Population Genetic divergence 229

5.4.6 Population history and demographic changes 236

5.5 Discussion 249

5.5.1 Genetic Diversity 249

5.5.2 Population structure and phylogeography 250

5.5.3 Historical demography 256

5.5.4 Taxonomic implications 257

5.5.5 Conservation 260

5.6 Conclusion 260

CHAPTER SIX: GENERAL DISCUSSION AND CONCLUSIONS 262

REFERENCES 267

APPENDICES

LIST OF PUBLICATIONS /CONFERENCE PRESENTATIONS

viii

LIST OF TABLES

Page

Table 3.1 Species name, locations, sample size (n) code, regional 48 locations within Peninsular Malaysia, Sarawak (Borneo), and Sumatra; and the newly assigned groups of Hemirhamphodon species (based on the constructed Neighbour-Joining COI gene tree).

Table 3.2 A total of 112 COI gene sequences of Hemirhamphondon 53 species obtained from GenBank and the sampling locations.

Table 3.3 Sample size (n), mean values, ranges of genetic 57 divergences based on K2P across taxonomic levels from 311 sequences of the genus Hemirhamphod on.

Table 3.4 Pairwise comparisons of the COI gene based on K2P 57 distance among six presumed (morphologically identified) Hemirhamphodon species.

Table 3.5 Pairwise comparison of the COI gene based on K2P 65 distance among newly assigned Hemirhamphodon groupings.

Table 4.1 Sampling locations, sample size (n), code, the hypothetical 84 Paleo-drainages in Sundaland where the sampling locations are situated, and regional locations within Peninsular Malaysia and Sumatra. Ma = Malacca, Si = Siam, nS = North Sunda

Table 4.2 Sample location in abbreviation, no. of sequences (n), no. 94 of haplotype (hp), no. of polymorphic sites (#V), nucleotide diversity (π), haplotype diversity (h) and expected heterozygosity per site based on number of segregating sites (ϴs) of cyt b for H. pogonognathus populations.

Table 4.3 The 39 haplotypes with 209 variable sites within a 883 bp 96 segment of cyt b for H. pogonognathus populations.

Table 4.4 Haplotype distribution of cyt b across 25 H. pogonognathus 100 populations in Sundaland.

Table 4.5 Best model across partitions after PartitionFinder analysis 105 for cyt b of H. pogonognathus.

Table 4.6 Pairwise genetic distances with TN93+G model of cyt b 108 between H. pogonognathus populations.

Table 4.7 Pairwise ФST values of cyt b (below diagonal) between H. 110 pogonognathus populations. Geographical distances (above diagonal in italics) between populations based on the shortest path following the Paleo-drainages systems (km).

Table 4.8 AMOVA result for hierarchical genetic subdivision for 111 percentage of variation and fixation indices (ФST, ФSC and ФCT) of cyt b of H. pogonognathus populations. Bold values indicate significant value (p < 0.05).

Table 4.9 Summary of population neutrality tests and demographic 113 analyses based on Tajima’s D, Fu’s Fs, Rasmos-Onsins & Rozas (R2) and Harpending’s raggedness index (Hri) for H. pogonognathus populations based on cyt b gene.

Table 4.10 Sample location in abbreviation, no of sequences (n), no. 119 of haplotype (hp), no. of polymorphic sites (#V), nucleotide diversity (π), haplotype diversity (h) and expected heterozygosity per site based on number of segregating sites (ϴs) of Hp5 and Hp54 for H. pogonognathus populations.

Table 4.11 The 71 haplotypes with 79 variable sites generated from 123 264 nucleotide bases of Hp5 in H. pogonognathus populations.

Table 4.12 Haplotype distribution of Hp5 across 25 H. pogonognathus 125 populations.

Table 4.13 The 48 haplotypes with 41 variable sites generated from 128 264 nucleotide bases of Hp54 for H.pogonognathus.

Table 4.14 Haplotype distribution of Hp54 in 25 H. pogonognathus 130 populations.

Table 4.15 Pairwise genetic distances with K2P+G model of Hp5 142 between H. pogonognathus populations.

Table 4.16 Pairwise genetic distances with TN92+G distribution 143 model of Hp54 between H. pogonognathus populations.

Table 4.17 Pairwise ФST values of Hp5 (below diagonal) between H. 145 pogonognathus populations. Bold values indicate non significant ФST values after Bonferroni corrections (p > 0.05). Geographical distances (above diagonal in italics) between populations of the shortest path follow the Paleo- drainages (km).

Table 4.18 Pairwise ФST values of Hp54 (below diagonal) between H. 146 pogonognathus populations. Bold values indicate non significant ФST values after Bonferroni corrections (p > 0.05). Geographical distances (above diagonal in italics) between populations of the shortest path follow the Paleo- drainages systems (km).

Table 4.19 AMOVA results for hierarchical genetic subdivision for 148 percentage of variation and fixation indices (ФST, ФSC and ФCT) of Hp5 and Hp54 of H. pogonognathus populations. Bold values indicate significant value (p < 0.05).

Table 4.20 Summary of population neutrality tests and demographic 151 analyses. Tajima’s D, Fu’s Fs, Rasmos-Onsins & Rozas (R2) and Harpending’s raggedness index (Hri) of Hp5 and HP54 of H. pogonognathus populations.

Table 5.1----Sampling locations, sample size (n) and code of H. 184 byssus and H. kuekenthali. The central Sarawak region for H. byssus starts from Sri Aman until south of Rajang river while for H. kuekenthali starts form north of Rajang river until Bintulu.

Table 5.2 Sampling location in abbreviation, no. of sequences (n), 192 no. of haplotype (hp), no. of polymorphic sites (#V), nucleotide diversity (π), haplotype diversity (h) and expected heterozygosity per site based on number of segregating sites (ϴs).

Table 5.3 The generated haplotypes of three DNA markers of H. 197 byssus. (a) Nine haplotypes with 91 variable sites of 883

nucleotide bases of cyt b. (b) Seven haplotypes with 14 variable sites of 261 nucleotide bases of Hp5 and (c) Fourteen haplotypes with 17 variable sites of 243 nucleotide bases of Hp54.

Table 5.4 Haplotype distribution of cyt b, Hp5 and Hp54 across five 198 H. byssus populations.

Table 5.5 The generated haplotypes of three DNA markers of H. 200 kuekenthali. (a) Twenty-three haplotypes with 166 variable sites of 883 nucleotide bases of cyt b. (b) Thirteen haplotypes with 64 variable sites of 261 nucleotide bases of Hp5 and (c) Fifteen haplotypes with 16 variable sites of 243 nucleotide bases of Hp54.

Table 5.6 Haplotype distribution of cyt b, Hp5 and Hp54 across 202 eleven H. kuekenthali populations.

Table 5.7 Best model across partitions after PartitionFinder analysis 204 for BI cyt b tree of H. byssus and H.kuekenthali.

Table 5.8 Pairwise ФST values of cyt b, Hp5 and Hp54 (below 217 diagonal) between H. byssus populations. Geographical distance (above diagonal in italics) between populations is based on the shortest path following the hypothetical Paleo-drainage systems (km).

Table 5.9 AMOVA results for hierarchical genetic subdivision for 220 percentage of variation and fixation indices (ФST, ФSC and ФCT) of cyt b, Hp5 and Hp54 of H. byssus populations.

Table 5.10 Pairwise ФST values of cyt b, Hp5 and Hp54 (below 223 diagonal) between H. kuekenthali populations. Geographical distances (above diagonal in italics) between populations based on the shortest path following the hypothetical Paleo-drainage systems (km).

Table 5.11 AMOVA results for hierarchical genetic subdivision for 227 percentage of variation and fixation indices (ФST, ФSC and ФCT) of cyt b, Hp5 and Hp54 of H. kuekenthali populations.

xii

Table 5.12 Pairwise genetic distances with Tamura-Nei+G model of 230 cyt b between H. byssus and H. kuekenthali populations.

Table 5.13 Pairwise genetic distances with Tamura-3-parameter 232 model of Hp5 between H. byssus and H. kuekenthali populations.

Table 5.14 Pairwise genetic distances with Tamura-3-parameter+G 235 model of Hp54 between H. byssus and H. kuekenthali populations.

Table 5.15 Summary of population neutrality tests and demographic 237 analyses based on Tajima’s D, Fu’s Fs, Rasmos-Onsins & Rozas (R2) and Harpending’s raggedness index (Hri) for H. byssus and H. kuekenthali populations.

xiii

LIST OF FIGURES

Page

Figure 1.1 Map of Southeast Asia indicating contemporary 1 landmasses, straits, seas, arcs, and faunal lines (Modified from de Bruyn, 2005).

Figure 1.2 The four biogeographic sub-regions (bioregions or 3 hotspots) in SE Asia. (Adapted from Woodruff, 2010).

Figure 2.1 Shading in dark grey is present day landscape. The 11 Sundaland landmass during the middle Pleistocene period when sea level was 120m below the present level is shaded in light grey. (Adapted and modified from Anderson & Collette, 1991).

Figure 2.2 Distribution of continental blocks, fragments and 12 terranes, and principal sutures of SE Asia. Colour blocks explain the formation of SE Asia (Sunda shelf) through amalgamation of continental blocks/terranes at different periods. LT = Lincang Terrane; CT = Chanthaburi Terrane. (Metcalfe, 2011a).

Figure 2.3 The geological timescale. Ages are in million years. 13 (Adapted from Hall, 1998)

Figure 2.4 Paleogeography of the SE Asia region and key habitat 16 availability between 30 and 5Ma. (Adapted from Hall, 2012a)

Figure 2.5 The Paleo-drainage systems in Sundaland during the low 18 sea level period based on Voris, (2000). Ma = Malacca, Si = Siam, nS = North Sunda, eS = East Sunda, Me = Mekong. Map was reprinted/modified with kind permission from H. Voris and Field Museum of Natural History, Chicago.

Figure 2.6 Sea level fluctuation from 3mya with 100ka’s cycle and 19 41ka’s cycle; 500kya and 20kya fluctuation. (Adopted and modified from Chen, 2016)

xiv

Figure 2.7 (a) Hypothetical sampling locations of freshwater taxa in 26 SE Asia based on the Paleo-drainage systems by Voris (2000). (b). A priori hypotheses phylogenetic relationships of the freshwater taxa as a result of the occurrences of connectivity facilitated by paleo- drainages dispersal model, followed by allopatric speciation due to rising sea levels (Adapted from de Bruyn, et al., 2012). Me = Mekong, Si = Siam, Ma = Malacca, nS = North Sunda, eS = East Sunda.

Figure 2.8 Distribution of the genus Hemirhamphodon (Modified 28 from Anderson & Collette, 1991; Tan & Lim, 2013). Shaded areas indicate land area of Sundaland during the middle Pleistocene period (Voris, 2000).

Figure 2.9 The melanophores on dorsal fin in (a) H. kuekenthali are 30 continuous to the but not continuous to the base in (b) H. pogonognathus. (Photo was reprinted/modified with kind permission from Piscesilim, 2012)

Figure 2.10 Structure of a specimen record in Barcode of Life 35 Datasystem (BOLD) (Hubert, et al., 2015).

Figure 2.11 Schematic representation of the Inferred Barcoding Gap. 36 (A) Existence of ‘barcoding gap’ with discrete distributions of intraspecific and interspecific with no overlapping. (B) Overlapping of intraspecific and interspecific distribution, no ‘barcoding’ gap (possibility of cryptic species) (Meyer & Paulay, 2005).

Figure 2.12 An example of the maximum intraspecific divergence 37 plotted against the nearest-neighbour distance of presumed monophyletic species. Plots that fall above the 1:1 line indicate the presence of a “barcode gap” (more than one species exist in the presumed monophyletic species). (Adapted and modified from (Robinson, et al., 2009)

Figure 3.1 Sampling locations of H. pogonognathus, H. byssus and 47 H. kuekenthali (Modified from Anderson & Collette, 1991), for location abbreviation see Table 3.1.

Figure 3.2 Maximum intraspecific divergence compared with 58 nearest-neighbour distance for four initial presumed morphological Hemirhamphodon species excluding two singleton species. Diagonal line represents 1:1 line to separate “barcode gap” presence and absence area.

Figure 3.3 Neighbor-Joining COI gene tree among 60 Hemirhamphodon species generated through K2P. Values at nodes represent bootstrap confidence levels (10000 replicates). A Dermogenys species was employed as an outgroup. The scale bar refers to genetic distance.

Figure 3.4 Neighbor-Joining COI gene tree with representative male 61 specimen live photo from selected locations. The scale bar refers to genetic distance and values on branches are bootstrap values.

Figure 3.5 Bayesian Inference COI gene tree generated through 62 HKY+G+I. Value at nodes represents the Bayesian posterior probability. A Dermogenys species was employed as an outgroup. The scale bar refers to genetic distance.

Figure 3.6 The number of genetically distinct OTUs according to 63 the prior intraspecific divergence value generated by ABGD based on K2P.

Figure 3.7 Maximum intraspecific divergence compared with 66 nearest-neighbour distance for newly assigned Hemirhamphodon species grouping. Diagonal line represents 1:1 line to separate “barcode gap” presence and absence area.

Figure 3.8 Sampling location according to Paleo-drainage systems 67 mapped with NJ COI gene tree. Ma = Malacca, Si = Siam, nS = North Sunda, eS = East Sunda, Me = Mekong. H. pogonognathus (shaded circle), H. byssus (shaded triangle) and H. kuekenthali (shaded four-pointed star). Colours of shapes and bars are identical in representing new assigned grouping. The embedded map was reprinted/modified from Voris, 2000 under a CC BY license, with permission from Field Museum of Natural History, Chicago, original copyright 2000.

xvi

Figure 4.1 Sampling locations of H. pogonognathus in contemporary 83 geography landmass (Modified from Anderson & Collette, 1991), for location abbreviation see Table 4.1.

Figure 4.2 Map showing the hypothetical Paleo-drainages existed 90 in Sundaland during the low sea level around 20,000 BP and the sampling sites of H. pogonognathus. Number at the sampling sites is following the location sequence from top to bottom in Table 4.1. (Modified from Irwanto, 2015).

Figure 4.3 Nucleotide substitution saturation analysis for cyt b of H. 95 pogonognathus. Transitions (s) and transversions (v) plotted against p distance.

Figure 4.4 Neighbor-Joining cyt b tree among 39 H. pogonognathus 103 haplotypes generated through TN93+G model. Values at nodes represent bootstrap confidence level (1000 replicates). A single H. kuekenthali (KJI OG) sequence was used as outgroup. Three major clades indicated by black vertical bars: Main, southern Sumatra and Kelantan. The scale bar refers to genetic distance.

Figure 4.5 Bayesian Inference cyt b tree among 39 H. pogonognathus 104 haplotypes. Values at nodes represent the Bayesian posterior probabilities. A single H. kuekenthali (KJI OG) sequence were used as outgroup. Three major clades indicated by black vertical bars: Main, southern Sumatra

and Kelantan. The scale bar refers to genetic distance.

Figure 4.6 Minimum Spanning Network of 39 cyt b haplotypes of 106 H. pogonognathus. Crossbars on connecting line indicate the number of substitutions separating haplotypes and the number indicate additional mutation steps. The size of the circles is proportional to haplotype frequency. Small black dots are missing haplotypes linking the clades. (b) Map of Sundaland with Paleo-drainages systems indicated by colours.

Figure 4.7 Pairwise ФST comparisons of cyt b (below diagonal) 109 between H. pogonognathus populations.

Figure 4.8 The ФST plotted against the shortest path following the 112 hypothetical Paleo-drainages systems (km) of H.

xvii

pogonognathus cyt b data. Trendline (black dashed line) shows the general pattern of only slight increase in genetic distance with a higher leap in geographical distance (IBD).

Figure 4.9 Mismatch distribution of cyt b of twelve H. 115 pogonognathus populations showing the expected and observed pairwise differences between the sequences with the respective frequencies under constant population size. The solid lines represent the expected distribution and the dotted lines represent the observed distribution. The dotted line shows that the left edge of distribution converges rapidly toward the new equilibrium. Multimodal was detected for three populations (JMP, KJ and ST) and total samples dataset.

Figure 4.10 Extended Bayesian Skyline Plots (EBSPs) showing the 117 demographic history of H. pogonognathus based on cyt b. Solid blue line is the median effective population size, the dashed lines are the upper and lower 95% HPD for those estimates.

Figure 4.11 Nucleotide substitution saturation analysis for Hp5. 12 0 Transitions (s) and transversions (v) plotted against p distance.

Figure 4.12 Nucleotide substitution saturation analysis for Hp54. 120 Transitions (s) and transversions (v) plotted against p distance.

Figure 4.13 Neighbour-Joining HP5 tree among 71 haplotypes of H. 133 pogonognathus generated through K2P+G model. Values at nodes represent bootstrap confidence level (1000 replicates). A H. kuekenthali (KJI501 OG) sequence was used as outgroup. Two major clades are indicated by black vertical bars: Main and Kelantan. The scale bar refers to genetic distance.

Figure 4.14 Bayesian Inference Hp5 tree generated through K2P+G 134 model. Value at nodes represents the Bayesian posterior probabilities. A H. kuekenthali (KJI501 OG) sequence was used as outgroup. Two major clades are indicated by black vertical bars: Main and Kelantan. The scale bar refers to genetic distance.

xviii

Figure 4.15 Neighbour-Joining Hp54 tree among 48 haplotypes of H. 135 pogonognathus generated through TN92+G distribution model. Values at nodes represent bootstrap confidence level (1000 replicates). A H. kuekenthali (KJI OG) sequence was used as an outgroup. The scale bar refers to genetic distance.

Figure 4.16 Bayesian Inference Hp54 tree generated through 136 TN92+G model. Values at nodes represents the Bayesian posterior probabilities. A H. kuekenthali (KJI OG) sequence was used as outgroup. The scale bar refers to genetic distance.

Figure 4.17 Minimum Spanning Network of 71 Hp5 haplotypes 138 obtained from 458 individuals. Crossbars on connecting line indicate the number of substitutions separating haplotypes and the numbers indicate additional mutation steps. The size of the circles is proportional to haplotype

frequency. Small black dots are missing haplotypes linking the clades. (b) Map of Sundaland with Paleo- drainages systems indicated by colours.

Figure 4.18 Minimum Spanning Network of 48 Hp54 haplotypes 140 obtained from 375 individuals. Crossbars on connecting line indicate the number of substitutions separating haplotypes and the numbers indicate additional mutation steps. The size of the circles is proportional to haplotype frequency. Small black dots are missing haplotypes linking the clades. (b) Map of Sundaland with Paleo- drainages systems indicated by colours.

Figure 4.19 Pairwise ФST comparisons of Hp5 (below diagonal) 147 between H. pogonognathus populations.

Figure 4.20 Pairwise ФST comparisons of Hp54 (below diagonal) 147 between H. pogonognathus populations.

Figure 4.21 The ФST plotted against the shortest path following the 149 hypothetical Paleo-drainages systems (km) of H. pogonognathus Hp5 data. Trendline (Black dashed line) shows the general pattern of little increasing genetic distance with greater geographical distance (IBD).

xix

Figure 4.22 The ФST plotted the shortest path following the 149 hypothetical Paleo-drainages systems (km) of H. pogonognathus Hp54 data. Trendline (black dashed line) shows the general pattern of little increasing genetic distance with greater geographical distance (IBD).

Figure 4.23 Mismatch distribution of Hp5 of 12 H. pogonognathus 152 populations showing the expected and observed pairwise differences between the sequences with the respective frequencies under constant population size. The solid lines represent the expected distribution and the dotted lines represent the observed distribution. The dotted line shows that the left edge of distribution converges rapidly toward the new equilibrium.

Figure 4.24 Mismatch distribution of Hp54 of 12 H. pogonognathus 154 populations showing the expected and observed pairwise differences between the sequences with the respective frequencies under constant population size. The solid lines represent the expected distribution and the dotted lines represent the observed distribution.

Figure 4.25 Extended Bayesian Skyline Plots (EBSPs) showing the 158 demographic history of H. pogonognathus of (a) Hp5 and (b) Hp54. Solid blue line is the median effective population size, the dashed lines are the upper and lower 95% HPD for those estimates.

Figure 4.26 Divergence time of H. pogonognathus populations in 162 Sundaland river basins based on ultrametric Bayesian mitochondrial DNA trees of COI and control region variation. Values at nodes are median ages (in millions of years; bars=95% highest posterior densities). (Adopted and modified from de Bruyn, et al., 2013).

Figure 5.1 Morphology of dorsal fin. (A) H. byssus male, anterior 176 half of dorsal fin with black streaks on the inter-radial membrane. (B) H. kuekenthali male, black streaks in the middle of dorsal fin. (Adapted from Tan & Lim, 2013)

Figure 5.2 Map showing the main rivers and boundaries within 179 Borneo Island. The embedded map illustrates the five biogeographical sub-regions of Borneo. (Modified from Yap, 2002 and Tan, 2006).

Figure 5.3 Sampling locations of H. byssus and H. kuekenthali in 183 contemporary geographical landmass, Sarawak (northwest Borneo). (Modified from Nguyen, et al., 2006b). For location abbreviation see Table 5.1.

Figure 5.4 Map showing the hypothetical Paleo-drainages existed 188 in Borneo during the low sea level around 20,000 BP and the sampling sites of H. byssus and H. kuekenthali. Number at the sampling sites follows the location sequence from top to bottom in Table 5.1. (Modified from Irwanto, 2015).

Figure 5.5 Nucleotide substitution saturation analysis for (a) cyt b, 193 (b) Hp5 and (c) Hp54 of H. byssus. Transitions (s) and transversions (v) plotted against p distance.

Figure 5.6 Nucleotide substitution saturation analysis for (a) cyt b, 195 (b) Hp5 and (c) Hp54 of H. kuekenthali. Transitions (s) and transversions (v) plotted against p distance.

Figure 5.7 Neighbor-Joining and Bayesian Inference trees among 207 cyt b haplotypes of H. byssus and H. kuekenthali. Values at nodes represent bootstrap confidence level and Bayesian posterior probabilities. A single H. pogonognathus (BP01 OG) sequence was used as outgroup. The major clades were indicated by coloured vertical bars. The scale bar refers to genetic distance.

Figure 5.8 Neighbor-Joining and Bayesian Inference trees among 209 Hp5 haplotypes of H. byssus and H. kuekenthali. Values at nodes represent bootstrap confidence level and Bayesian posterior probabilities. . Clade (A) that consists of admixed haplotypes from both species was indicated by black vertical bar. A single H. pogonognathus (BP01 OG) sequence was used as outgroup. The scale bar refers to genetic distance.

Figure 5.9 Neighbor-Joining and Bayesian Inference trees among 210 Hp54 haplotypes of H. byssus and H. kuekenthali. Values at nodes represent bootstrap confidence level and Bayesian posterior probabilities. A single H. pogonognathus (BP01 OG) sequence was used as outgroup. The scale bar refers to genetic distance.

xxi

Figure 5.10 Haplotype Minimum Spanning Network of (a) cyt b, (b) 213 Hp5 and (c) Hp54 of H. byssus. Number in red on the connecting line indicated the number of substitutions separating haplotypes. The size of the circles is proportional to haplotype frequency. Small black dots are missing haplotypes linking the clades.

Figure 5.11 Minimum Spanning Network of (a) cyt b, (b) Hp5 and 215 (c) Hp54 haplotypes of H. kuekenthali. Number in red on the connecting line indicated the number of substitutions separating haplotypes. The size of the circles is proportional to haplotype frequency. Small black dots are missing haplotypes linking the clades.

Figure 5.12 Pairwise ФST comparisons of (a) cyt b, (b) Hp5 and (c) 218 Hp54 (below diagonal) between H. byssus populations.

Figure 5.13 The ФST plotted against the shortest path following the 221 hypothetical Paleo-drainage systems (km) of H. byssus. (a) cyt b, (b) Hp5 and (c) Hp54 data.

Figure 5.14 Pairwise ФST comparisons of (a) cyt b, (b) Hp5 and (c) 225 Hp54 (below diagonal) between H. kuekenthali populations.

Figure 5.15 The ФST plotted against the shortest path following the 228 hypothetical Paleo-drainage systems (km) of H. kuekenthali. (a) cyt b, (b) Hp5 and (c) Hp54 data.

Figure 5.16 Mismatch distribution of H. byssus populations based on 239 three markers (cyt b, Hp5 and Hp54) showing the expected and observed pairwise differences between the sequences with the respective frequency under constant population size. The solid lines represent the expected distribution and the dotted lines represent the observed distribution. The dotted line shows that the left edge of distribution converges rapidly toward the new equilibrium in most populations.

Figure 5.17 Extended Bayesian Skyline Plots (EBSPs) showing the 242 demographic history of H. byssus based on cyt b, Hp5 and Hp54. Solid blue line is the median effective population size, the dashed lines are the upper and lower 95% HPD for those estimates.

xxii

Figure 5.18 Mismatch distribution of H. kuekenthali populations 245 based on three markers (cyt b, Hp5 and Hp54) showing the expected and observed pairwise differences between the sequences with the respective frequency under constant population size. The solid lines represent the expected distribution and the dotted lines represent the observed distribution. The dotted line shows that the left edge of distribution converges rapidly toward the new equilibrium in most populations.

Figure 5.19 Extended Bayesian Skyline Plots (EBSPs) showing the 248 demographic history of H. kuekenthali based on cyt b, Hp5 and Hp54. Solid blue line is the median effective population size, the dashed lines are the upper and lower 95% HPD for those estimates.

Figure 5.20 Divergence time of H. byssus and H. kuekenthali 253 populations based on ultrametric Bayesian mitochondrial DNA trees of COI and control region variation. Values at nodes are median ages (in millions of years; bars=95% highest posterior densities). (Adopted and modified from de Bruyn, et al., 2013).

xxiii

LIST OF PLATES

Page

Plate 2.1 Photo of male Hemirhamphodon spp., (a) H. phaiosoma; 29 (b) H. pogonognathus; (c) H. kuekenthali; (d) H. chrysopunctatus; (e) H. kapuasensis; (f) H. tengah; (g) H. byssus (live); (h) H. sesamun and (i) H. kecil. All specimens presented are live colouration except H. kapuasensis, H. sesamun and H. kecil.

Plate 3.1 Live (above) and preserved (below) colouration of 51 male specimen: (a) H. pogonognathus from Peninsular Malaysia, (b) H. pogonognathus from southern Sumatra (c) H. byssus and (d) H. kuekenthali.

Plate 3.2 Different colouration among localities of male 72 H. pogonognathus.

Plate 4.1 Live (above) and preserved (below) colouration of male H. 76 pogonognathus specimen.

Plate 5.1 Live (above) and preserved (below) colouration of male 175 H. byssus specimen.

Plate 5.2 Live (above) and preserved (below) colouration of male H. 175 kuekenthali specimen.

xxiv

LIST OF ABBREVIATIONS

Cyt b cytochrome b

SCNP Single Copy Nucleotide Polymorphism

COI cytochrome c oxidase subunit I mtDNA Mitochondrial DNA nDNA nuclear DNA

NJ Neighbour-Joining

BI Bayesian Inference

AMOVA Analysis of Molecular Variance

MSN Minimum Spanning Network

EBSPs Extended Bayesian Skyline Plots

xxv

KEPELBAGAIAN GENETIK DAN FILOGEOGRAFI IKAN JULUNG-

JULUNG GENUS Hemirhamphodon (TELEOSTEI: HEMIRAMPHIDAE:

ZENARCHOPTERINEA) DI LEMBANGAN SUNGAI PELANTAR SUNDA

ABSTRAK

Genus ikan air tawar Hemirhamphodon atau biasa dikenali sebagai ikan julung-julung terdiri daripada sembilan spesies morfologi. Taburan yang luas dan ciri endemiknya di dalam Pelantar Sunda menjadikan ia sesuai untuk kajian filogenetik, populasi genetik, filogeografi dan juga mempunyai potensi penemuan spesies baru.

“Kod bar DNA” (DNA barcoding) daripada gen sitokrom oksidase subunit I (COI) telah digunakan untuk menjelaskan sistematik genus ikan air tawar julung-julung,

Hemirhamphodon dengan melibatkan 201 individu daripada 46 lokasi di Semenanjung

Malaysia, barat laut Borneo (Sarawak) dan Sumatra. Perbandingan jarak genetik intra- spesies mempamerkan julat yang tinggi dari 0.0% hingga ke 14.8% mencadangkan potensi kewujudan spesies novel atau kriptik. Keputusan ini turut disokong oleh analisis jurang kod bar (Barcode Gap Analysis), ABGD (Automatic Barcode Gap

Discovery) dan pepohon filogenetik gen COI. Penilaian di tahap populasi dan filogeografi bagi spesies H. pogonognathus, H. byssus dan H. kuekenthali juga telah dijalankan dengan gabungan kedua-dua penanda mtDNA (cyt b) dan nDNA (penanda

Single Copy Nucleotide Polymorphism (SCNP), HP5 dan Hp54) untuk menilai perbezaan genetik, penstrukturan populasi, sejarah demografi dan juga sebagai pendekatan untuk penemuan kepelbagaian diversiti yang tersembunyi. Dua puluh lima, lima dan sebelas populasi, masing-masing dianalisis untuk penilaian H. pogonognathus, H. byssus dan H. kuekenthali. Keputusan telah mendedahkan struktur populasi yang tinggi untuk ketiga-tiga spesies yang menunjukkan bahawa kebanyakan

xxvi populasi (jika bukan semua) telah menyumbang kepada kolam gen secara keseluruhan.

Di samping itu, penemuan menekankan kehadiran kepelbagaian spesies tersembunyi atau spesies kriptik sejajar dengan hipotesis awal dengan jangkaan penemuan spesies baharu. Tambahan pula, analisa sejarah demografi mendedahkan bahawa kebanyakan populasi daripada ketiga-tiga spesies Hemirhamphodon mengalami pengurangan saiz populasi yang pesat. Oleh itu, setiap satu daripada populasi Hemirhamphodon harus dianggap sebagai unit pengurusan yang berasingan dalam konteks pemuliharaan.

Projek ini juga mendedahkan bahawa sistem Saliran-Paleo di Pelantar Sunda seolah- olah mempunyai pengaruh yang terhad dalam memacu kepelbagaian spesies

Hemirhamphodon. Bagaimanapun, sejarah geologi seperti kitaran glasier semasa zaman Pleistocene ditambah pula dengan penyusunan semula Saliran-Paleo telah banyak mempengaruhi kepelbagaian genetik populasi bagi ketiga-tiga spesies

Hemirhamphodon. Secara keseluruhan, rangka kerja penyelidikan yang lebih bersepadu perlu dijalankan untuk menyelesaikan isu kompleks spesies berserta dengan strategi pemuliharaan dan pelan pengurusan perlu dilaksanakan terhadap setiap satu populasi Hemirhamphodon untuk kemampanan populasi jangka panjang dan pemuliharaan spesies.

xxvii

GENETIC DIVERSITY AND PHYLOGEOGRAPHY OF THE

FRESHWATER HALFBEAK, GENUS Hemirhamphodon (TELEOSTEI:

HEMIRAMPHIDAE: ZENARCHOPTERINEA) IN SUNDALAND RIVER

BASINS

ABTRACT

The freshwater fish genus Hemirhamphodon commonly known as halfbeak consists of a total nine morphological species. Its wide distribution and endemism within Sundaland makes it suitable for phylogenetic, population and phylogeographic studies, and even the potential discovery of new species. DNA barcoding of the cytochrome oxidase subunit I (COI) gene was utilized to elucidate the systematics of the freshwater halfbeak genus Hemirhamphodon which involved 201 individuals from

46 locations in Peninsular Malaysia, northwest Borneo (Sarawak) and Sumatra.

Pairwise within species comparisons exhibited a high range of intraspecific diversity from 0.0% to 14.8%, suggesting potential occurrence(s) of novel or cryptic species.

This finding was further supported by the barcode gap analysis, ABGD (Automatic

Barcode Gap Discovery) and the constructed COI gene tree. Population level assessment and phylogeography of H. pogonognathus, H. byssus and H. kuekenthali were also conducted with a combination of both mtDNA (cyt b region) and nDNA markers (Single Copy Nucleotide Polymorphism (SCNP) markers, Hp5 and Hp54) to assess the genetic variabilities, population structuring, historical demography as well as attempt to discover hidden diversity. Twenty five, five and eleven populations were analysed in the assessments of the H. pogonognathus, H. byssus and H. kuekenthali, respectively. The results revealed high population structure for all three species indicating that most (if not all) populations contribute to the total gene pool. In addition,

xxviii the findings highlighted the presence of hidden diversity or cryptic species which is in agreement with the initial hypothesis of new species discovery. Furthermore, the historical demographic analyses revealed that most of the populations of the three

Hemirhamphodon species had experienced rapid population size reduction. Therefore, each of the Hemirhamphodon populations should be treated as a separate management unit in the context of conservation. This project also revealed that the Paleo-drainage systems of Sundaland seemed to have limited influence in driving Hemirhamphodon species diversity. On the other hand, the geological history such as the cyclical glaciation events during the Pleistocene epoch coupled with the Paleo-drainage rearrangements have greatly influenced the genetic diversity of the three

Hemirhamphodon species populations. Overall, a more integrated investigation framework needs to be carried out to resolve the species complex issue, and conservation strategies and management plans should be implemented on each of the

Hemirhamphodon populations for long term population sustainability and conservation of species.

xxix

CHAPTER 1

GENERAL INTRODUCTION

1.1 Introduction

Biological contrasts between different geographical zones have been recorded since the 19th century viz; between Australian and Asian biotic transition zone

(Wallace, 1860), between east of Palawan (the most westerly of the Philippine islands) and the rest of the Philippine Archipelago (Huxley, 1868) and also between Sulawesi islands with Sundaland (Kottelat, et al., 1993). The Wallace’s observations (Wallace,

1860;1863) had drawn great attention to naturalists and later several ‘lines’ (Huxley’s line, Weber’s line and Lydekker’s line) (Figure 1.1) were proposed to demarcate biogeographical divisions between the Asian and Australian faunal zones (Mayr, 1944;

Simpson, 1977). Since then, much work has been carried out to understand the evolutionary events in Indo-Australia Archipelago (Lohman, et al., 2011; de Bruyn, et al., 2012).

Figure 1.1: Map of Southeast Asia indicating contemporary landmasses, straits, seas, arcs, and faunal lines (Modified from de Bruyn, 2005).

It is believed that numerous ancient earth geological events and climatic fluctuations have had great influence on shaping these major contemporary zoogeographic boundaries. How species respond to recent climatic fluctuations and associated eustatic changes during these historical phases could be revealed through genealogical studies of the spatial and temporal dynamics of populations (Woodruff,

2010). It can also be used to explain how biogeographic barriers play a major role in shaping species distributions.

During the last glaciation at Pleistocene, the Sundaland experienced radical shifts in sea-land distribution or landscape due to the dramatic fluctuations in sea levels and drastic climate changes. These cyclical glaciations led to the larger islands in the

Sundaland namely Borneo, Sumatra and Java to be repeatedly connected (formed exposed landmass) and disconnected (isolated islands) to each other and to the mainland of Indochina (Lohman, et al., 2011) consequently influencing the floral and faunal distribution in SE Asia.

The SE Asian region is one of the 17 global biodiversity hotspots housing ca.

25% of global fauna and flora with a high level of endemic species within only a 4% surface area of the earth (Myers, et al., 2000). In terms of freshwater biodiversity

(species-richness and endemicity), SE Asia is the second richest globally after the

Amazon. However, many believed that these figures are greatly underestimated and its diversity potential has been understudied and undervalued (Kottelat, 2002;

Dudgeon, et al., 2006). Within the SE Asian region, there are four biodiversity hotspots

(Figure 1.2); Indo-Burma (Thailand, Cambodia, Laos, Vietnam and Myanmar),

Sundaland (Malaysia, Indonesia), Wallacea (Indonesian islands between Sundaland

2 and Australia and New Guinea) and Philippines (Myers, et al., 2000; Woodruff, 2010).

The non-random distribution of biodiversity in SE Asia is believed to be the result of these environmental fluctuations that shaped the biogeographic patterns over the last few millions of years (Woodruff, 2010).

Figure 1.2: The four biogeographic sub-regions (bioregions or hotspots) in SE Asia. (Adapted from Woodruff, 2010).

In order to have a good resolution for many biogeographic questions such as the link between genealogy and the relative effects of earth and climatic history, freshwater organisms are ideal models due to their restricted movement as they require freshwater habitats, as opposed to terrestrial taxa that are able to disperse widely across continuous habitats (Adamson, et al., 2012). Thus, the biogeographical history of a given region can be inferred based on the observed phylogeographical structure of freshwater taxa because their historical connections among discrete drainages rely directly on the underlying earth and climatic history and also recent sea-level fluctuations (de Bruyn, et al., 2012).

The freshwater halfbeak of the genus Hemirhamphodon was used as a model group in this thesis project. This genus consists of nine morphologically recognized species; H. phaiosoma, H. pogonognathus, H. kuekenthali, H. chrysopunctatus, H. kapuasensis, H. tengah, H. sesamum, H. byssus and H. kecil. They inhabit small freshwater rivers and streams, and are distributed only in specific areas of the

Sundaland. Hemirhamphodon pogonognathus is the most widespread species and distributed in almost all of the Sundaland river basins and shows colour differentiation amongst localities. Most of the Hemirhamphodon species are endemic to a particular region. The species distribution patterns, locally abundant across its natural range and endemism within Sundaland makes this genus suitable for phylogenetic and phylogeography/population studies, and even the potential discovery of new species as well as an environmental indicator.

Advances in genetic technologies and utilisation of molecular markers have provided information on phylogenetic relationships at the interspecific and intraspecific level which is relevant in addressing conservation goals. It also allows better understanding of the relationship between geography and spatial patterns of genetic diversity and thus further provides insight into how historical geography changes have shaped the evolution of regional biotas (Avise, 2009; Hickerson, et al.,

2010).

Mitochondrial DNA has a number of unique characteristics such as maternal inheritance, haploidy and lack of recombination, which is especially useful for interspecific and intraspecific phylogeographic analysis (Avise, 1994; 2000; 2009).

There are many well documented freshwater fish phylogeographic studies which have

4 clarified the phenomena of postglacial colonization in SE Asia utilising mtDNA

(Nguyen, et al., 2008; Jamsari, et al., 2010; 2011; Adamson, et al., 2012; de Bruyn, et al., 2012; 2013; Tan, et al., 2012; 2015). In the last decade, the DNA barcoding (Hebert, et al., 2003a) approach has been widely utilised as an accurate and rapid species identification technique. It was initially proposed to circumvent the lack of traditional taxonomists who mainly rely on morphologically based identification (Stoeckle, 2003;

Blaxter, 2004; Janzen, et al., 2005). The effectiveness of DNA barcoding in identifying both marine and freshwater fishes has been well documented in several studies (Ward, et al., 2005; Hubert, et al., 2008; Rock, et al., 2008). In addition, DNA barcoding also has the ability in discovering deep intraspecific divergence in fish species indicating cryptic speciation (Ward, 2009; Lara, et al., 2010; Sriwattanarothai, et al., 2010; Smith, et al., 2011; Kadarusman, et al., 2012; Hubert, et al., 2012; Puckridge, et al., 2013).

Although mtDNA has proven powerful for genealogical and evolutionary studies, it has some limitations due to its uniparental haploid feature. As only a segment of the story is derived through mtDNA, involvement of nuclear loci/markers is essential in order to understand the complex processes that shape genetic diversity.

According to a review by Hare (2001) based on published results and insights provided by intraspecific nuclear gene trees, incorporation of the diploid nuclear loci in phylogeographical analyses is feasible and productive. Huang, et al. (2008) suggested that Single Copy Nuclear Polymorphic (SCNP) DNA is potentially a powerful molecular marker for evolutionary studies of populations. This is because the data produced by SCNP are more directly comparable to mtDNA sequence data and can be analysed using the same suite of phylogenetic and coalescent-based analyses (Zhang

& Hewitt, 2003; Moyer, et al., 2005; Dolman & Moritz, 2006).

This project aimed to investigate the genetic diversity of the freshwater halfbeak, genus Hemirhamphodon which is endemic to Sundaland river basins, and to test the Paleo-drainage divergence hypothesis as an influential driver of speciation and population distribution of this taxon through phylogeographic analysis. The phylogeographic study involved the most widely distributed Hemirhamphodon species across Sundaland river basins, H. pogononathus, and also two endemic species in northwest Borneo (Sarawak), H. byssus and H. kuekenthali.

1.2 Problem statement

This study utilises two mitochondrial makers (cytochrome b and cytochrome c oxidase subunit I) and two nuclear markers (SCNP: Hp5 and Hp54) to address three aspects as will be discussed below.

Firstly, this project will be focused in assessing the genetic diversity of the genus Hemirhamphodon. This is because the existing classification of this genus is mainly based on morphological approach. It is hypothesised that significant levels of cryptic diversity exist. The application of molecular approach can contribute towards more accurate lineage information and thus provide better interpretation for phylogeography studies in the following chapters. Furthermore, the influence of Paleo- drainage systems in speciation events will also be examined.

Secondly, the wide distribution of H. pogonognathus makes it very suitable for population level studies in genetic variability assessment which would be critical for

6 any future management or conservation program. Furthermore, different colour morphs of the species amongst localities warrant further investigation. Several studies have shown that the differentiation in colouration coupled with morphometric traits might be indicative of a different species (Knowlton, 1993; Gusma˜o, et al., 2000; Tsoi, et al., 2005). According to Anderson & Collete (1991), H. pogonognathus is the only species found in Peninsular Malaysia. Thus, this investigation aims to reveal the existence of cryptic species in Peninsular Malaysia (if existing) that might have been misidentified and assigned to H. pogonognathus. The observed population structure or genetic variation will then be linked with the Paleo-drainage systems to investigate the influences of geographical changes in contemporary gene flow.

Finally, genetic assessment at population level will be conducted for H. byssus and H. kuekenthali, which are endemic in northwest Borneo (Sarawak). This is an attempt to understand the underlying factors that shape the contemporary population distribution. The re-assignment of H. byssus (Tan & Lim, 2013) in a recent study, which had been previously referred to as H. kuekenthali indicates that there might also be in existence of new species in Sarawak. Thus, this investigation also aims to discover the existence of new or cryptic species in Sarawak.

1.3 Objectives

This project is part of an effort to investigate the Sundaland taxogenesis, vicariance taxonomy and phylogeographic link to ecological, geological and geographical partitioning. The freshwater halfbeak of the genus Hemirhamphodon will be used as a model group in this thesis. Therefore, to address the knowledge gaps in this area of study;

Objectives of the study are;

1. To assess the genetic diversity of the genus Hemirhamphodon.

2. To conduct population and phylogeographic studies of H. pogonognathus, H.

kuekenthali and H. byssus in the Sundaland river basins.

3. To examine the influence of historical geographical changes particularly the

Paleo-drainage systems on the contemporary patterns of genetic diversity and

divergence.

To address the aims of this study, this thesis will be divided into six chapters made up of an Introduction (Chapter 1), Literature Review (Chapter 2), three working chapters (Chapters 3, 4 and 5), and general discussion and conclusion (Chapter 6).

Chapter 3 of this thesis describes the preliminary assessment of the species diversity within the genus Hemirhamphodon through DNA barcoding. Considering the dynamic history of this region, the broad distribution and locale-specific polymorphisms of the Hemirhampdodon spp., it is probable that significant levels of

8 cryptic diversity exist. This study aims to elucidate the levels of molecular divergence among Hemirhamphodon spp. and provide accurate lineage information

Chapter 4 describes the investigation on the phylogeography or population structure of H. pogonognathus across Sundaland river basins and examines the influence that historical landscape evolution and ecological traits particularly the

Paleo-drainage systems had on shaping the contemporary patterns of genetic diversity and divergence within this species.

Chapter 5 describes the analyses on the genetic diversity and the contemporary distribution of two endemic species in northwest Borneo (Sarawak), namely H. byssus and H. kuekenthali, in an attempt to understand the underlying factors that are involved.

CHAPTER 2

LITERATURE REVIEW

2.1 Sundaland

Located in SE Asia, Sundaland is one of the global biodiversity hotspots

(Myers, et al., 2000). During the low sea level epochs, the shallow continental

Sundaland or Sunda shelf (Figure 2.1) was a massive landmass encompassing a small area of southern Thailand (provinces of Pattani, Yala, and Narathiwat), Peninsular

Malaysia, Singapore, Borneo (Sarawak, Sabah, Brunei and Kaliamantan), Sumatra,

Java (Voris, 2000) and parts of the Philippine Archipelago (Palawan) (Esselstyn, et al.,

2010). The present day northern Sundaland is connected with mainland SE Asia of

Indochina or Indo-Burma (Indo-Burma hotspot) through the Isthmus of Kra (Sodhi, et al., 2004) and the Kangar-Pattani Line (Whitten, et al., 2004) which act as biogeographic boundaries for numerous species (Sodhi, et al., 2004; Whitten, et al.,

2004; de Bruyn, et al., 2005). The eastern boundary of Sundaland is delimited by

Wallace’s Line and Huxley’s line which runs between Palawan and Luzon, Borneo and Sulawesi, and Bali and Lombok.

Figure 2.1: Shading in dark grey is present day landscape. The Sundaland landmass during the middle Pleistocene period when sea level was 120m below the present level is shaded in light grey. (Adapted and modified from Anderson & Collette, 1991).

2.2 Palaeogeography of Southeast Asia focusing on Sundaland

SE Asia has a very complex geological history which have been discussed in great detail in numerous geological and biogeographical literature (Michaux, 1991;

Ridder-Numan, 1996; Hall, 1998; 2001; 2002; 2009; Morley, 2000; Metcalfe, 2002).

These events have involved the amalgamation and rifting of Gondwana, Eurasian and

Australian plates, with subsequent subduction, collision and arc volcanism at the plate margins (Hall, 2009). The formation of SE Asia (Sunda shelf) through amalgamation of these continental blocks/terranes at different periods of the earth’s history is shown in Figure 2.2 and the geological time scale is shown in Figure 2.3.

Figure 2.2: Distribution of continental blocks, fragments and terranes, and principal sutures of SE Asia. Colour blocks explain the formation of SE Asia (Sunda shelf) through amalgamation of continental blocks/terranes at different periods. LT = Lincang Terrane; CT = Chanthaburi Terrane. (Metcalfe, 2011).

Figure 2.3: The geological timescale. Ages are in million years. (Adapted from Hall, 1998)

In summary, mainland Southeast Asia originated from a few continental blocks mostly derived from the Southern Hemisphere supercontinent of Gondwana. South

China, Indochina and East Malaya blocks which once was part of Gondwana in the

Early Palaeozoic. These had rifted and separated from Gondwana in the Early

Devonian (Metcalfe, 2002; 2005; 2006). The West Sumatra block and possibly the

West Burma block separated from Gondwana in the Devonian (Barber, et al., 2005;

Meyer & Paulay, 2005; Barber & Crow, 2009; Metcalfe, 2009). The Sukhothai Arc

System (combination of the Lincang block in Southwest China, the Sukhothai block in central Thailand and the Chanthaburi block in SE Thailand–Cambodia) was a thin continental basement that formed the margin of the South China–Indochina–East

Malaya superterrane. It was derived from the margin of South China–Indochina–East

Malaya by back-arc spreading in the Late Carboniferous–Early Permian and was then accreted to Indochina by back-arc collapse in the Late Permian. The Sibumasu

Continental block was derived from eastern Gondwana in the Early Permian and accreted to mainland of SE Asia (Metcalfe, 1984; Şengör, 1984). During the Jurassic and Cretaceous, a number of smaller terranes, including the Gondwana-originated southwest Borneo, West Sulawesi and Java rifted from Australia in the Late Jurassic and accreted around the SE Asia promontory. Northeastern Borneo, derived from

South China/ Indochina and the Woyla intra-oceanic arc of Sumatra were added into the SE continental core to form the `Sundaland' in the Cretaceous–Cenozoic (Metcalfe,

2011).

The Sundaland had its greatest land area then, with predominantly southward flowing rivers and widespread, peat swamps (Harley & Morley, 1995) across the south eastern region. Southern Sundaland subsided starting from the later middle Eocene (ca.

42 Ma) onwards. By the end of the Oligocene (ca. 25 Ma) (Figure 2.4), much of the southern region (Java and Kalimantan) was submerged until the middle Miocene (~15

Mya). By the early Pliocene (~5 Mya) the conditions were hot, perhumid, and sea levels were higher around 25 m relative to today’s level (Haywood, et al., 2009; Naish

& Wilson, 2009). At this time the large landmass of Sundaland had been reduced to narrow corridors. However, the exact margins of the shallow sea and low-lying

14 adjacent areas are unknown. During the early Quaternary, the Sundaland experienced radical shifts in sea-land distribution or landscape due to the dramatic fluctuations in sea levels (glaciations) and drastic climate changes. These cyclical glaciations led to the larger islands on the shelf of Borneo, Sumatra and Java to be repeatedly connected

(formed exposed landmass) and disconnected (isolated islands) to each other and to the mainland of Indochina (Lohman, et al., 2011). The Sundaland then remained connected to one another until about 9500 years ago (Inger & Voris, 2001). However,

7000 years ago (Moss & Wilson, 1988), the Karimata Straits was formed separating

Borneo and Sumatra and shortly, afterwards ca. 6000 years ago the seas reached to present levels (Voris, 2000) establishing the present landscape.

Figure 2.4: Paleogeography of the SE Asia region and key habitat availability between 30 and 5Ma. (Adapted from Hall, 2012a)

There were numerous river systems (Paleo-drainages) during low stands in

Sundaland, most probably at the earlier times of low sea level (Plio- and Miocene)

(Lisiecki & Raymo, 2005) such as Malacca, Siam, north Sunda, east Sunda, and

Mekong drainages (Figure 2.5c). These river systems played a major role in the exchange of freshwater species between previously isolated islands and became populated by an extremely rich ichthyofauna. The Malacca drainage is made up of rivers from west Peninsular Malaysia and northeast Sumatra draining northwest into

Andaman Sea. The rivers from Singapore Straits region, central Sumatra and mostly the Gulf of Thailand together formed the Siam drainage which flowed to the north across Sundaland. The largest drainage of Sundaland, the north Sunda drainage flowed into the ocean northeast of Natuna Island from central east of Sumatra together with the Kapuas river from Borneo. The east Sunda drainage uniting rivers from northern

Java, southern Borneo and southern Sumatra drained into Java Sea near Bali. The

Mekong drainage which is believed not to have developed until present (Rainboth,

1996; Carling, 2009) drained southwards from the headwaters on Tibetan Plateau into

South China Sea. These great Paleo river systems and land bridges are now submerged under water due to the later rise in sea level leaving the newly isolated islands as in the present topography (Voris, 2000).

Figure 2.5: The Paleo-drainage systems in Sundaland during the low sea level period based on Voris, (2000). Ma = Malacca, Si = Siam, nS = North Sunda, eS = East Sunda, Me = Mekong. Map was reprinted/modified with kind permission from H. Voris and Field Museum of Natural History, Chicago.

2.3 The cyclical glaciations and the impact on SE Asian biota

Over billions of years of earth's evolution process, various geological processes

(such as plate tectonics, orogeny, volcano eruption) and oscillations occurred in the earth’s orbit. This had led to constant changes in the earth's environment consequently to consistent and periodic climate changes during the Quaternary (2.5mya) (Dam, et al., 2001), which had approximately 50 separate glacial cycles (Woodruff, 2010) causing sea level fluctuations.

Since the Quaternary period, the global climate is beginning to show regular changes. The climate between 2.5 - 1.0mya began to fluctuate under orbital control with an expansion and contraction cycle every 41ka. Then the cycle was transformed into a period of every 100ka at 1mya (Bintanja & van de Wal, 2008; Sosdian &

Rosenthal, 2009) (Fig. 2.5). The longer the cycle period, the greater the temperature differentiation and sea level fluctuation. The100ka cycle can cause sea level changes to reach more than 100 meters. The last glacial maximum (around 27-18kya) was among the most severe of these cycles (Morley, 2000) where the sea level was about

120 meters lower than it is now (Voris, 2000). During most of the Quaternary, the climate was dominated by long glacial phases where the landmass was exposed with cool and dry climate, followed with short warm humid interglacial periods, such as the present day (van der Kaars & Dam, 1995; Morley, 2000).

Figure 2.6: Sea level fluctuations from 3mya with 100ka’s cycle and 41ka’s cycle; 500kya and 20kya fluctuation. (Adopted and modified from Chen, 2016)

These geological processes coupled with sea-level fluctuations during late

Pleistocene have greatly influenced the floral and faunal distribution in Southeast Asia.

(Lohman, et al., 2011). The impact of the climatic fluctuation and the glacial events during the Pleistocene (around 2 million to 10 000 years before the present) resulted in active population differentiation and major shifts in species distributions (Avise,

2000; Hewitt, 2000; Davis & Shaw, 2001). During lowering of sea levels, the connected drainages network reduced the geographic distances between fresh watersheds that would have facilitated the movement of freshwater taxa, thus increased connectivity and range evolution of populations and/or sister taxa. However, during the maxima of sea levels, the populations were delineated into smaller and the isolated populations and formed discrete refugia. These cyclical Pleistocene sea level fluctuations repeatedly separate and fuse the landmasses with each other and/or the mainland and thus acted as a speciation ‘pump’, as populations were repeatedly isolated. On the other hand, the glacial events are also believed to have shrunk population ranges and sizes that may increase the extinction risk thus leading to the decline and eventual extinction of some species (Woodruff & Turner, 2009).

As a result of these cyclical events, the Sundaland basin harbours a rich diversity of endemic species. Numerous studies including on the giant freshwater prawn, Macrobrachium rosenbergii (de Bruyn, et al., 2005), mahseer, Tor spp.

(Nguyen, et al., 2008), catfish, Hemibagrus nemurus (Dodson, et al., 1995) cyprinid fish, Barbodes gonionotus (McConnell, 2004), chevron snakehead, Channa striata

(Adamson, et al., 2012; Tan, et al., 2012; 2015) have supported the hypothesis that

Pleistocene climate changes and the Sundaland paleodrainage rearrangements have influenced population structuring and speciation events.

2.4 Biodiversity and biogeography of Southeast Asia (Sundaland)

According to Whitten, et al. (2004), Sundaland consists of high variation of vegetation including lowland rainforest, beach forest, mangrove, peat swamp and montane forest where the total vascular plant diversity is estimated at roughly 25000 species, and the number of endemics at 15000. Referring to vertebrate diversity, 173 out of 381 mammal species are endemic in Sundaland. A total of 771 bird species with no less than 146 are endemic. Reptile diversity is estimated at some 449 species of which 249 species are endemic. Around 1600 fish species have been documented, where 500 are endemic to one or more of the main islands (Kottelat, 2013; Hubert, et al., 2015). In terms of population and species distribution, SE Asia exhibits high degree of structured and local endemism respectively (Whitten, et al., 2004; Woodruff, 2010;

Wong, 2011).

The description of faunal and floral patterns is essential in order to understand the biotic evolution in a particular region. However, the identification of processes

(event-based approaches) related to it is also important. According to Posadas, et al.

(2006), “ecology and history are indissolubly tied together” in a biogeographical sense.

This is because any change of an organism’s ecology is the response to the historical changes or modification of natural geographical landscapes. The impact of geological and global climatic cycles is believed to be the major force driving the biotic assembly of a given area (Hall, 2009; Lohman, et al., 2011).

Historical biogeography is the study concerned with documenting and explaining the relationship of past events and processes on the observed geographical distributions of taxa/species both in a spatial (geographic) and temporal (history)

21 framework (Whittaker, et al., 2005). It incorporates phylogenetics, geology, palaeontological and climatic information of a given study area, and has developed into a comprehensive research field particularly in recent years (Stelbrink, 2014). A phylogenetic approach could elucidate the evolutionary relationships among species or populations, the historical range expansions, local speciation and extinction dynamics and immigration of species from other places (Heaney, 2000; Wiens &

Donoghue, 2004), all of which contribute to the contemporary spatial distributions of organisms. These events are involved in the processes of dispersal and vicariance which are strongly related to an organism’s ecology (Kodandaramaiah, 2009). The changes of habitats or ecology certainly have been influenced by the historical changes in climate and geomorphology and hence driven changes in patterns of dispersal and vicariance (Wiens & Donoghue, 2004; Woodruff, 2010).

The flora and faunal distributions in SE Asia as observed by many biogeographers are significantly unequal even though experiencing the same tropical climate. Certainly, the current distribution of taxa in SE Asia is the result of the climatic fluctuations and associated sea level changes around the last 2mya. The complex geological history of SE Asia is challenging in interpretation of population/species distribution particularly involving cross regional heterogeneous terrestrial and freshwater taxa. However, the timing and distribution information of various landmasses and islands (Hall, 2009; 2012b) obtained from recent refined geological and tectonic models in this region have greatly assisted in the historical biogeographic studies on SE Asia. The understanding and maintenance of the regional biotic diversity can be enhanced with additional information such as estimates of

22 divergence times, vicariance and dispersal events combined with more empirical geological and biological data (Stelbrink, et al., 2012).

The fragmented nature of SE Asia’s geography is reflected in its extant biota distribution with high degree of regional and local endemism (e.g., Woodruff, 2010;

Wong, 2011). Several biogeographic studies have revealed some of the major zoogeographic transition zone in SE Asia. For example, plants show a different pattern between Continental Asiatic and Malesian floral regions with transition zone occurring at the Kangar-Pattani line (van Steenis, 1995; Morley, 2000; Wikramanayake, et al.,

2002). For freshwater taxa, the widespread Corbiculid clams have shown distinct lineages with several endemic species identified on Sumatra and Sulawesi islands

(Glaubrecht, et al., 2003; 2006; von Rintelen & Glaubrecht, 2006). A study on potamid freshwater crabs by Yeo, et al. (2007) revealed three deep lineages within the genus

Johora on the Malay Peninsula and Tioman Island. On the other hand, Shih, et al.

(2009) revealed that the widely distributed Potamiscinae shows two major clades; an

East Asian/Southeast Asian mainland clade, and a Sunda shelf lslands (Java, Borneo,

Sumatra). Takehana, et al. (2005) observed that the ricefishes (or medaka) of the genera Oryzias and Xenopoecilus could be divided into three well-supported species groups that show a clear geographical structure. The authors further suggested that a barrier for dispersal in ricefishes and other freshwater fishes is attributed to the formation of the Makassar Strait that separates Sulawesi and Borneo.

2.5 Phylogeography

The advancement of genetic techniques such as polymerase chain reaction

(PCR) and automated sequencing technology enables researchers to further their studies from phylogenetics which addresses intraspecific variability into phylogeography (Emerson, et al., 2001). Phylogeography is defined as ‘the study of principles and processes governing the geographical distributions of genealogical lineages, including those at the intraspecific level’ (Avise, 1994). Phylogeographical studies often focus on the within-species phylogenetic interrelationships through DNA molecules and geographic distribution of the phylogenetic groupings (Avise, et al.,

1987). Thus, historical biogeography could be better understood with the knowledge of phylogeographic structure of a species (Dodson, et al., 1995; Randi, 2000).

The distribution of species in space and time, influence of dispersal events, vicariance and demographic fluctuations can be interpreted from the patterns of variation of neutrally evolving genetic lineages (models of population genetics) (Avise, et al., 1987; Avise, 1998; 2009). The history of populations such as estimates of past population sizes, growth rates, and the time to most recent common ancestor among samples can be inferred using the Coalescent Theory of population genetics through simulations of gene genealogies back in time based on current gene variation. (Avise,

1989; Rand, 1996; Kingman, 2000; Kuhner, 2009). When geographical distributions of genetic variation are considered, interpretation on how past and contemporary patterns of gene flow have shaped natural geographical ranges and their contribution to the persistence of extant populations can be made.

The formation of modern phylogeographic patterning is highly related to factors such as Pleistocene ecological perturbations (glaciation, deglaciation, volcanic eruption or aridity) and other extrinsic regimes. Such factors have led to the formation of refugia as expansion and re-colonization events occurred rapidly when the conditions permitted (Hewitt, 1999; Nguyen, et al., 2006b). Furthermore, natural geomorphological events due to geological process such as mountain building, cascades, drainage rearrangement also promote relationship between an organism’s genetic makeup and geographic patterns by creating connections or disconnections among populations (Soltis, et al., 2006; Zemlak, et al., 2008).

The catfish (H. nemurus) population connectivity was documented to be largely influenced by Paleo-drainage rearrangements and exposure of the Sunda Shelf due to the eustatic fluctuations during Pleistocene (Dodson, et al., 1995). While the adjacent and nearby river populations harboured dissimilar haplotypes, the populations from within the Riau Pocket floristic province (Corner, 1960), including the Kapuas

River (western Borneo), Johor River (southern peninsular Malaysia), and Palembang

(Sumatra) shared identical haplotypes as there were from the same Paleo- drainage

(north Sunda drainage). Nguyen, et al. (2006a) also attributed the diversity patterns of the cyprinid genus, Tor to similar influence of landmass dynamics during the

Pleistocene resulting from eustatic fluctuations. Studies of three freshwater halfbeak genus Dermogenys, Nomorhamphus and Hemirhamphodon by de Bruyn, et al. (2013) revealed that species distribution were congruent to Palaeo-drainage boundaries during periods of low sea levels instead of current island boundaries. Parallel phylogeographic correlation to Paleo-drainage configurations have been highlighted in many aquatic

25 taxa (Waters, et al., 2001; McConnell, 2004; Poissant, et al., 2005; Adamson, et al.,

2012).

The above discussion underscores the close link of landscape changes such as drainage re-alignments with phylogeography of freshwater aquatic taxa as it is believed to be the only avenue for historical dispersal. Thus, observed freshwater fauna patterns would provide strong insights in interpreting the biogeographical history of a given region (Lundberg, 1993). The phylogeographic data from freshwater taxa will also help in understanding the past drainage history in the region where an a priori hypotheses (Figure 2.7) can be made through paleodrainage dispersal model regarding the arrangement and evolution of these drainage systems (de Bruyn, et al., 2004).

Figure 2.7: (a) Hypothetical sampling locations of freshwater taxa in SE Asia based on the Paleo-drainage systems by Voris (2000). (b). A priori hypotheses phylogenetic relationships of the freshwater taxa as a result of the occurrences of connectivity facilitated by paleodrainage dispersal model, followed by allopatric speciation due to rising sea levels (Adapted from de Bruyn, et al., 2012). Me = Mekong, Si = Siam, Ma = Malacca, nS = North Sunda, eS = East Sunda.

2.6 The genus Hemirhamphodon

The freshwater fish genus Hemirhamphodon (Bleeker, 1866-1872) belongs to the family Hemiramphidae, subfamily Zenarchopterinae in the order Beloniformes

(Collette, 2004). In Greek, ‘Hemir’ means half and ‘ramphos’ means beak or bill, hence the common name of Hemiramphidae species as halfbeak. The

Zenarchopterinae subfamily consists of five genera and 59 estuarine or freshwater species. The five genera include Dermogenys (13 species), Hemirhamphodon (9 species), Nomorhamphus (16 species), Tondanichthys (1 species) and Zenarchopterus

(approximately 20 species) (Collette, 2004). Two genera (Zenarchopterus and

Dermogenys) inhabit marine or brackish waters (Katherine, 2006) while three

(Hemirhamphodon, Nomorhamphus, Tondanichthys) live in freshwater (Tan & Lim,

2013). Halfbeaks basically have a long needle-like lower jaw (beak) with a shorter and triangular upper jaw (less than half the length of the lower jaw).

The genus Hemirhamphodon is a livebearer (viviparous). It inhabits small freshwater rivers and streams, and is distributed only in specific areas of the Sundaland as shown in Figure 2.8. Hemirhamphodon is unique from other halfbeak genera in having anteriorly directed teeth along the entire length of the lower jaw, posteriorly directed extension of the teeth on the upper jaw (Anderson & Collette, 1991), and pleural ribs originating from the second vertebra instead of the third (Tan & Lim, 2013).

Figure 2.8: Distribution of the genus Hemirhamphodon (Modified from Anderson & Collette, 1991; Tan & Lim, 2013). Shaded areas indicate land area of Sundaland during the middle Pleistocene period (Voris, 2000).

Anderson & Collette (1991) described a total of six species (Plate 2.1a-f) based on the number of vertebrae, dorsal fins, anal fins, and the pattern of melanophores and pigments on the dorsal fin (Figure 2.9). The six species are H. phaiosoma (Bleeker,

1852), H. pogonognathus (Bleeker, 1853), H. kuekenthali (Steindachner, 1901), H. chrysopunctatus (Brembach, 1978), H. kapuasensis (Collette, in Anderson & Collette,

1991) and H. tengah (Collette, in Anderson & Collette, 1991). More recently, three new species (Plate 2.1g-i) from Borneo have been assigned as H. byssus, H. sesamun and H. kecil by Tan & Lim (2013).

Plate 2.1: Photo of male Hemirhamphodon spp., (a) H. phaiosoma; (b) H. pogonognathus; (c) H. kuekenthali; (d) H. chrysopunctatus; (e) H. kapuasensis; (f) H. tengah; (g) H. byssus (live); (h) H. sesamun and (i) H. kecil. All specimens presented are live colouration except H. kapuasensis, H. sesamun and H. kecil.

Figure 2.9: The melanophores on dorsal fin in (a) H. kuekenthali are continuous to the base but not continuous in (b) H. pogonognathus. (Photo was reprinted/modified with kind permission from Piscesilim, 2012)

According to Roberts (1989), H. pogonognathus is the most widespread species in this genus. Its distribution covers almost all of the Sundaland river basins except northern Borneo (Sabah) and even extends out of Sundaland to the Moluccas

(Halmahera) (Anderson & Collette, 1991). Observations by Roberts (1989), Brembach

(1978), Wickman (1981) and Hartl (1983), revealed that H. pogonognathus colouration shows some differentiation amongst some localities. Tan & Lim (2013) observed that H. kuekenthali is only found in the northern region of Sarawak and is believed to be endemic to the river basins in Sarawak. On the other hand, the newly assigned species, H. byssus is also endemic to the Sarawak river basin specifically in the southern region of Sarawak and H. kuekenthali seems to be its closest congener but apparently allopatric. Although eight species (H. byssus, H. kuekenthali, H. kapuasensis, H. tengah, H. chrysopunctatus, H. phaiosoma, H. sesamum and H. kecil)

30 occur in Borneo, H. byssus and H. kuekenthali, which are endemic to Sarawak, are geographically separated from the other six species by mountain ranges traversing the island of Borneo. Hemirhamphodon sesamum that has been discovered in eastward flowing lowland coastal basins of South Kalimantan is most closely related to H. kuekenthali in terms of external morphology (Tan & Lim, 2013). Hemirhamphodon kecil where the term of “kecil” means small in Malay language have smaller adult size compared to its closest congener, H. pogonognathus, is found in lower Mahakam basin in east Kalimantan. Hemirhamphodon kapuasnesis is endemic to the Kapuas river basin in Kalimantan Barat. The most morphologically distinct species is H. phaiosoma with relatively higher numbers of dorsal-fin rays. However, based on reproductive behavior, the H. tengah seems to be the most distinct species. As noted above, the

Hemirhamphodon genus is livebearer or viviparous, but, H. tengah has been observed to lay fertilized eggs (Bork & Mayland, 1998; Grell, 1998; Dorn & Greven, 2007).

In recent years, Hemirhamphodon has generated considerable global interest among the scientific biodiversity community (Lovejoy, 2000; de Bruyn, et al., 2013;

Tan & Lim, 2013) working in the hotspot region of Sundaland. Undoubtedly, the key factors in the rising interest is attributed to its species distribution pattern, locally abundant across its natural range, viviparity and endemism within Sundaland particularly in freshwater habitat which makes it suitable for phylogenetic and phylogeography/population studies, and even the potential discovery of new species.

2.7 Molecular markers in phylogenetics and phylogeographic studies

Based on recent biogeographic studies, it appears that the molecular approach offers a holistic interpretation of information complementing morphology-based methods (Avise, 2009; Hickerson, et al., 2010; Lim, et al., 2011; Lohman, et al., 2011;

Stelbrink, 2014). This advancement is attributed to several reasons including the larger number of characters of DNA sequence data in resolving the relationships among taxa.

Besides, molecular data can be used to assess the confidence in strength of relationships compared to morphologically-based analysis of taxa. In addition, molecular data also allows the different evolutionary hypotheses to be tested (Avise,

2000; Knowles, 2009).

The advances of genetic technologies coupled with the improvements in the understanding of molecular evolution and population genetics have strongly facilitated the elucidation of relationships between geography and spatial patterns of genetic diversity. Besides, it also provides useful insights in understanding the drivers that have shaped contemporary biogeographical patterns (Avise, 2009; Hickerson, et al.,

2010). These improvements have led to increasing number of biogeographic studies involving phylogenetic analyses at the interspecific and intraspecific levels across a variety of taxonomic groups and geographic areas including plants, invertebrates and vertebrates in both aquatic and terrestrial environments (Beheregaray, 2008; Parr, et al., 2012). Molecular markers also permit estimation of divergence times between taxa.

A molecular clock analysis can be implemented based on the principle that changes in nucleotide sequences (mutations) are accumulated and assumed to be fixed in a population over time. Thus, if the genetic distance and rate of nucleotide changes (e.g., substitutions per site per year) are available, the divergence time between two (or more)

32 genetically isolated taxa and also the degree of relatedness can be estimated (Bromham

& Penny, 2003; Pulquério & Nichols, 2007). Hence, interpretation of the regional biogeographical patterns and speciation events can be easily linked with historical geological and climatic changes through statistical analyses and time-calibrated phylogenetic trees.

2.8 Mitochondrial DNA markers- DNA Barcoding and cytochrome b

Molecular markers such as mitochondrial and nuclear DNA are important tools for addressing various biological questions; taxonomic identification, genetic assessment, systematic classification and many others (Witt, et al., 2003).

Mitochondrial DNA (mtDNA) is an ideal genetic marker for studies of population and evolutionary biology of fish (Verheyen & Rubber, 2000; von der Heyden, et al., 2007;

Kochzius & Nuryanto, 2008) and also in identifying and managing stocks of fish species (Martins, et al., 2003) due to its different levels of nucleotide variation.

Additionally, the evolution rate of mitochondrial genes is fairly rapid. Its maternal inheritance and lack of recombination provides an unbroken maternal genealogical record of species. Due to its maternal and haploid nature, it has a small effective population which is approximately one-fourth that of nuclear genome (Avise, et al.,

1987). Under assumptions of neutrality and coalescent theory, these factors make them sensitive to phylogeographic investigations in addressing questions such as population history, demographic changes, range expansions, and spatial patterns of genetic diversity and divergence (Wang, et al., 2000; Schultheis, et al., 2002; Avise, 2009).

The importance of mtDNA analysis has been extensively reviewed in terms of its effectiveness over other techniques for investigating phylogenetic relationships and

33 systematics among fishes (Kocher & Stepien, 1997). There are also many well documented phylogeographic studies of freshwater fishes particularly in SE Asia in clarifying the phenomena of postglacial colonization utilising mtDNA (Nguyen, et al.,

2008; Jamsari, et al., 2010; 2011; Adamson, et al., 2012; de Bruyn, et al., 2012; 2013;

Tan, et al., 2012; 2015).

In year 2003, a standardized segment of the mtDNA (an approximately 650bp region of the cytochrome c oxidase subunit I, COI) was proposed by Hebert, et al.

(2003a) as a global bioidentification system to differentiate the vast majority of animal species, referred to as the DNA-barcoding concept. Its initial objective was to overcome the taxonomic impediment in order to provide accurate and fast species identification. DNA barcodes can overcome the difficulties of a morphologically based identification (Stoeckle, 2003; Blaxter, 2004; Janzen, et al., 2005) where intraspecific phenotypic variation often overlaps in congener, which can lead to incorrect identification or species delineation (Hebert, et al., 2004; Packer, et al., 2009). It is effective at the various life stages of the organism which often appear morphologically different in any one species (Hubert, et al., 2010; Ko, et al., 2013 ) and the various forms of the available biological materials for identification including processed products (Wong & Hanner, 2008; Holmes, et al., 2009; Too, et al., 2016).

Now, DNA barcoding has become a mature field of biodiversity sciences filling the conceptual gap between traditional taxonomy and different fields of molecular systematics (Stoeckle, 2003; Blaxter, 2004; Janzen, et al., 2005) through the establishment of the global information systems in taxonomy as proposed by the international Barcode of Life project (iBOL). An interactive online database system

Barcode of Life Datasystem (BOLD) (www.boldsystems.org) shown in Figure 2.10 has been created for sequence repository, and also offering a collective assembly and curation of the DNA barcode libraries workbench (Godfray, 2006; Ratnasingham &

Hebert, 2007). With the establishment of DNA barcode reference library, users can assign unknown specimen to known species through DNA barcodes. This would be very useful in biosecurity (Collins, et al., 2012), marketplace substitution (Lowenstein, et al., 2010), food product regulation (Becker, et al., 2011), conservation (Francis, et al., 2010), and investigating species interactions (Valentini, et al., 2009).

Figure 2.10: Structure of a specimen record in Barcode of Life Datasystem (BOLD) (Hubert, et al., 2015).

DNA barcoding not only enables accurate species delimitation but also flags the likely existence of morphologically cryptic species (Hebert, et al., 2004; Janzen, et al., 2005; Smith, et al., 2011; Hubert, et al., 2012; Mat Jaafar, et al., 2012). These discoveries were mainly assisted by the ‘barcoding gap’ approach introduced by

Meyer & Paulay (2005). This system proposed that intraspecific COI divergence would be less than interspecific divergence (Figure 2.11). A display of distance data for specimen identification/discovery is dotplots. In this method, for each individual in the data set, the maximum intraspecific divergences is plotted against the minimum nearest-neighbour divergences (Meier, et al., 2008). A 1:1 slope represent the point at which the intraspecific and interspecific differences are the same (i.e. no local barcoding gap) (Figure 2.12).

Figure 2.11: Schematic representation of the Inferred Barcoding Gap. (A) Existence of ‘barcoding gap’ with discrete distributions of intraspecific and interspecific with no overlapping. (B) Overlapping of intraspecific and interspecific distribution, no ‘barcoding’ gap (possibility of cryptic species) (Meyer & Paulay, 2005).

Figure 2.12: An example of the maximum intraspecific divergence plotted against the nearest-neighbour distance of presumed monophyletic species. Plots that fall above the 1:1 line indicate the presence of a “barcode gap” (more than one species exist in the presumed monophyletic species). (Adapted and modified from (Robinson, et al., 2009)

Application of DNA barcoding to both marine and freshwater fishes in the biodiversity hotspot of SE Asia have been documented in several studies (Mohd Lutfi, et al., 2010; Jamsari, et al., 2011; Kadarusman, et al., 2012; Noor Adelyna, 2013;

Hubert, et al., 2015; Duangjai Boonkusol & Wuttipong Tongbai, 2016). It is important to emphasize that although DNA barcoding is effective in species identification and discoveries, it can also be affected by several biases. Therefore, it be must utilised in an integrative framework to incorporate other independent genes, as well as morphological, geographical or ecological data to clearly delimit species (Will, et al.,

2005; Damm, et al., 2010; Padial, et al., 2010).

In addition to the barcoding gene, the cytochrome b (cyt b) is also a very well- studied mtDNA gene in fishes (Rüber, et al., 2007; Bohlen, et al., 2011; Lim, et al.,

2014; Noor Adelyna, et al., 2015). The cyt b gene which contains both slowly and moderately fast evolving regions have been commonly used to infer phylogenies within and among genera (Farias, et al., 2001; Planes, et al., 2001; Slechtova, et al.,

2006; Zaki, et al., 2008). The gene codes for a membrane protein involved in the respiration chain and energy transport, and as a coding region, evolution is constrained by reading frame, resulting in conserved sequence lengths. Several phylogenetic and phylogeographic studies of freshwater fishes have been conducted in SE Asia utilising the cyt b gene including; Kamarudin & Esa (2009) in the study of tinfoil barb

(Barbonymus schwanenfeldii) between Peninsular Malaysia region and Borneo;

Adamson (2012) who investigated the historical drainage evolution based on phylogeography of the snakehead fish C. striata in Mekong basin; Lee & Sulaiman

(2015) in the study of the genetic structure of walking catfish (Clarias batrachus); de

Bruyn, et al. (2013) in the study of the influence of Paleo-drainage in the species distribution of freshwater halfbeak and many more illustrations (Pouyaud, et al., 2003;

Slechtova, et al., 2006; Rüber, et al., 2007; Bohlen, et al., 2011).

2.9 Nuclear DNA markers

While mtDNA is recognized as a powerful tool for genealogical and evolutionary studies of animal populations, it also has limitations. Basically, the analyses of mtDNA genes correspond to a single locus. Thus, the mtDNA only represents a small part of the evolutionary history (Godinho, et al., 2008) that might be able to address issues of incomplete lineage sorting, introgression hybridization, or ancestral polymorphisms (Zhang & Hewitt, 2003). In order to have a more

38 comprehensive understanding of the history and evolutionary potential of populations, the genealogical data from nuclear loci should be an integral part of the analyses.

According to a review by Hare (2001), based on the results and insights provided by intraspecific nuclear gene trees, utilisation of nuclear loci in phylogeographical analyses are feasible and productive. Nuclear gene phylogeographic analysis in complement with mtDNA allows a more comprehensive interpretation of genealogical patterns evolving in genomes in response to the effects of both historical evolutionary processes and contemporary ecological factors on populations and species. There are increasing interest in utilising nuclear sequence data, often multiple loci in analyses of population structure and history. These nuclear loci/markers can be derived from introns (Tchaicka, et al., 2007; Townsend, et al.,

2007), anonymous loci derived from genomic libraries (e.g. Jennings & Edwards, 2005;

Rosenblum, et al., 2007), and microsatellite flanking regions (e.g. Hey, et al., 2004;

Feulner, et al., 2006). Such data are more directly comparable to mtDNA sequence data and can be analysed using the same suite of phylogenetic and coalescent-based analyses (e.g. Zhang & Hewitt, 2003; Moyer, et al., 2005; Dolman & Moritz, 2006).

For a more comprehensive investigation, nuclear markers such as single copy nuclear polymorphic (SCNP) DNA is becoming the marker of choice (Zhang & Hewitt,

2003) due to its common availability in the nuclear genome of eukaryotic organism.

SCNP is a DNA segment that is present in a single copy in the haploid genome and highly polymorphic in the population. Analysis normally involves noncoding regions that is more variable than the coding regions (in contrast to mtDNA markers where coding regions are mostly used because of the lack of introns). In diploid organisms,

39 the sequence copies of a pair of homologous alleles will be present either in a homozygous or heterozygous form. SCNP and mtDNA markers reveal different aspects of a complex story at different evolutionary depths, thus they are considered complementary. Recent advances in technology and statistical power enable nuclear

DNA analyses to be widely applicable e.g. allele-discriminating characterization of

SCNP loci. Thus, there is no doubt on the utility of SCNP markers to address more complex questions, and thereby the sophistication of genetic analyses of populations as shown by data from different genome sequencing projects and studies in model organisms (Zhang & Hewitt, 2003). The utility of SCNP markers for addressing population-genetic questions has been clearly demonstrated in empirical studies, as recently reviewed by Hare (2001). There are also many studies utilising SCNP markers in phylogenetics and phylogeographic studies involving various organisms such as freshwater halfbeak (de Bruyn, et al, 2013) and killifish (Beck, et al., unpublished); insects (Carstens & Knowles, 2007; Huang & Zhang, 2008; Hamm, 2011; Ren, et al.,

2013); lizard (Godinho, et al., 2008) and fungi (Gale, et al., 2011).

2.10 Biodiversity and conservation in Southeast Asia focusing on Sundaland

The separation of Sundaland into islands after the Pleistocene epoch had reduced land area by 50% and the region is presently in a vulnerable refugial state

(Cannon, et al., 2009). Compared to other tropical areas, SE Asia has the highest tropical deforestation rates and highest proportions of threatened flora and fauna (Koh

& Sodhi, 2010; Woodruff, 2010). According to Sodhi, et al. (2010), up to 85% of the region’s species is estimated to become extinct by year 2100 based on the current rates of habitat loss in a “business as usual” scenario. Both terrestrial and freshwater biota

40 in SE Asia are threatened by unprecedented anthropogenic pressures (Sodhi & Brook,

2006) such as high human population growth, indiscriminate deforestation and overexploitation of natural resources. Further threatening forces include the introduction of invasive species (unintentionally or otherwise), pollution, climatic changes, and also natural, geography-induced population bottleneck (Sodhi & Brook,

2006; Sodhi, et al., 2007; Nijman, 2010; Peh, 2010; Wilcove & Koh, 2010). Several studies reported that Asian rivers and wetlands have been seriously degraded due to overfishing, invasion of alien species, erosion and flow regulation which likely led to higher rates of extinction of freshwater biodiversity than its terrestrial relatives

(Dudgeon, et al., 2006; Sodhi, et al., 2007). Thus, intensive conservation efforts should be implemented by not only the regional authorities but also individuals to preserve the region’s biotic novelty and wealth.

In strategising effective conservation efforts, fundamental knowledge such as the regional biodiversity is required in order to understand the genetic structure of the target organisms and the drivers that influence the biodiversity composition. Genetic analysis at interspecific and intraspecific levels will provide information on the genetic diversity (number of species or existence of cryptic species) and also the population structure (e.g. discrete, semi-independent, or panmictic populations) (Carvalho &

Hauser, 1994; Ward, 2000). Inclusion of regional geological information such as historical geology and climate changes will assist in better interpretation of regional biogeography. Understanding these regional processes that shape population configuration is thus essential in maintaining the region’s biological integrity.

In the context of conservation implications, populations that have significant genetic diversity and the most ancestral haplotypes should be protected in each geographic region to conserve the genetic pool. Furthermore, strategic planning of conservation efforts can be assisted through accurate information on taxonomic status,

Evolutionary Significant Unit (ESU) and Management Unit (MU) derived from phylogeographic studies (Moritz, 1994; 2002). Identification of cryptic species and populations with unusually high or unique genetic composition through molecular genetic approaches can also facilitate conservation efforts. According to Bickford, et al. (2007), low conservation priority is frequently given to widespread species.

However cryptic species-level diversity is frequently detected in such species through molecular approaches used in biodiversity research. In addition, molecular approaches also allow incorporation of model-based approaches to be used to infer historical demographic and distributional changes in response to past climates (Lim, et al., 2011).

Such information is critical for conserving the natural resources particularly in this biodiversity hotspot region.

CHAPTER 3

SPECIES DIVERSITY ASSESSMENT OF THE GENUS Hemirhamphodon

THROUGH DNA BARCODING

3.1 Introduction

Biological diversity is believed to be entering an era of mass extinction where it is disappearing worldwide at unprecedented rates (Hubert & Hanner, 2015). Thus, precise taxonomic delineation of species is crucial in the context of biodiversity conservation (Hubert & Hanner, 2015). Traditionally, species description and identification are based on morphological traits, however, the morphological approach alone has intrinsic limitations. Phenotypic plasticity and genotypic variation may mask the morphological characters used in species descriptions, leading to misdiagnoses.

Furthermore, cryptic diversity cannot be easily detected, and variable life stage morphologies and sexual dimorphism may add to the confusion, thus requiring high dependence on experts, often a time consuming enterprise (Hebert, et al., 2003b).

Consequently, molecular methods are proposed to contribute to a new ‘integrative taxonomy’ approach, which can enhance speed and accuracy in species discovery

(Padial, et al., 2010; Hubert & Hanner, 2015).

The DNA barcoding concept was first proposed by Hebert, et al. (2003a), based on an approximately 650 base pair (bp) fragment of cytochrome c oxidase subunit I

(COI), which has since been adopted as the gold standard for global bioidentification to differentiate among animal species. This system proposed that intraspecific COI divergence would be less than interspecific divergence. Thus, the delimitation of

43 species in DNA barcoding is based on this so-called “barcoding-gap”. The evolutionary rate, absence of indels, and limited occurrence of recombination are the primary reasons why COI was chosen as the DNA barcoding gene, in addition to the existence of robust universal primer sets that can amplify across diverse taxonomic groups (Ribeiro, et al., 2012). Rapid identification of potentially unidentified species in global biodiversity assessment and conservation, such as cryptic species, juveniles and organisms with ambiguous morphological characters, is one of the main goals of

DNA barcoding.

Species identification of freshwater and marine fishes through the utilisation of the DNA barcoding method have resulted in a greater than 90% success rate in species discrimination (Ward, et al., 2005; Hubert, et al., 2008; Steinke, et al., 2009;

Valdez-Moreno, et al., 2009; Kim, et al., 2010; Ribeiro, et al., 2012; Puckridge, et al.,

2013). The inconsistencies from expectations as determined by other approaches might be attributable to several factors such as maintenance of ancestral polymorphism, recent divergence among lineages and introgressive hybridization (León-Romero, et al., 2012; Puckridge, et al., 2013). Since the development of DNA barcoding, there has been extensive documentation of species discovery and cryptic species revelations from DNA barcoding data for both freshwater and marine fishes (Zhang & Hanner,

2011; Hubert, et al., 2012; Mat Jaafar, et al., 2012; Rosso, et al., 2012; Ruocco, et al.,

2012).

On the other hand, geographic information has been frequently incorporated in species delimitation. This is because most species exhibit geographic variation, and the divergence between populations within older and geographically widespread

44 species is probably larger than between populations from different but recently separated species (de Queiroz & Good, 1997). In order to determine the true linage separation that occurs within species that is due to clines or isolation by distance, the geographic information is thus necessary. It has been proposed that the abnormal changes or differentiation detected by genetic or phenotypic characters that are indicative of at least partial lineage separation can be better interpreted by utilising geographic information. (e.g., Manel, et al., 2003; Manni, et al., 2004; Miller, 2005).

3.1.1 Current status/ problems

As described in the previous chapter, the genus Hemirhamphodon is widely distributed in Sundaland and shows endemism for some species. Widespread species are given low conservation priority, although cryptic species-level diversity is frequently detected through molecular data in biodiversity research (Bickford, et al.,

2007). Furthermore, observation of different colouration amongst localities and a recent discovery of new species suggests that there is potential for cryptic diversity in this genus. The presence of cryptic species will be problematic in phylogenetic or phylogeography studies as it is not reflected in the current taxonomy (Gómez, et al.,

2007). In order to have better interpretation of the phylogeography study, further investigation and accuracy of lineages information is thus essential.

3.2 Objectives

This chapter focuses on the preliminary assessment of the species diversity within the genus Hemirhamphodon through DNA barcoding. Based on the findings of

Anderson & Collette (1991), Tan & Lim (2013), and observation by Roberts (1989), it is hypothesised that there is potentially high cryptic diversity in this genus due to its broad distribution and locale-specific polymorphisms. Furthermore, the influence of

Paleo-drainage systems in speciation events or genetic divergence will also be assessed.

3.3 Materials and methods

3.3.1 Sample collection, preservation and DNA extraction

Collection of H. pogonognathus, H. kuekenthali and H. byssus species samples were conducted in Peninsular Malaysia, Sarawak and Sumatra as shown in Figure 3.1 and Table 3.1. The information on geographical locations and voucher specimens is shown in Appendix A. No protected species were collected in this study. Permission was granted from the State Forestry Department for field sampling in Peninsular

Malaysia and Sarawak. Samples from Sumatra were obtained from collaborators from

Syiah Kuala University and Riau University in Sumatra.

Figure 3.1: Sampling locations of H. pogonognathus, H. byssus and H. kuekenthali (Modified from Anderson & Collette, 1991), for location abbreviation see Table 3.1.

Table 3.1: Species name, locations, sample size (n), code, regional locations within Peninsular Malaysia, Sarawak (Borneo), and Sumatra; and the newly assigned groups of Hemirhamphodon species (based on the constructed Neighbour-Joining COI gene tree).

Species Locations n Code Present region of locations Newly assigned groups Hemirhampodon pogonognathus Kampung Wang Kelian 5 KWK Northwest-Peninsular Main H. pogonognathus Sungai Teroi 5 TE Northwest-Peninsular Main H. pogonognathus Teluk Bahang 5 TB Northwest-Peninsular Main H. pogonognathus Bukit Pancor 5 BP Northwest-Peninsular Main H. pogonognathus Pondok Tanjung 5 PTG Northwest-Peninsular Main H. pogonognathus Sungkai 5 SK Northwest-Peninsular Main H. pogonognathus Sungai Panjang 5 PJG West-Peninsular Main H. pogonognathus Damansara 5 DM West-Peninsular Main H. pogonognathus Serting Ulu 5 SU Central-Peninsular Main H. pogonognathus Kampung Som 5 SOM Central-Peninsular Main H. pogonognathus Kamung Salong 5 KS Central-Peninsular Main H. pogonognathus Pos Iskandar 5 PI Central-Peninsular Main H. pogonognathus Jeram Pasu 5 JP Northeast-Peninsular Kelantan H. pogonognathus Lata Belatan 5 LB Northeast-Peninsular Kelantan H. pogonognathus Sekayu 5 S Northeast-Peninsular Main H. pogonognathus Rantau Abang 5 RA East-Peninsular Main H. pogonognathus Padang Ah Hong 5 PA East-Peninsular Main H. pogonognathus Panti 5 PT Southeast-Peninsular Main H. pogonognathus Kahang Jemaluang 5 KJ Southeast-Peninsular Main H. pogonognathus Sungai Tarai 3 ST Central-east-Sumatra Main H. pogonognathus Sungai Gerigeng 3 SG Central-east-Sumatra Main H. pogonognathus Sungai Timek 3 STK Central-east-Sumatra Main

Table 3.1: continue…

Species Locations n Code Present region of locations Newly assigned groups H. pogonognathus Sungai Baung 3 SBG Central-east-Sumatra Southern Sumatra H. pogonognathus Jambi 3 JBI Southeast- Sumatra Southern Sumatra H. pogonognathus Jambi Palembang 3 Jmp Southeast- Sumatra Southern Sumatra H. pogonognathus Rambat 3 Rbt North-Bangka Island, Southeast-Sumatra Southern Sumatra H. pogonognathus Merawang 3 MRW Central-Bangka Island, Southeast-Sumatra Southern Sumatra H. pogonognathus Petaling 3 PLG Central-Bangka Island, Southeast-Sumatra Southern Sumatra H. pogonognathus Koba 2 KA South-Bangka Island, Southeast-Sumatra Southern Sumatra H. byssus Kampung Semunin Cina 5 SC Southern-Sarawak, Borneo Southern H. byssus Sungai Stuum Toman 5 SST Southern-Sarawak, Borneo Southern H. byssus Sungai Duyoh 4 SD Southern-Sarawak, Borneo Central H. byssus Tapang Rumput 4 TR Central-Sarawak, Borneo Central H. byssus Sungai Paku 5 SP Central-Sarawak, Borneo Central H. kuekenthali Nangan Lassi 5 LS Central-Sarawak, Borneo Central H. kuekenthali Nangan Lanang 4 LG Central-Sarawak, Borneo Central H. kuekenthali Bukit Kemunyang 5 BK Central-Sarawak, Borneo Central H. kuekenthali Mukah 5 MK Central-Sarawak, Borneo Central H. kuekenthali Tatau 5 TT Central-Sarawak, Borneo Central H. kuekenthali Sungai Kemenda 5 KMD Northern-Sarawak, Borneo Northern H. kuekenthali Sungai Liku 3 LK Northern-Sarawak, Borneo Northern H. kuekenthali Long Lama 5 LL Northern-Sarawak, Borneo Northern H. kuekenthali Sunagi Kejin 5 KJI Northern-Sarawak, Borneo Northern H. kuekenthali Labi-Linei 4 LBI Northern-Sarawak, Borneo Northern H. kuekenthali Limbang 5 LMB Northern-Sarawak, Borneo Northern

Specimens were euthanised with Transmore (NIKA Trading Co.), a commercial fish stabilizer commonly used in aquatic trading in Malaysia before the fin clips were excised from the pectoral fin of the specimens and stored in 95% ethanol for DNA extraction. Specimens were subsequently formalin fixed and preserved in 70% ethanol as vouchers. Voucher specimens were identified based on taxonomic keys by Anderson &

Collette (1991) and Tan & Lim (2013). Images of voucher specimens (Plate 3.1) were captured by digital camera prior to being placed in 95% ethanol for long-term storage.

DNA extractions were conducted using the high-salt DNA extraction protocol (Aljanabi

& Martinez, 1997) with modification and then stored at -20oC until required. For further details on extraction method please refer to Appendix B.

Plate 3.1: Live (above) and preserved (below) colouration of male specimen: (a) H. pogonognathus from Peninsular Malaysia, (b) H. pogonognathus from southern Sumatra (c) H. byssus and (d) H. kuekenthali.

3.3.2 Gene amplification and sequencing

A partial segment of the COI gene of 3 to 5 individuals per location (except only

2 individuals from Koba) were PCR amplified using BIO-RAD T100 Thermal Cycler

(BioRad Laboratories Inc., USA) with primers Fish F2 and Fish R2 (Ward, et al., 2005).

The PCR conditions were conducted as described in Ward, et al. (2005). Full details of primer sequences used are listed in Appendix C and PCR condition is detailed in Appendix

D. The PCR products were electrophoresed on 1.5% agarose gels for band characterization and purified (PCR Clean-up System, Promega, Madison, WI, USA). Sequencing was conducted by a service provider (First Base Laboratories Sdn. Bhd., Malaysia) using an

ABI3730XL Genetic Analyzer (Applied Biosystems, Foster City, CA, USA).

3.3.3 Data Analysis

Several additional species and two out-group sequences totaling 112 individuals from GenBank (Table 3.2) were included in the analyses. Three other documented species,

H. kapuasensis, H. kecil and H. sesamun were not included in the analysis as no specimens were obtained and no GenBank sequences were available.

Table 3.2: A total of 112 COI gene sequences of Hemirhamphondon species obtained from GenBank and the sampling locations.

Species Locations Sequence label GenBank Accession Hemirhamphodon Yong Peng, Johor H_7140 JQ430593.1 pogonognathus H. kuekenthali Kuching, Sarawak H_7136 JQ430639.1 *(H. byssus) H. kuekenthali Sungai Belait, Brunei H_7138 JQ430651.1 H. phaiosoma Sekadau, Kalimantan Tengah H_7135 JQ430652.1 H. tengah Kuala Kurun, Kalimantan H_7129 JQ430653.1 Tengah H. tengah Kuala Kurun, Kalimantan H_7142 JQ430654.1 Tengah H. chrysopunctatus Kuala Kurun, Kalimantan H_7141 JQ430655.1 Tengah H. sp. Sekadau, Kalimantan Tengah H_7125 JQ430557.1 H. sp. Jambi, Sumatra H_7127 JQ430054.1 H. sp. Tampines, Singapore H_7126 JQ430561.1 H. sp. Yong Peng, Johor H_SGK1 JQ430601.1 H. sp. Pekanbaru, Sumatra H_Pek1 - H_Pek10 JQ430558.1, JQ430560.1, JQ430563.1 - JQ430565.1, JQ430567.1, JQ430571.1, JQ430577.1, JQ430582.1, JQ430586.1 H. sp. Bukit Pancor, Penang H_Pen1 - H_Pen11 JQ430573.1, JQ430575.1, JQ430579.1, JQ430584.1, JQ430594.1, JQ430599.1, JQ430602.1, JQ430603.1, JQ430605.1, JQ430607.1, JQ430609.1 H. sp. Tampines, Singapore H_Sel2 - H_Sel3 JQ430559.1, JQ430562.1

Table 3.2: continue…

Species Locations Sequence label GenBank Accession H. sp. Jambi H_Jam1- H_Jam10 JQ430546.1 - JQ430553.1, JQ430555.1 - JQ430556.1 H. sp. Bukit Merah, Perak H_LR_5250 - JQ430569.1, JQ430572.1, JQ430576.1, JQ430578.1, H_LR5260 JQ430580.1, JQ430583.1, JQ430585.1, JQ430600.1, JQ430608.1, JQ430610.1, JQ430611.1 H. sp. Jeram Linang, Kelantan H_LR_5317 - JQ430612.1 - JQ430621.1 H_LR5326 H. sp. Jemaluang, Johor H_LR_5422 - JQ430588.1, JQ430589.1, JQ430591.1, JQ430606.1 H_LR5425 H. sp. Jemaluang, Johor H_LR_5445 - JQ430574.1, JQ430590.1, JQ430592.1, JQ430595.1, H_LR5449 JQ430604.1 H. sp. Sungai Panjang, Selangor H_LR_5452 - JQ430566.1, JQ430568.1, JQ430570.1, JQ430581.1, H_LR5459 JQ430587.1, JQ430596.1, JQ430597.1, JQ430598.1 H. sp. Sebako, Sarawak H_LR_6979 - JQ430634.1 - JQ430638.1 H_LR6980d H. sp. Mukah, Sarawak H_LR_6993 - JQ430640.1 - JQ430650.1 H_LR7003 H. sp. Lundu, Sarawak H_Sar1- H_Sar12 JQ430622.1 - JQ430633.1 Dermogenys sp. Nibong Tebal, Penang D_Pen4, D_Pen7 JQ430526.1, JQ430524.1

*New species name according to the latest species revision by Tan & Lim (2013).

All sequences were edited using MEGA v6.0 (Tamura, et al., 2013). Multiple alignments were evaluated using MUSCLE (Edgar, 2004) that is integrated within MEGA v6.0. Minor adjustments were then made by eye to manually remove any false homologies.

The pairwise comparison matrices were constructed using the Kimura 2 Parameter (K2P) model as it is most widely used in COI barcoding studies (Hebert, et al., 2003b; Hebert &

Gregory, 2005). To check the presence of a “barcode gap” in the dataset, the maximum intraspecific divergences against the minimum nearest-neighbour divergences was plotted.

The Automatic Barcode Gap Discovery (ABGD) species delineation tool on a web interface (http://wwwabi.snv.jussieu.fr/public/abgd/abgdweb.html) with default settings for the K2P distance matrix was employed (Puillandre, et al., 2012) to determine the number of operational taxonomic units (OTUs) based on pairwise sequence distances between individuals within the dataset. A Neighbour-Joining (NJ) COI gene tree was constructed using K2P models with the bootstrap procedure (Felsenstein, 2005) of 10000 pseudoreplicates. A Bayesian Inference (BI) COI gene tree was included to examine any likelihood of different positioning of OTUs. The BI tree was constructed using MrBayes v3.2.2 (Ronquist, et al., 2012) with HKY+G+I model (optimal substitution model under

Bayesian Information Criterion using ModelTest in MEGA v6.0) for 1000 pseudoreplicates. The length of the MCMC chain was 5 million with a sample frequency of 500.

3.4 Results

3.4.1 Genetic diversity, intraspecific and interspecific divergences

A total of 201 individuals were successfully PCR amplified for the COI gene consisting of three morphologically recognized species. All sequences have been deposited in GenBank (accession number: KM405651 – KM405787 and KX216532 –

KX216595) and are available on BOLD public datasets DS-HMRD. The final COI gene segment consisted of 616bp with average nucleotide composition of A = 24%, T =35%,

C = 25% and G = 16%, 231 variable sites, of which 213 were parsimony informative. No insertions, deletions or stop codons after translation were found, indicating that the amplified sequences constitute functional mitochondrial COI sequences and did not harbour any nuclear mitochondrial pseudogenes (numts).

The summary values of genetic divergence across taxonomic level based on K2P are shown in Table 3.3. The mean genetic divergence within species was 2.5% while the within genus divergence was around six times greater at 15.2%. Pairwise comparisons of interspecific divergence among six species (Table3.4) ranged from 8.7% (H. byssus vs H. kuekenthali) to 20.1% (H. phaiosoma vs H. chrysopunctatus - both from GenBank sequences) with a mean of 15.2%. Intraspecific divergence (Appendix E) revealed considerable heterogeneity ranging from low to deep divergence for H. pogonognathus

(0% to 14.8%), H. byssus (0% to 6.7%) and H. kuekenthali (0% to 8.7%). A barcode Gap analysis (Fig 3.2) with two singleton species (H. phaiosoma and H. chrysopunctatus) excluded showed that a barcode gap was present in H. tengah and H. byssus while no barcode gap was present in H. pogonognathus and H. kuekenthali.

Table 3.3: Sample size (n), mean values, ranges of genetic divergences based on K2P across taxonomic levels from 311 sequences of the genus Hemirhamphodon.

Taxon n Min Max Mean Within Species *4 0 4.8 2.5 Within Genus 311 8.7 20 15.2 *Only four species analysed as the other two species were represented by a single individual.

Table 3.4: Pairwise comparisons of the COI gene based on K2P distance among six presumed (morphologically identified) Hemirhamphodon species.

1 2 3 4 5 6 7 1 H_pogonognathus 0.045 2 H_byssus 0.126 0.030 3 H_kuekenthali 0.120 0.087 0.048 4 H_phaiosoma 0.144 0.143 0.148 n/c 5 H_chrysopunctatus 0.171 0.186 0.178 0.201 n/c 6 H_tengah 0.150 0.149 0.145 0.177 0.161 0.000 7 Outgroup_Dermogenys 0.186 0.215 0.206 0.201 0.208 0.150 0.000 n/c = no calculation due to single sample.

Barcode Gap

No Barcode Gap

Figure 3.2: Maximum intraspecific divergence compared with nearest-neighbour distance for four initial presumed morphological Hemirhamphodon species excluding two singleton species. Diagonal line represents 1:1 line to separate “barcode gap” presence and absence area.

3.4.2 Gene tree and OTU counts based on DNA Barcoding

The constructed NJ (Fig 3.3 and 3.4) COI gene tree consisted of nine OTUs including one outgroup. Most individuals formed monophyletic species clusters consistent with their morphological identifications. However, H. pogonognathus was split into three distinct clusters with one major cluster of H. pogonognathus from

Peninsular Malaysia and central Sumatra, a second cluster of a southern Sumatran group, and a third discrete Kelantan (JP, LB and H_LR53) cluster. Further inspection also revealed that population divergence patterns occurred in H. byssus and H. kuekenthali. Hemirhamphodon byssus of the southern group (SC and SST) formed one cluster while H. byssus from the central group (SD, SP and TR) formed another cluster.

For H. kuekenthali, the northern group (KMD, LK, KJI, LL, LBI and LMB) formed one cluster and the central group (LS, LG, BK, MK and TT) formed another cluster.

The BI COI gene tree (Fig 3.5) showed nearly the same OTU clusters except for H. pogonognathus from Kelantan, where it branched as a sister clade of H. kuekenthali, instead of the H. pogonognathus_Main group as shown in the NJ tree.

Figure 3.3: Neighbor-Joining COI gene tree among Hemirhamphodon species generated through K2P. Values at nodes represent bootstrap confidence levels (10000 replicates). A Dermogenys species was employed as an outgroup. The scale bar refers to genetic distance.

Figure 3.4: Neighbor-Joining COI gene tree with representative male specimen live photo from selected locations. The scale bar refers to genetic distance and values on branches are bootstrap values.

Figure 3.5: Bayesian Inference COI gene tree generated through HKY+G+I. Value at nodes represents the Bayesian posterior probability. A Dermogenys species was employed as an outgroup. The scale bar refers to genetic distance.

The number of OTUs generated by ABGD based on K2P varied from 1 to 74 (Fig

3.6). The initial partition at a prior intraspecific divergence (P) (P = 0.0077– 0.0359) produced 9 OTUs, and was in concordance with the NJ and BI trees. The additional OTU identified by ABGD was H. pogonognathus from Rambat.

Figure 3.6: The number of genetically distinct OTUs according to the prior intraspecific divergence value generated by ABGD based on K2P.

3.4.3 Genetic divergence based on the newly assigned groups

Pairwise comparisons (Table 3.5) were computed for the newly assigned groupings according to the constructed NJ COI gene tree and the ABGD results. As expected, intraspecific divergence exhibited a decreased value with a mean of 1.5% ranging from 0% to 4.3%. However, the intraspecific divergence for H. pogonognathus

(southern Sumatra) remained high (4.3%). Pairwise divergence between groups for H. byssus (southern vs central) and H. kuekenthali (northern vs central) were 4.9% and 6.9% respectively. These values might be considered high but did not exceed the minimum nearest-neighbour interspecific divergence (7.9%). On the other hand, pairwise divergence among the three H. pogonongnathus groups ranged from 7.8% to 11.8%, where divergences of the Kelantan group from other populations exceeded the minimum interspecific divergence value. A re-analysis of the barcode gap was conducted (Fig 3.7) for the new grouping. Again, the results revealed absence of a barcode gap in H. pogonognathus from Sumatra, H. kuekenthali from northern and H. byssus from central.

Table 3.5: Pairwise comparison of the COI gene based on K2P distance among newly assigned Hemirhamphodon groupings.

1 2 3 4 5 6 7 8 9 10 11 1 H. pogonognathus_Main 0.011 2 H. pogonognathus_southern_Sumatra 0.078 0.043 3 H. pogonognathus_Kelantan 0.107 0.118 0.004 4 H. byssus_southern 0.123 0.124 0.124 0.010

5 H. byssus_central 0.134 0.133 0.120 0.049 0.029 6 H. kuekenthali_central 0.119 0.121 0.124 0.090 0.095 0.025 7 H. kuekenthali_northern 0.120 0.121 0.120 0.079 0.087 0.069 0.030 8 H. phaiosoma 0.141 0.147 0.163 0.138 0.153 0.147 0.148 n/c

9 H. chrysopunctatus 0.168 0.178 0.178 0.193 0.174 0.174 0.182 0.201 n/c 10 H. tengah 0.149 0.149 0.159 0.147 0.154 0.147 0.141 0.177 0.161 0.000 11 Outgroup_Dermogenys 0.181 0.201 0.201 0.214 0.216 0.208 0.203 0.201 0.208 0.150 0.000 n//c = no calculation due to single sample

Barcode Gap

No Barcode Gap

Figure 3.7: Maximum intraspecific divergence compared with nearest-neighbour distance for newly assigned Hemirhamphodon species grouping. Diagonal line represents 1:1 line to separate “barcode gap” presence and absence area.

The results also revealed that members of the Hemirhamphodon genus appear to be allopatrically distributed. To determine whether the high intraspecific divergence was influenced by Paleo-drainage systems in Sundaland as discussed in de Bruyn, et al., (2013), the NJ COI gene tree clusterings were mapped against the Paleo-drainages (Fig 3.8) as suggested by Voris (2000). The mapping results revealed that only the divergence of H. pogonognathus from southern Sumatra is consistent with the Paleo-drainage (north Sunda) hypothesis. Samples from Malacca and Siam Paleo-drainages formed a single mixed group instead of two groups. Although there is no record of a Paleo-drainage for northwest

Borneo (Sarawak), the divergence of H. byssus and H. kuekenthali seem likely to be congruent with a presently unknown barrier.

Figure 3.8: Sampling location according to Paleo-drainage systems mapped with NJ COI gene tree. Ma = Malacca, Si = Siam, nS = North Sunda, eS = East Sunda, Me = Mekong. H. pogonognathus (shaded circle), H. byssus (shaded triangle) and H. kuekenthali (shaded four-pointed star). Colours of shapes and bars are identical in representing new assigned grouping. The embedded map was reprinted/modified from Voris, 2000 under a CC BY license, with permission from Field Museum of Natural History, Chicago, original copyright 2000.

3.5 Discussion

3.5.1 High levels of intraspecific divergences

The DNA barcoding approach is now widely recognized as an efficient tool to facilitate rapid identification of unidentified or unknown taxa through a DNA barcode reference library and also in assessment for conservation purposes, including of cryptic and microscopic organisms, particularly those with morphologically ambiguous characters (Hebert, et al., 2003a; Hubert & Hanner, 2015). In this study, DNA barcode analysis was used in an attempt to assess cryptic diversity in the genus

Hemirhamphodon. The analysis also permitted insights into the influence of Paleo- drainage systems of Sundaland in driving species diversity.

The high levels of intraspecific divergence in H. pogononathus, H. byssus and

H. kuekenthali suggest that this genus exhibits high cryptic diversity. Several studies have reported the same phenomenon, for instance of the fighting fish Betta

(Sriwattanarothai, et al., 2010), flathead fishes (Puckridge, et al., 2013) and fishes from

Nujiang River (Chen, et al., 2015), which revealed very high species diversity in the absence of apparent morphological differences. The constructed COI gene trees were generally consistent with the current morphological delimitation of Hemirhamphodon species, although several species were only represented by a single sequence here.

However, the clustering of H. pogonognathus split into three well-supported clusters with high levels of divergence, except H. pogononathus from Kelantan (bootstrap value of 59). This result further supports the existence of cryptic diversity within the

H. pogonognathus group. Given its broad distribution in a biodiversity hotspot and the recent documentation of species discovery in Hemirhamphodon (Tan & Lim, 2013),

68 this is therefore not surprising. This pattern could be interpreted as any of these: a recent speciation event, interspecies hybridization, or as (morphological) misidentification (Coyne & Orr, 2004; Ivanova, et al., 2007). When hybridization occurs, the divergent sequence will cluster with the clade of one of the hybridizing species. Conversely, for cryptic species, a new divergent clade will be apparent that is different from that of any currently recognized species (Puckridge, et al., 2013). Our results clearly exhibit the split of different clusters indicating the probable presence of cryptic diversity.

3.5.2 Influence of Paleo-drainage systems

Multiple lineages generated through tree construction demonstrated the potential occurrence of sources from different drainages. The mapping of Sundaland

Paleo-drainage systems against the NJ COI gene tree revealed that Paleo-drainages also likely played a role in the high intraspecific divergence values recovered here.

This was evident in the H. pogonognathus southern Sumatran group which follows the north Sunda Paleo-drainage, diverging from the H. pogononathus_Main group

(combined Malacca and Siam Paleo-drainages) as shown in Fig 3.8. Additionally, the

NJ COI gene tree also shows further splits within H. byssus and H. kuekenthali into multiple lineages. However, these bifurcations did not split out from the currently recognized species to form a new cluster as in H. pogonognathus. de Bruyn, et al.

(2013) found little evidence for Paleo-drainage systems driving divergence in

Hemirhamphodon, but strong support for this mechanism influencing diversity in the halfbeak genus Dermogenys. They postulated that life history strategies

(Hemirhamphodon = forest stream specialists; Dermogenys = brackish water generalists) could have been an important determinant of ability to migrate via these vast Pleistocene paleo-river systems, before subsequent allopatric splitting as sea- levels fell and paleo-systems waned. Although the mapping result demonstrated no obvious Paleo-drainage assigned for northwest Borneo (Sarawak), the different lineages of H. byssus and H. kuekenthali seem very likely to be geographically restricted lineages resulting from unidentified barriers. On the other hand, these divergences could also be associated with ecosystem-dependent adaptive radiation. In addition, the Rajang River seems to have acted as an effective barrier leading to endemism of H. byssus and H. kuekenthali with H. byssus only found in the south of the Rajang basin and H. kuekenthali in the north.

3.5.3 Evidence of cryptic species

The tree topology showing several distinct clusters within certain species coupled with high levels of intraspecific diversity indicated probable occurrence of cryptic species (Chen, et al., 2015). High genetic divergence within nominal species can be interpreted as misidentification, or more importantly as cryptic or unrecognized speciation events (Kadarusman, et al., 2012; León-Romero, et al., 2012; Puckridge, et al., 2013). There are several criteria proposed for species delineation based on the

DNA barcoding approach. Hebert, et al. (2004) proposed the ‘10X rule’ as an indicator of cryptic speciation. On the other hand, Ward, et al. (2009), who analysed barcode data from about 1000 fish species, showed that individuals were much more likely to be congeneric than conspecific at a level of 2% distance or greater. Another criterion is the use of the barcode gap, which is the distance or gap between the maximum

70 intraspecific and minimum interspecific distances (Hebert, et al., 2004; Meyer &

Paulay, 2005; Hajibabaei, et al., 2006; Meier, et al., 2006; Puillandre, et al., 2012).

Barcode gap analysis for this dataset for both initial groupings and new groupings (Figure 3.2 and Figure 3.6) revealed that no barcode gap was present in H. pogonognathus, H. kuekenthali and H. byssus, which indicated the probable existence of more than one species within each of these taxa. In addition, the ABGD method generated 9 OTUs, which is nearly concordant with the COI gene trees. The H. pogonognathus complex formed three OTUs, which potentially implies detectable intraspecific diversity. Thus, referring to the results of the genetic distance, COI gene trees and barcode gap analyses, the existence of high cryptic diversity among our dataset is apparent.

The three genetic lineages across H. pogonognathus most likely represent species-level taxa, suggesting that H. pogonognathus consists of at least two distinct species. Hemirhamphodon pogonognathus was the most widely distributed species and showed high intraspecific divergence values up to 14.8% with a mean of 4.5 %, separated by three geographical splits (Figure 3.8). Further re-grouping revealed that the H. pogonognathus complex exhibited deep divergence among the three H. pogonongnathus groups, ranging from 7.8% to 11.8% (Table 3.4), where divergences of the Kelantan group from the other two groups exceeded the minimum interspecific divergence (7.9%). No obvious morphological characters were found to distinguish them into different species even though colour differentiation was observed among some localities (Plate 3.2). In fact, the genetically distant Kelantan group shared the same colour pattern with the Main group. In addition, the Kelantan group was separated from the Main cluster to form its own clade in the COI gene trees, further

71 supporting the probable existence of cryptic species in the H. pogonognathus group.

Thus, it is proposed that H. pogonognathus from Kelantan has the potential of being a new species record.

Plate 3.2: Different colouration among localities of male H. pogonognathus.

The six morphological species included in this study generated 9 OTUs. This

DNA barcoding study shows the high potential in cryptic species assessment particularly within ‘hyperdiverse’ SE Asia. The species status for H. byssus and H. kuekenthali remains unclear, and each could be considered as a species complex.

Based on this study, the genus of Hemirhamphodon is proposed to consist of at least

10 species namely H. pogonognathus, H. kuekenthali, H. byssus, H. phaiosoma, H. chrysopunctatus, H. tengah, H. kapuasensis, H. sesamun, H. kecil (the last three species not included in this analysis) and the newly proposed H. pogonognathus sp. of

Kelantan. Nevertheless, further investigations with combined molecular and morphological approaches, and also population level analyses with larger sample sizes are needed to clarify Hemirhamphodon taxonomy.

3.5.4 Conservation and management

According to Lambert & Collar (2002), the lowland tropical rainforests are the most biodiverse. Unfortunately, the rapid economic development in SE Asia such as transformation of forests into palm oil plantations and direct logging have been led to the serious destruction of large parts of tropical forest habitat, mainly lowland rain forests (Curran, et al., 2004; Sodhi, et al., 2004). Since the genus Hemirhamphodon is not involved in commercial or ornamental fish trading (personal observation), it is thus not exposed to overfishing threat. Consequently, habitat destruction seems to be the main threat to this genus which inhabits forest stream. Habitats destruction could lead to high risk of loss in species diversity due to the high cryptic diversity and endemism of this genus. Hence, conservation and management efforts is urgent in order to sustain the species diversity of the genus Hemirhamphodon.

3.6 Conclusions

In conclusion, the present study highlights the efficiency of DNA barcoding in species diversity assessment. In addition, the findings highlight the high cryptic diversity of this genus, which is in agreement with the initial hypothesis. However, a more integrated study including molecular and morphological approaches needs to be conducted to resolve the issue of paraphyletic or species complexes of several

Hemirhamphodon species. Lastly, more studies of freshwater fishes with a range of distributional, life history and other datasets (genetic, phenotypic and geographical) incorporating multiple research tools need to be conducted in order to have a better understanding of the drivers and maintenance of biodiversity within the SE Asian region.

CHAPTER 4

POPULATION STUDY AND PHYLOGEOGRAPHY OF Hemirhamphodon

pogonognathus IN SUNDALAND RIVER BASINS

4.1 Introduction

According to Roberts (1989), H. pogonognathus (Plate 4.1) is the most widespread species in the genus Hemirhamphodon. Its distribution covers almost all of the Sundaland river basins except northern Borneo (Sabah) and even extends out of

Sundaland to the Moluccas (Halmahera) (Anderson & Collette, 1991). Observations by Roberts (1989), Brembach (1978), Wickman (1981) and Hartl (1983), revealed that

H. pogonognathus colouration shows differentiation amongst some localities. The wide distribution of H. pogonognathus makes it very suitable for population level studies specifically in genetic variability assessment, which would be critical for any future management or conservation program. Furthermore, its variable colour patterns also warrant further investigation. Several studies have shown that differentiation in colouration coupled with morphometric traits from the typical characteristics might be indicative of a different species (Knowlton, 1993; Gusma˜o, et al., 2000; Tsoi, et al.,

2005). As revealed in Chapter 3, the three genetic lineages across H. pogonognathus most likely represent species-level taxa, suggesting that H. pogonognathus consists of at least three distinct taxa (very likely species level). High intraspecific divergence with values up to 14.8% and a mean of 4.5 %, separated by three geographical splits

(Figure 3.8, in Chapter 3) were observed. Further re-grouping revealed that within the

H. pogonognathus complex, divergences of the Kelantan group from the other two groups exceeded the minimum interspecific divergence (7.9%) (Table 3.4, Chapter 3).

No obvious morphological characters were found to distinguish them into different species even though colour differentiation was observed among some localities (Plate

3.2, Chapter 3). In fact, the genetically distant Kelantan group shared the same colour pattern with the H. pogonognathus_Main group. The separation of Kelantan was sugggested due to the probable existence of cryptic species in the H. pogonognathus. de Bruyn, et al. (2013) found little evidence for Paleo-drainage systems driving divergence in Hemirhamphodon. The authors postulated that life history strategies could have been an important determinant of ability to migrate via these vast

Pleistocene paleo-river systems, before subsequent allopatric splitting as sea levels fell and paleo-systems waned. This was evident in the salt-tolerant Dermogenys spp., which inhabit brackish water, usually estuarine areas and their speciation is more influenced by Paleo-drainage systems in contrast to Hemirhamphodon speciation

Plate 4.1: Live (above) and preserved (below) colouration of male H. pogonognathus specimen.

The initial findings as reported in Chapter 3, highlights that taxonomic issues within H. pogonognathus remain unclear. Thus, a further study involving population study and phylogeography encompassing increased samples over the Sundaland river basins is required. Analyses of the phylogeography of H. pogonognathus across the

Sundaland river basins coupled with the findings documented in Chapter 3 regarding the species distribution would be valuable in understanding the evolutionary history and the ambiguous status of this taxon. In a conservation context, comprehensive and precise information on the genetic diversity of the species is vital; existence of cryptic lineages is not reflected in the taxonomy and systematics of the organisms in question

(Gómez, et al., 2007) and therefore missed out in the strategic planning. Additionally, as it is well known, widespread species is often given low conservation priority in biodiversity research (Bickford, et al., 2007). Thus, documenting intraspecific diversity through population study or phylogeographic study is necessary as it can assist conservation efforts in identifying precise genetic diversity status and uncover the geographic areas with unusually high or unique genetic diversity through molecular approaches (Fraser & Bernatchez, 2001; Habel, et al., 2011).

According to Avise (2000; 2009) phylogeography is integrally linked to landscapes and to landscape history. It is well recognised that intra-species distribution of genetic diversity across its natural geographical range could be strongly influenced by earth history and/or environmental and ecological processes. Paleo-drainage flowages (Figure 2.5, Chapter 2) across Sundaland during Pleistocene have facilitated connectivity to varying degrees among populations of several freshwater taxa (Dodson, et al., 1995; McConnell, 2004; de Bruyn, et al., 2013). The findings reported in Chapter

3 are also congruent with these earlier studies. However, although this pattern is

77 common (Luo, et al., 2004; Su, et al., 2007; Fuchs, et al., 2008; Vidya, et al., 2009), it is not universal (Gorog, et al., 2004; Esselstyn & Brown, 2009) and requires a multi- species comparative analysis for a holistic study.

During the middle Pleistocene when the sea level was at its lowest at approximately 120m lower than today, there were numerous river systems (Paleo- drainages) during low stands in Sundaland such as north Sunda, east Sunda, Siam and

Malacca drainages (Voris, 2000) (Figure 2.5, Chapter 2) which played a major role in the exchange of freshwater species between previously isolated islands and these river basins became populated by an extremely rich ichthyofauna (de Bruyn, et al., 2012).

These great river systems and land bridges are now submerged under water leaving the newly isolated islands as in the present topography due to the rise in sea level

(Voris, 2000).

As a result of differences in sea levels, the isolated populations of biota and many fish groups now harbour numerous endemic genera and species in the Sundaland river basins (Lohman, et al., 2011). Much work has been carried out to understand the evolutionary events in Sundaland. These have involved documenting past events and biogeographical patterns (de Bruyn, et al., 2012) that are believed to be the result of environmental fluctuations over the last few millions of years. According to Woodruff

(2010), how species responded to fluctuating environmental conditions during the

Pleistocene can be revealed through genealogical studies of the spatial and temporal dynamics of populations. In addition, many population structures or speciation events were influenced by Pleistocene changes and have been well documented (Nguyen, et al., 2008; Jamsari, et al., 2010; 2011; Adamson, et al., 2012; de Bruyn, et al., 2012;

2013; Tan, et al., 2012; 2015). In order to have a good resolution for many biogeographic questions, freshwater organisms are ideal models due to their restricted movement as they require freshwater habitats, as opposed to terrestrial taxa that are able to disperse widely across continuous habitats and consequently making it difficult to determine their biogeographic histories (de Bruyn, et al., 2012).

Population structure assessment is fundamental for the management of intraspecific genetic diversity and sustenance and conservation of fisheries (Nelson &

Soulé, 1987). Population genetic analyses can provide information on demographic history such as effective population size (Fu, 1997), changes in population size (past bottlenecks, range expansions related to habitat availability, etc.; Rogers &

Harpending, 1992), population structuring that may result from a probable combination of life-history characteristics and complex historical paleodrainage rearrangements (So, et al., 2006b; Hurwood, et al., 2008; Adamson, et al., 2009).

Currently, population genetic analyses based on coalescent theory (Kingman, 1982a;

1982b), including minimum-spanning trees/networks (Smouse, 1998) or nested clade analysis involving spatial and temporal tests of association for detection of geographical structure with increased statistical power are complementing traditional analyses such as fixation indices (Slatkin, 1989; Hudson, et al., 1992).

Mitochondrial DNA is an ideal genetic marker for studies of population and evolutionary biology of fish (Verheyen & Rubber, 2000; von der Heyden, et al., 2007) and also in identifying and managing stocks of fish species due to its different levels of nucleotide variation (Martins, et al., 2003). It is maternally inherited without recombination, thus suitable for tracing the individual’s maternal lineage. Its haploid

79 nature makes it more sensitive for detecting population structure (Wang, et al., 2000;

Schultheis, et al., 2002). The smaller effective population size of mtDNA enables it to better detect demographic events compared to nuclear markers that require larger effective population size. The cyt b gene is one of the markers of choice for these purposes. It is a very well-studied mtDNA gene particularly in phylogenetic and phylogeographic studies of freshwater fishes (Kamarudin & Esa, 2009; Adamson, et al., 2012; Lee & Sulaiman, 2015) and has been previously described for

Hemirhamphodon (de Bruyn, et al., 2013).

While mtDNA is recognized as a powerful tool for genealogical and evolutionary studies of animal populations, it also has limitations. Basically, the analyses of mtDNA genes correspond to a single locus and thus elucidate only a small part of the evolutionary history (Godinho, et al., 2008). In order for a more comprehensive understanding of the history and evolutionary potential of populations, the genealogical data from nuclear loci are recommended to be included in the analyses.

Due to its diploid nature, nuclear DNA furnishes replicate samples for inferring demographic history as well as the coalescence process. To reduce the variance of parameter estimates, sampling multiple nuclear markers is preferred (Felsenstein, 2005;

Brito & Edwards, 2009; Hey, 2010). The utility of single copy nuclear polymorphic

(SCNP) DNA markers for addressing population-genetic questions has been demonstrated clearly in many empirical studies, as reviewed by Hare (2001) and also in Hemirhamphodon (de Bruyn, et al., 2013). A combination of SCNP and mtDNA markers is thus complementary in demographic studies as they reveal different aspects of a complex story at different depths of perception.

4.2 Objectives

This study uses a combination of the mtDNA marker, cytochrome b gene (cyt b) and two nDNA markers (SCNP: Hp5 and Hp54) to address three aspects. Firstly, a population study was conducted to describe the population genetics and phylogeography of the widespread H. pogonognathus species. Secondly, to assess how the Pleistocene events or geomorphology has influenced the distribution patterns of this species in the Sundaland river basins. Thirdly, it was proposed in Chapter 3 that

H. pogonognathus from Kelantan has the potential of being a new species. Thus, as a follow up, a detailed population-level study among the diverse populations of H. pogonognathus focusing on phylogeography and demographic history analyses was conducted to clarify its taxonomic status.

4.3 Materials and methods

4.3.1 Sample collection, preservation and DNA extraction

Samples were collected from 25 sites as shown in Figure 4.1 and Table 4.1.

Wherever possible, fifteen individuals were collected from each sampling site. A sample of H. kuekenthali was employed as an outgroup in the analyses. Specimen preparation and DNA extraction of the individuals are as described in Chapter 3.

4.3.2 Gene amplification and sequencing

Ten to fifteen individuals per site were PCR amplified for the mtDNA protein coding cyt b gene region (cyt b) and two nuclear DNA markers (SCNP: Hp5 and Hp54).

PCR amplification conditions for cyt b was conducted as described in Lovejoy (2000) with primers L14504 and Cyt-b3-3’ (Miya & Nishida, 2000; Lovejoy, 2000). SCNP markers (HP5 and Hp54) were PCR amplified according to de Bruyn, et al. (2010).

Full details of primer sequences used are listed in Appendix C and PCR conditions are detailed in Appendix D. The PCR products were electrophoresed on 1.5% agarose gels for band characterization and purified (PCR Clean-up System, Promega, Madison, WI,

USA). Sequencing was conducted by a service provider (First Base Laboratories Sdn.

Bhd. Malaysia) on an ABI3730XL Genetic Analyzer (Applied Biosystems, Foster City,

CA, USA).

Figure 4.1: Sampling locations of H. pogonognathus in contemporary geography landmass (Modified from Anderson & Collette, 1991), for location abbreviation see Table 4.1.

Table 4.1: Sampling locations, sample size (n), code, the hypothetical Paleo-drainages in Sundaland where the sampling locations are situated, and regional locations within Peninsular Malaysia and Sumatra. Ma = Malacca, Si = Siam, nS = North Sunda

Locations n Code Paleo- Present region of drainages locations Peninsular Malaysia 1. Kampung Wang Kelian 15 KWK Ma Northwest-Peninsular 2. Sungai Teroi 10 TE Ma Northwest-Peninsular 3. Teluk Bahang 15 TB Ma Northwest-Peninsular 4. Bukit Pancor 15 BP Ma Northwest-Peninsular 5. Pondok Tanjung 15 PTG Ma Northwest-Peninsular 6. Sungkai 15 SK Ma Northwest-Peninsular 7. Sungai Panjang 15 PJG Ma West-Peninsular 8. Damansara 15 DM Ma West-Peninsular 9. Serting Ulu 10 SU Si Central-Peninsular 10. Kampung Som 15 SOM Si Central-Peninsular 11. Kamung Salong 15 KS Si Central-Peninsular 12. Pos Iskandar 15 PI Si Central-Peninsular 13. Jeram Pasu 15 JP Si Northeast-Peninsular 14. Lata Belatan 15 LB Si Northeast-Peninsular 15. Sekayu 15 S Si Northeast-Peninsular 16. Rantau Abang 13 RA Si East-Peninsular 17. Padang Ah Hong 15 PA Si East-Peninsular 18. Panti 10 PT Si Southeast-Peninsular 19. Kahang Jemaluang 10 KJ Si Southeast-Peninsular Sumatra 20. Sungai Tarai 15 ST Ma Central-east-Sumatra 21. Sungai Sekor 15 SKR Ma Central-east-Sumatra 22. Jambi 15 JBI nS Southeast- Sumatra 23. Jambi Palembang 10 JMP nS Southeast- Sumatra 24. Rambat 15 RBT nS North-Bangka Island, Southeast-Sumatra 25. Merawang 15 MRW nS Central-Bangka Island, Southeast- Sumatra

4.3.3 Data sorting and haplotype generation

4.3.3(a) Mitochondrial DNA analysis

All mtDNA cyt b sequences were edited using MEGA v6.0 (Tamura, et al., 2013).

Multiple sequence alignments were evaluated by using MUSCLE (Edgar, 2004) which is integrated in the same program. Minor adjustments were then made by eye to manually remove any false homologies. DNA Sequence Polymorphism (dnaSP) program v5.10.01

(Librado & Rozas, 2009) was used to generate haplotypes.

4.3.3(b) Nuclear DNA analysis

The nDNA markers SCNP (Hp5 and Hp54) vary in length among individuals due to the presence of indels in heterozygous sequences. So more complex methods were utilised to generate robust sequences. Haplotypes from diploid nuclear sequences of SCNP

(HP5 and HP54) were inferred using Bayesian inference, PHASE v2.1.1 (Stephens, et al.,

2001; Stephens & Scheet, 2005). The PHASE input files were generated from FASTA sequence alignments using SeqPHASE (www.mnhn.fr/jfflot/seqphase: Flot, 2010).

PHASE was run twice with different starting seeds that were run for 100 iterations preceded by a burn-in phase of 100 iterations. To be accepted for further analyses, the

PHASE results must reach a probability threshold of > 0.7 (Harrigan, et al., 2008). Further haplotype assignment was conducted using DNA Sequence Polymorphism (dnaSP) v5.10.01.

4.3.3(c) Genetic diversity

The datasets were tested for optimal substitution models under the Bayesian

Information Criterion (BIC: Schwarz, 1978) using the ModelTest that is incorporated in

MEGA v6.0. However, when the best model was unavailable, the closest available substitution model was utilized for subsequent analyses. Genetic distances among populations and among major clades identified in the tree and network analyses were calculated using MEGA v6.0 with the best substitution model. For each locus, saturation test was performed and the saturation plots (the number of transition and transversion observed between pairs of sequences against p-distance) were drawn using DAMBE (Xia

& Xie, 2001). This is to avoid weak interpretation. If the nucleotide substitution was saturated, it may mask the true evolutionary rate, and hence the true level of divergence

(Kocher & Carleton, 1997).

To describe DNA variation in each population, ARLEQUIN v3.5 (Excoffier &

Lischer, 2010) was used to calculate three estimations of diversity measurement. These included nucleotide diversity (π) (Tajima, 1983), the mean number of pairwise nucleotide differences between all individuals in the sample; haplotype/gene diversity (h), the probability that two randomly chosen haplotypes within a sample will be different (Nei,

1987); and theta S (ϴs) (Watterson, 1975), which is a measure of the number of segregating sites among haplotypes in a sample.

4.3.3(d) Gene tree construction

Genetic relationships among haplotypes were assessed by constructing gene trees through Neighbor-Joining (NJ) and Bayesian Inference (BI) methods. An individual of H. kuekenthali was employed as outgroup to root the tree. The NJ gene trees were constructed with the optimal substitution model as derived from the ModelTest for both cyt b and

SCNP markers using MEGA v6.0. Confidence limits were assessed with the bootstrap procedure (Felsenstein, 2005) of 10000 pseudoreplicates. The BI tree was implemented in MrBayes v3.2.2 (Ronquist, et al., 2012). Different levels of partitioning among codons for cyt b was a priori tested using PartitionFinderV1.1.1 (Lanfear, et al., 2012). All rates and partitions were unlinked and allowed to vary across partitions, for implementation in the BI tree construction. For SCNP markers, the best model was also chosen after

ModelTest. Parsimony informative indels (gaps common to two or more individuals) were included as binary codes (presence/absence) data following the “Simple Indel Coding” method of Simmons & Ochoterena (2000) and were allowed to vary according to the beta distribution. According to Dwivedi & Gadagkar (2009), accuracy in phylogenetic reconstruction can be improved by incorporating the parsimoniously informative signal from indels data by treating gaps as binary characters. The length of the MCMC chain was

10 million with a 4 by 4 nuclear model and a sample frequency of 1000. The first 25% of samples were discarded as burn in and the summarised consensus trees with posterior probabilities for clades were viewed with FigTree v1.4.0 (Rambaut, 2006).

4.3.3(e) Haplotype network

Minimum Spanning Network (MSN) that explains the relationships among the haplotypes and spatial distribution were constructed by using the maximum parsimony- median joining (MP-MJ) method as implemented in NETWORK v4.2.0.1 (Bandelt, et al.,

1999). Compared to phylogenetic trees, Network estimation with the MP-MJ method is regarded as a more reliable method in providing true genealogy for interpreting the relationships among haplotypes especially when node (internal) haplotypes are absent from the data set (Cassens, et al., 2005).

4.3.3(f) Population structure

Analysis of molecular variance (AMOVA) was performed using ARLEQUIN v3.5 to test the significance of pairwise ФST distances and population structure. Population pairwise ФST (Excoffier, et al., 1992) explains genetic variation among sites for examining the level of differences among populations and spatial population structuring. The analysis used Tamura & Nei (1993) with Gamma correction distance method and significance of each pairwise comparison was tested with a nonparametric permutation procedure with

10000 Markov steps. Bonferroni correction was used to compensate for simultaneous tests in estimating significance with a global significance level of 0.05. Hierarchical AMOVA was also conducted using the same program to infer the relative contribution of variances among groups within total (ФCT), among populations within groups (ФSC). Two analyses were conducted based on partitioning of groups; populations were grouped according to

88 either the Paleo-drainages systems (Malacca, Siam, east Sunda and north Sunda drainages) as in Table 4.1 or according to the DNA lineages that were recovered in the phylogenetic analysis.

Mantel’s test (Mantel, 1967) for isolation by distance (IBD) was conducted to test for significance in spatial structuring in patterns of genetic differentiation (the significance of the correlation between pairwise ФST and geographical distance). Geographical distances were measured with Google Earth v7.1.7.2606 (2016). Distances representing the shortest path between two populations follow the hypothetical Paleo-drainage systems

(Figure 4.2). Calculations were performed in ARLEQUIN v3.5 with 10,000 permutations.

Figure 4.2: Map showing the hypothetical Paleo-drainages in Sundaland during the low sea level around 20,000 BP and the sampling sites of H. pogonognathus. Number at the sampling sites is following the location sequence from top to bottom in Table 4.1. (Modified from Irwanto, 2015).

4.3.3(g) Historical demographic analysis

To evaluate the demographic history of the populations, the following analyses were conducted for each population; neutrality tests such as deviation from mutation-drift and gene flow-drift equilibrium, mismatch distribution analysis and the distribution of the number of mutation differences between pairs of sequences (Rogers & Harpending, 1992).

The Tajima’s D (Tajima, 1989) and Fu’s Fs (Fu, 1997) were used to check for deviation from neutrality. Zero value of Tajima’s D indicates a mutation-drift equilibrium; negative values indicates population growth or selective sweep, whereas positive values are a signature of genetic subdivision, population contraction, or diversifying selection. Fu’s Fs test compares the number of haplotypes observed with the expected number of haplotypes under the assumption of an infinite-site model without recombination (Fu, 1997). The R2 test (Ramos-Onsins & Rozas, 2002) which utilises the mismatch distribution of pairwise differences in detecting past population demographic changes was conducted using dnaSP v5.10.01. The Fu’s Fs and R2 performs better in revealing population demographic changes for small sample size (Liao, et al., 2010) with large negative values of Fu’s Fs and small positive values of R2, a signature for population growth. Both Tajima’s D and

Fu’s FS were calculated using ARLEQUIN v3.5.

A unimodal pattern and Poisson distribution of mismatch distribution is expected for populations that have recently undergone a demographic expansion (Slatkin & Hudson,

1991), while long term constant population size was expected to display a multimodal distribution (Rogers & Harpending, 1992). The demographic historical expansion of each population was assessed using dnaSP program. Harpending’s raggedness index (Hri)

(Harpending, 1994) was also used to test against the null distribution of recent population expansion with ARLEQUIN v3.5. A significant Hri value means a rejection of population expansion model (Schneider & Excoffier, 1999).

Pattern of historical demography for overall population (total sample dataset of 25

H. pogonognathus populations) inferred from estimates of effective population size over

91 time was reconstructed with Extended Bayesian skyline plots (EBSPs) (Heled &

Drummond, 2008) methods as implemented in BEAST v1.8.0 (Drummond, et al., 2012).

Model of evolution was assigned based on the result of ModelTest. The rate of molecular clock was assigned based on the estimation as discussed below. Relaxed molecular clock with uncorrelated lognormal distribution (Drummond, et al., 2006) was assigned for cyt b and strict clock model for SCNP loci (de Bruyn, et al., 2013). The analysis was run for

25,000,000 generations with parameters logged every 2500th generation. The log files were examined using TRACER v1.5 (Rambaut & Drummond, 2007). If the effective sample size (ESS) reached >200 for all parameters (Drummond, et al., 2006) it indicates stationary. Multiple independent runs were conducted in order to reach suffusion ESS.

The generated log csv file was used to construct EBSPs using Microsoft Office Excel.

The substitution rate (r) was estimated from the net genetic distance/divergence

(Da) between lineages generated by MEGA v6.0. The mean time since the most recent common ancestor between the lineages (tMRCA) was set at 3.3 Mya as referred to de

Bruyn et al. (2013). For example, given the net genetic divergence (Da) = 0.069, the substitution rates will then be,

r = [(Da /2) / tMRCA]

= [(0.062/2)/3.3]

= 0.00939 (equal to 0.93% per million years)

4.4 Results

4.4.1 Mitochondrial DNA Analysis

4.4.1(a) Nucleotide composition and genetic diversity

A total of 323 individuals from 25 populations were successfully PCR amplified for the cyt b gene region. Samples successfully amplified per population varied as shown in Table 4.2. The 883 base pair (bp) cyt b region after editing consisted of 134 bp of ND6,

73 bp of t-RNA-Glu and 676 bp of cyt b gene (ND 6-tRNAglu-cyt b region). The average nucleotide composition was A = 29%; T = 32%; C = 26% and G = 13% with 209 variable sites, of which 205 were parsimony informative sites. The protein coding regions were successfully translated indicating that there were no nuclear mitochondrial pseudogenes

(numts). The saturation plots (Figure 4.3) illustrate absence of saturation of nucleotide substitution.

Nucleotide diversity (π) of cyt b across the 25 populations was generally low ranging from 0% to 0.34% with 13 populations showing absence of diversity (π = 0%).

The JMP population showed the highest nucleotide diversity (Table 4.2). Haplotype diversity (h) across populations ranged from 0 to 0.7810. The highest haplotype diversity was in SKR population while 13 populations had diversity equaled to zero. Several adjacent populations (mostly at the extended river mouth or tip of each Paleo-drainage) including KS, KJ, ST, SKR and JMP presented moderately high haplotype diversity. More than half (13) of the populations showed absence of both haplotype and nucleotide

93 variation. However, interestingly in total sample dataset, the cyt b of H. pogonognathus showed high nucleotide and haplotype diversity at 4.69% and 0.9627 respectively.

Table 4.2: Sample location in abbreviation, no. of sequences (n), no. of haplotype (hp), no. of polymorphic sites (#V), nucleotide diversity (π), haplotype diversity (h) and expected heterozygosity per site based on number of segregating sites (ϴs) of cyt b for H. pogonognathus populations.

Locus: cyt b Location n hp #V π h ϴs KWK 15 1 0 0.0000 0.0000 0.0000

TE 6 1 0 0.0000 0.0000 0.0000 TB 15 1 0 0.0000 0.0000 0.0000 BP 15 2 1 0.0002 0.1333 0.3075 PTG 14 2 1 0.0002 0.1429 0.3145 SK 13 1 0 0.0000 0.0000 0.0000

PJG 15 2 1 0.0002 0.1333 0.3075 DM 15 1 0 0.0000 0.0000 0.0000 SU 7 1 0 0.0000 0.0000 0.0000 SOM 15 1 0 0.0000 0.0000 0.0000 KS 13 3 3 0.0010 0.5128 0.9667 PI 15 1 0 0.0000 0.0000 0.0000 JP 15 3 2 0.0007 0.5619 0.6151

LB 14 1 0 0.0000 0.0000 0.0000 S 15 1 0 0.0000 0.0000 0.0000 RA 13 2 1 0.0003 0.2821 0.3222 PA 15 2 1 0.0004 0.3429 0.3075 PT 9 1 0 0.0000 0.0000 0.0000

KJ 7 2 3 0.0016 0.4762 1.2245 ST 13 2 4 0.0023 0.5128 1.2890

SKR 15 5 4 0.0012 0.7810 1.2302 JBI 15 1 0 0.0000 0.0000 0.0000 JMP 4 2 6 0.0034 0.5000 3.2727 RBT 15 1 0 0.0000 0.0000 0.0000 MRW 15 2 1 0.0002 0.1333 0.3075

Total (sample) 323 @39 @209 0.0469 0.9627 32.896 @- actual total no. of haplotype and no. of polymorphic sites as shown in Table 4.3 and 4.4.

Figure 4.3: Nucleotide substitution saturation analysis for cyt b of H. pogonognathus. Transitions (s) and transversions (v) plotted against p distance.

4.4.1(b) Haplotype distribution

Table 4.3 and Table 4.4 (haplotype frequencies in Appendix F) show the generated haplotypes of cyt b and its distribution across 25 H. pogonognathus populations. A total of 39 haplotypes with 209 variable sites within an 883 bp segment were revealed from

323 cyt b sequences. The haplotypes of Hpb17 to 20 from JP and LB, and Hpb34 to 39 from southern Sumatra have high variable sites compared to other haplotypes. As shown in Table 4.4, most of the haplotypes were unique to their own population with no haplotype sharing among populations except for haplotype Hpb02 (shared between TE and TB), Hpb07 (shared between SK and ST) and Hpb29 (shared between ST and SKR) which are from same drainage system (Malacca drainage). Fourteen populations were monomorphic for a single haplotype (private haplotype). The SK population had the highest number of haplotypes with five haplotypes.

Table 4.3: The 39 haplotypes with 209 variable sites within a 883 bp segment of cyt b for H. pogonognathus populations.

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 1 2 2 3 3 3 4 4 5 6 6 7 7 8 8 8 9 9 9 9 9 0 0 0 0 1 1 1 1 2 2 2 2 3 3 3 4 4 6 6 8 8 8 8 9 0 0 1 1 3 4 4 5 5 5 6 6 1 3 9 1 4 7 3 5 9 2 7 2 3 9 2 8 1 7 9 0 3 5 6 9 2 5 7 8 0 6 7 8 0 2 5 6 0 5 9 0 8 2 5 0 1 7 8 6 3 6 6 7 5 0 7 2 3 5 1 7 Hpb01 C C A T A T T C C A A C C G T C T T A G T A T C A C A T A T T C C C C A G G C C A C C G C T T C T T C T C C A T A C A T Hpb02 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb03 ...... G T ...... C . . . . G ...... T . G ...... T G C . . . . Hpb04 ...... G T ...... C . . . . G ...... T . G ...... T G C . . . . Hpb05 . T . C ...... G T ...... T ...... T . G . T ...... T G . . . . . Hpb06 . T . C ...... G T ...... T ...... T . G . T ...... T G . . . . . Hpb07 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb08 . . . C ...... G T ...... G ...... T . G ...... T G . . . . . Hpb09 . . . C ...... G T ...... G ...... T . G ...... T G . . . . . Hpb10 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb11 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb12 ...... G T A ...... G ...... A . T . G ...... T ...... Hpb13 ...... G T ...... G ...... T ...... T G . . . . . Hpb14 ...... G T ...... A . . . . G ...... T ...... T G . . . . . Hpb15 ...... G T ...... G ...... T ...... T G . . . . . Hpb16 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb17 T . C ...... G T . . T . C . A C . C . . T . C C . . . . T T . . A T T . T T . T . A . C . . C . T G C . . . C Hpb18 T . C ...... G T . . T . C . A C . C . . T . C C . . . . T T . . A T T . T T . T . A . C . . C . T G C . . . C Hpb19 T . C ...... G T . . T . C . A C . C . . A . C C . . . . T T . . A T T . T T . T . A . C . . C . T G C . . . C Hpb20 T . C . . . . . T . . G T . . T . C . A C . C . . T . C C . . . . T T G . A T T . T T . T C A T C . T C . T G C . . . C Hpb21 ...... G T ...... T . G ...... T G . . . . . Hpb22 ...... G T ...... G ...... T . G ...... T G . . . G . Hpb23 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb24 ...... G T ...... T . G ...... T G . . . . . Hpb25 ...... G T ...... T . G ...... T G . . . . . Hpb26 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb27 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb28 ...... G T ...... A . . . . G ...... T . G ...... T G . . . . . Hpb29 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb30 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb31 ...... G T ...... G ...... T . G ...... T G . . . . . Hpb32 ...... G T ...... G ...... T ...... T G . . . . . Hpb33 ...... G T ...... G ...... T . G ...... T G C . . . . Hpb34 . . . C . . C . T T G G T A . . C . . A C G . T . . . . . C . A . . . . . A T . G T T A . . . . C C . C T T G C . T . C Hpb35 . . . C . . C . T . . G T A . . C . T A C . . T . . . . . C . A . . . . . A T . . T T . . . . . C . . C . T G C . T . . Hpb36 . . . C . . C . T . . G T A . . C . . A C . . T . . . . . C . A . . . . . A T . . T T . . . . . C . . C . T G C . T . . Hpb37 . . . C T C . A . . . G T A . T . . . A ...... G ...... A T . . T T A . . . . C . . C . T G C G . . . Hpb38 A . G C . . . . T C . G T . C A C . . A . G . T . . G . . C C A T . . . . A T . . T T . . . . . C . . C . T G C . T . . Hpb39 A . G C . . . . T C . G T . . A C . . A . G . T . . G . . C C A T . . . . A T . . T T . . . . . C . . C . T G C . T . .

Table 4.3: continue…

2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 6 7 7 8 8 9 9 0 0 1 2 2 2 3 3 4 5 5 5 5 6 6 7 8 9 0 0 0 0 1 1 2 2 2 3 3 3 4 4 5 5 6 7 7 7 8 8 8 8 9 9 9 0 0 1 1 1 1 1 2 8 3 9 5 8 5 7 3 9 3 1 7 8 0 9 3 1 2 4 7 3 9 2 4 0 2 3 5 6 1 7 3 6 9 0 2 5 1 4 0 6 2 1 4 7 3 6 7 9 2 5 8 1 3 0 1 3 6 9 2 Hpb01 C T C T C C A C C T C C T A G C C C A C T C C T A C G T G C T T T C T T T T T C T C A C T C T A T C C C T C C T A T C A Hpb02 ...... A ...... Hpb03 ...... A ...... Hpb04 ...... A ...... Hpb05 ...... A ...... C . . . . Hpb06 ...... A ...... C . . . . Hpb07 ...... A ...... A . . . . . Hpb08 ...... T . . . . . A ...... Hpb09 ...... T . . . . . A ...... Hpb10 ...... A ...... A . . . . . Hpb11 . . . C ...... A ...... Hpb12 . . . C ...... A ...... Hpb13 ...... A ...... T ...... Hpb14 ...... A ...... T ...... Hpb15 ...... A ...... T ...... Hpb16 . C ...... A . . . . . C . . . G ...... C ...... Hpb17 . C A C T . G . . . . . C T A T . . . . A ...... T . C C . C C C A . T C T C T C T . . . T T T . G . C . C . C Hpb18 . C A C T . G . . . . . C T A T . . . . A ...... T . C C . C C C A . T C T C T C T . . . T T T . G . C . C . C Hpb19 . C A C T . G . . . . . C T A T . . . . A ...... T . C C . C C C A . T C T C T C T . . . T T T . G . C . C . C Hpb20 . C A C . . G . . . . . C T A T . . . . A ...... T . C C . C C C G . T C T C T C T . . . T T T . G . . . C . C Hpb21 ...... A ...... C ...... Hpb22 ...... A ...... C ...... Hpb23 ...... A ...... C ...... Hpb24 ...... A ...... C . C ...... Hpb25 ...... A ...... C ...... Hpb26 . . . C ...... A ...... Hpb27 . . . C ...... A ...... Hpb28 . . . C ...... A ...... T ...... Hpb29 ...... A ...... Hpb30 ...... A ...... T ...... Hpb31 ...... A ...... A ...... Hpb32 ...... A ...... Hpb33 ...... A ...... Hpb34 . C . . T T . T . . T . . . A T . . . T . T T C . T . . . . C . C . . C . C ...... T . G . . T . C G . C . C . . Hpb35 . C . . . T . T . C . . . . A T . . G T . . T C . T . . A . C ...... C ...... C . T . C G . C . C . . Hpb36 . C . . . T . T ...... A T . . G T . . T C . T . . A . C ...... C ...... C . T . C G . C . C . . Hpb37 T ...... A T . . . . . T . C . T . . A . C ...... C . T ...... T . . G G . . . . . Hpb38 T C . C . T G . . . . T . . A T T A G T C T . C . T A C . . C C . . . . C C C ...... T C . . . T . C G . C G . T . Hpb39 T C . C . T G . . . . T . . A T T A G T C T . C . T A C . . C C . . . . C C C ...... T C . . . T . C G . C G . T .

Table 4.3: continue…

7 7 7 7 7 7 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 9 9 9 9 9 0 0 1 1 1 2 3 3 4 4 4 4 4 5 5 5 5 6 6 6 7 7 8 3 0 2 6 8 9 1 4 1 6 9 5 4 7 0 3 6 7 9 0 1 2 5 1 2 4 3 6 2 Hpb01 A A C T A A C T G A T C A T T T A G T A a T T C T C G C C Hpb02 ...... T ...... C . . . . . A . . Hpb03 ...... T ...... C . . . . . A . . Hpb04 ...... T ...... C . . . . . A . . Hpb05 ...... T ...... C . . . . . A T . Hpb06 ...... T ...... C . . . . . A T . Hpb07 ...... T ...... C . . . . . A . . Hpb08 ...... T ...... C . . . . . A T . Hpb09 ...... T ...... C . . . . . A T . Hpb10 ...... T ...... g . . C . . . . . A . . Hpb11 ...... T ...... C . . . . . A . . Hpb12 ...... T ...... C . . . . c A . . Hpb13 ...... t . . . A . . C ...... Hpb14 ...... A . . C . . . . c . . . Hpb15 ...... T . . . . A . . C . . . . c . . . Hpb16 ...... T ...... C . . . . . A . . Hpb17 T G A C T G T C . . C . . . . C G . . . C C . T . A A . A Hpb18 T G A C T G T C . . C . . . . C G . . . C C . T . A C . A Hpb19 T G A C T G T C . . C . . . . C G . . . C C . T . A A . A Hpb20 T G A C T G T C . . C . . . t C G . . G C C . T . A A . a Hpb21 ...... T ...... T . . . . . A . T Hpb22 ...... T ...... C . . . . . A . . Hpb23 ...... T ...... C . t . . . A . . Hpb24 ...... T . g ...... C t . . A . . . . Hpb25 ...... T ...... C . . . A . . . . Hpb26 ...... T ...... C . . . . . A . . Hpb27 ...... T ...... C . . . . . A . . Hpb28 ...... T ...... C . . . . . A . . Hpb29 ...... T ...... C . . . . . A . . Hpb30 ...... T ...... C . . . . . A . . Hpb31 ...... T ...... C . . . . . A . . Hpb32 ...... T ...... C . . . . . A . . Hpb33 ...... T ...... C . . . . . A . . Hpb34 C . T C T ...... T . A . C . A . . C . . . . . A . T Hpb35 T . . C T . . . A . . . . A . C . a A C C . C . . . A T T Hpb36 T . . C T ...... A . C . . . . C . C . . . A T T Hpb37 C . T C T . . . . G . . . C C C . . . T C . C . . . A T . Hpb38 C . T C C . . . . . C . . A . . . A C . C . C . . . A T . Hpb39 C . T C C . . . . . C . . A . . . A C . C . C . . . A T .

Table 4.4: Haplotype distribution of cyt b across 25 H. pogonognathus populations in Sundaland.

Central-east Southeast and Region Northwest, West PM Central PM Northeast, East, Southeast PM SM Bangka SM Haplotype KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW Total (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Ma) (Ma) (nS) (nS) (nS) (nS) n 15 6 15 15 14 13 15 15 7 15 13 15 15 14 15 13 15 9 7 13 15 15 4 15 15 323 Hpb01 15 15 Hpb02 6 15 21 Hpb03 14 14 Hpb04 1 1 Hpb05 13 13 Hpb06 1 1 Hpb07 13 8 21 Hpb08 14 14 Hpb09 1 1 Hpb10 15 15 Hpb11 7 7 Hpb12 15 15 Hpb13 9 9 Hpb14 2 2 Hpb15 2 2 Hpb16 15 15 Hpb17 9 9 Hpb18 1 1 Hpb19 5 5 Hpb20 14 14

100

Table 4.4: continue…

Central-east Southeast and Region Northwest, West PM Central PM Northeast, East, Southeast PM SM Bangka SM Haplotype KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW Total (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Ma) (Ma) (nS) (nS) (nS) (nS) n 15 6 15 15 14 13 15 15 7 15 13 15 15 14 15 13 15 9 7 13 15 15 4 15 15 323 Hpb21 15 15 Hpb22 11 11 Hpb23 2 2 Hpb24 3 3 Hpb25 12 12 Hpb26 9 9 Hpb27 5 5 Hpb28 2 2 Hpb29 5 6 11 Hpb3 1 1 Hpb31 4 4 Hpb32 2 2 Hpb33 2 2 Hpb34 15 15 Hpb35 1 1 Hpb36 3 3 Hpb37 15 15 Hpb38 14 14 Hpb39 1 1 n = number of individuals. Abbreviation in parenthesis after location is hypothetical Paleo-drainage where it is situated. PM = Peninsular Malaysia, SM = Sumatra.

101

4.4.1(c) Phylogeography and evolutionary relationships among haplotypes

The construction of NJ cyt b tree (Figure 4.4) employed Tamura-Nei with Gamma distribution (TN93+G) model (the third best model) as there is no option for HKY model

(best model obtained from the optimal substitution models under the BIC) to construct NJ tree in MEGA v6.0. The BI cyt b tree (Figure 4.5) was conducted with best model across partitions (Table 4.5) following the results of the PartitionFinder v1.1.1. Both NJ and BI cyt b trees showed nearly the same topology which divided the haplotypes into three major clades excluding the outgroup (H. kuekenthali) with high bootstrap value (>70) and high

Bayesian posterior probabilities (pp). A closer examination revealed that all 39 haplotypes were assigned into the three clades; 1. southern Sumatra clade 2. Kelantan clade and 3. a clade that encompasses all the other populations and therefore referred to as the Main clade. These clades revealed a very interesting geographical pattern which conformed with the classification deduced from the DNA barcoding study. The Main clade was represented by the majority of haplotypes (Hpb01 – Hpb16 and Hpb21 - Hpb33) which were found across two Paleo-drainages (Malacca and Siam). All six haplotypes from north Sunda drainage were grouped to the southern Sumatra clade (Hpb34 – Hpb39) which is the sister clade to the Main clade. Four haplotypes (Hpb17 – Hpb20) classified as belonging to the Kelantan clade were found from two sites (Jeram Pasu and Lata

Belatan) from Kelantan within the Siam drainage. Within all clades, all individuals were well assigned into their own respective population. Therefore, this analysis clearly shows that the Paleo-drainage system was not the major factor for population structuring.

102

Figure 4.4: Neighbor-Joining cyt b tree among 39 H. pogonognathus haplotypes generated through TN93+G model. Values at nodes represent bootstrap confidence level (1000 replicates). A single H. kuekenthali (KJI OG) sequence was used as outgroup. Three major clades indicated by black vertical bars: Main, southern Sumatra and Kelantan. The scale bar represents 2% of substitution divergence.

103

Figure 4.5: Bayesian Inference cyt b tree among 39 H. pogonognathus haplotypes. Values at nodes represent the Bayesian posterior probabilities. A single H. kuekenthali (KJI OG) sequence were used as outgroup. Three major clades indicated by black vertical bars: Main, southern Sumatra and Kelantan. The scale bar refers to genetic distance.

104

Table 4.5: Best model across partitions after PartitionFinder analysis for cyt b of H. pogonognathus.

Partition Best Model Subset Partitions Subset Sites 1 HKY+G nd_6, cytb_3 1-134, 210-883\3 2 K80+G cytb_1, tglu 135-207, 208-883\3 3 F81 cytb_2 209-883\3

The evolutionary relationships among these 39 haplotypes are also illustrated in the Minimum Spanning Network (MSN - Figure 4.6). The MSN is colour-coded according to the Paleo-drainage systems to describe the haplotypes relationships found within and across three Paleo-drainages in Sundaland. Three clades are observed; Firstly is the southern Sumatra clade consisting of haplotypes (Hpb34 to Hpb39) from north Sunda drainage populations (JBI, JMP, RBT, MRW). Secondly the haplotypes (Hpb17 to Hpb20) derived from JP and LB populations represent a distinct component (Kelantan clade) of the central Siam drainage. Although these haplotypes from Kelantan clade belong to the

Siam drainage; however, they are however not closely related to other haplotypes within the same drainage. Thirdly, is the Main clade that consists of the rest of the haplotypes from Malacca and Siam drainages. The MSN clearly shows that the Kelantan clade is highly divergent from the widespread Main clade with high mutation sites of 120 and the southern Sumatra clade is separated by 55 mutation sites from the Main clade.

105

Figure 4.6: Minimum Spanning Network of 39 cyt b haplotypes of H. pogonognathus. Crossbars on connecting line indicate the number of substitutions separating haplotypes and the number indicate additional mutation steps. The size of the circles is proportional to haplotype frequency. Small black dots are missing haplotypes linking the clades. (b) Map of Sundaland with Paleo-drainages systems indicated by colours.

106

4.4.1(d) Population genetic divergence and population structure

The cyt b pairwise genetic distances (Table 4.6) within population generated using

Tamura-Nei +G model ranged from 0% to 0.3% with the JMP population having the highest value while the majority of populations possessed no variation, 0%. Intraspecific comparisons ranged from 0% (ST vs TB) to 17.9% (LB vs MRW). As expected, JP and

LB populations (Kelantan) have the highest distance when measured to other populations ranging from 13.5% to 17.9%. Comparisons between the populations from southern

Sumatra (JBI, JMP, RBT and MRW) and other populations have divergence value ranging from 8.1% (RBT vs RA) to 12.9 %. (JBI vs KWK). Apart from these two groups, the divergences among other populations (Peninsular Malaysia except Kelantan and central

Sumatra) were relatively low ranging from 0% to 2.6%. These results are congruent with the findings in the generated cyt b trees and the MSN.

107

Table 4.6: Pairwise genetic distances with TN93+G model of cyt b between H. pogonognathus populations.

KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW KWK 0.000 TE 0.021 0.000 TB 0.021 0.000 0.000 BP 0.024 0.002 0.002 0.000 PTG 0.026 0.008 0.008 0.011 0.000 SK 0.022 0.003 0.003 0.006 0.009 0.000 PJG 0.024 0.005 0.005 0.007 0.006 0.006 0.000 DM 0.023 0.005 0.005 0.007 0.010 0.001 0.007 0.000 SU 0.026 0.008 0.008 0.010 0.014 0.009 0.010 0.010 0.000 SOM 0.025 0.009 0.009 0.012 0.015 0.010 0.012 0.012 0.008 0.000 KS 0.018 0.011 0.011 0.013 0.017 0.012 0.013 0.013 0.013 0.017 0.001 PI 0.026 0.007 0.007 0.009 0.013 0.008 0.009 0.009 0.013 0.014 0.016 0.000 JP 0.157 0.144 0.144 0.142 0.144 0.142 0.148 0.144 0.141 0.146 0.146 0.141 0.001 LB 0.165 0.152 0.152 0.150 0.155 0.150 0.155 0.151 0.148 0.154 0.154 0.148 0.013 0.000 S 0.022 0.007 0.007 0.009 0.012 0.008 0.009 0.009 0.013 0.014 0.016 0.009 0.142 0.150 0.000 RA 0.023 0.004 0.004 0.007 0.010 0.006 0.007 0.007 0.010 0.011 0.013 0.007 0.140 0.151 0.007 0.000 PA 0.020 0.006 0.006 0.008 0.011 0.007 0.008 0.008 0.012 0.013 0.012 0.008 0.142 0.150 0.006 0.006 0.000 PT 0.024 0.006 0.006 0.008 0.012 0.007 0.008 0.008 0.005 0.006 0.013 0.010 0.141 0.148 0.010 0.008 0.009 0.000 KJ 0.025 0.007 0.007 0.009 0.013 0.008 0.010 0.010 0.004 0.007 0.014 0.012 0.135 0.142 0.012 0.009 0.011 0.004 0.002 ST 0.022 0.003 0.003 0.006 0.009 0.002 0.006 0.003 0.008 0.009 0.012 0.008 0.142 0.150 0.008 0.006 0.007 0.006 0.007 0.002 SKR 0.021 0.004 0.004 0.006 0.010 0.005 0.007 0.006 0.008 0.009 0.011 0.009 0.143 0.150 0.009 0.006 0.008 0.005 0.007 0.003 0.001 JBI 0.129 0.115 0.115 0.113 0.112 0.116 0.115 0.118 0.116 0.120 0.117 0.117 0.157 0.169 0.115 0.118 0.117 0.115 0.118 0.115 0.112 0.000 JMP 0.110 0.101 0.101 0.099 0.094 0.102 0.098 0.104 0.102 0.106 0.102 0.102 0.156 0.159 0.099 0.104 0.103 0.104 0.103 0.100 0.097 0.053 0.003 RBT 0.089 0.081 0.081 0.080 0.078 0.082 0.078 0.083 0.084 0.086 0.082 0.087 0.146 0.155 0.083 0.081 0.083 0.081 0.083 0.082 0.081 0.096 0.078 0.000 MRW 0.128 0.118 0.118 0.116 0.111 0.119 0.114 0.121 0.118 0.123 0.116 0.118 0.175 0.179 0.121 0.121 0.119 0.114 0.119 0.119 0.118 0.091 0.079 0.095 0.000 OG 0.188 0.174 0.174 0.171 0.177 0.175 0.178 0.177 0.169 0.168 0.174 0.172 0.206 0.208 0.175 0.175 0.175 0.170 0.171 0.173 0.169 0.183 0.180 0.200 0.182

108

Based on AMOVA, all ФST comparisons were highly significant (p < 0.05) after

Bonferroni corrections, except for two (TE vs TB; SK vs ST) (Table 4.7). Overall the ФST comparisons plots (Figure 4.7) of cyt b indicate that the H. pogonognathus populations in

Sundaland river basins are highly structured.

Figure 4.7: Pairwise ФST comparisons of cyt b (below diagonal) between H. pogonognathus populations.

109

Table 4.7: Pairwise ФST values of cyt b (below diagonal) between H. pogonognathus populations. Geographical distances (above diagonal in italics) between populations based on the shortest path following the Paleo-drainages systems (km).

KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW KWK 0.000 317 320 719 742 995 992 1056 2225 2324 2203 2258 2461 2504 2304 2307 2137 1505 2120 1351 1474 1794 1981 1812 2339 TE 1.000 0.000 233 750 804 1025 1019 1115 2324 2359 2168 2246 2524 2536 2388 2324 2198 1550 2100 1362 1488 1851 2023 1851 2397 TB 1.000 0.000 0.000 772 790 1101 1077 1176 2346 2375 2197 2277 2546 2590 2411 2389 2249 1566 2145 1357 1496 1903 2069 1866 2322 BP 0.997 0.953 0.968 0.000 27 1062 1014 1121 2295 2314 2174 2285 2549 2600 2375 2319 2178 1520 2164 1356 1499 1851 1958 1803 2303 PTG 0.997 0.985 0.990 0.985 0.000 1088 1041 1150 2326 2343 2200 2310 2579 2635 2398 2350 2201 1536 2190 1389 1531 1890 1975 1841 2321 SK 1.000 1.000 1.000 0.986 0.991 0.000 90 441 1617 1716 1478 1626 2091 2145 1715 1657 1652 912 1447 774 894 1207 1340 1190 1686 PJG 0.997 0.976 0.984 0.978 0.973 0.986 0.000 480 1601 1706 1457 1601 2080 2127 1705 1633 1631 901 1413 812 950 1212 1321 1180 1655 DM 1.000 1.000 1.000 0.989 0.992 1.000 0.989 0.000 1500 1514 1382 1444 1669 1610 1580 1481 1324 693 1304 548 662 1050 1130 1001 1477 SU 1.000 1.000 1.000 0.990 0.992 1.000 0.990 1.000 0.000 163 130 92 1365 1400 1383 1143 528 1014 648 1525 1314 2273 2417 2416 2189 SOM 1.000 1.000 1.000 0.993 0.995 1.000 0.993 1.000 1.000 0.000 166 147 1308 1341 1168 1126 530 1050 660 1554 1346 2253 2386 2200 2236 KS 0.974 0.938 0.959 0.959 0.967 0.960 0.960 0.966 0.951 0.973 0.000 126 1076 1111 1004 984 501 819 521 1384 1146 2128 2280 2141 2103 PI 1.000 1.000 1.000 0.992 0.994 1.000 0.992 1.000 1.000 1.000 0.971 0.000 1142 1186 1035 1030 583 885 570 1413 1323 2173 2328 2159 2163 JP 0.997 0.996 0.997 0.996 0.996 0.997 0.996 0.997 0.996 0.997 0.993 0.997 0.000 53 334 414 998 1035 1043 1751 1645 2302 2507 2300 2245 LB 1.000 1.000 1.000 0.999 0.999 1.000 0.999 1.000 1.000 1.000 0.996 1.000 0.972 0.000 380 455 1042 1080 1088 1700 1596 2347 2560 2390 2299 S 1.000 1.000 1.000 0.992 0.993 1.000 0.992 1.000 1.000 1.000 0.971 1.000 0.997 1.000 0.000 334 944 1022 968 1639 1454 2229 2354 2176 2155 RA 0.993 0.949 0.966 0.966 0.976 0.971 0.966 0.978 0.979 0.987 0.950 0.978 0.995 0.999 0.978 0.000 936 923 915 1663 1431 2155 2339 2121 2038 PA 0.990 0.952 0.967 0.967 0.973 0.970 0.967 0.976 0.977 0.985 0.946 0.976 0.995 0.998 0.967 0.938 0.000 819 457 1440 1263 2242 2383 2213 2095 PT 1.000 1.000 1.000 0.988 0.991 1.000 0.988 1.000 1.000 1.000 0.956 1.000 0.996 1.000 1.000 0.975 0.974 0.000 781 716 551 2142 2248 2417 2125 KJ 0.980 0.877 0.929 0.934 0.951 0.933 0.936 0.947 0.783 0.929 0.913 0.957 0.991 0.996 0.957 0.917 0.928 0.809 0.000 1343 1169 2145 2327 2176 2086 ST 0.949 0.565 0.677 0.796 0.868 0.333 0.796 0.617 0.818 0.884 0.856 0.863 0.988 0.991 0.863 0.759 0.816 0.772 0.721 0.000 727 1012 1194 980 1586 SKR 0.971 0.796 0.855 0.891 0.930 0.878 0.896 0.906 0.892 0.931 0.903 0.931 0.992 0.995 0.931 0.873 0.898 0.859 0.801 0.501 0.000 903 1003 825 1421 JBI 1.000 1.000 1.000 0.999 0.999 1.000 0.999 1.000 1.000 1.000 0.995 1.000 0.997 1.000 1.000 0.998 0.998 1.000 0.995 0.989 0.994 0.000 646 485 920 JMP 0.993 0.985 0.993 0.991 0.990 0.992 0.991 0.993 0.987 0.993 0.983 0.993 0.990 0.995 0.993 0.989 0.989 0.989 0.975 0.970 0.981 0.987 0.000 173 1089 RBT 1.000 1.000 1.000 0.999 0.999 1.000 0.999 1.000 1.000 1.000 0.994 1.000 0.997 1.000 1.000 0.998 0.997 1.000 0.993 0.985 0.992 1.000 0.991 0.000 912 MRW 0.999 0.999 0.999 0.998 0.998 0.999 0.998 0.999 0.999 0.999 0.995 0.999 0.997 0.999 0.999 0.998 0.997 0.999 0.994 0.988 0.993 0.999 0.989 0.999 0.000 Bold values indicate non significant ФST values after Bonferroni corrections (p > 0.05). 110

The hierarchical AMOVA examined in all hierarchical levels revealed significant genetic structure (Table 4.8). The high and significant ФCT values (0.816) for lineage analysis shows that the most of the variance was attributed to differences among the cyt b lineages. On the other hand, the moderately high and significant ФCT value (0.378) for

Paleo-drainage analysis that the variance could also be attributed to differences among

Paleo-drainages which thus indicates geographical subdivision with some influences by

Paleo-drainages. The high variation among groups (cyt b lineages) and ФST value at 81.62% and 0.997 respectively suggested that the populations are highly structured between the lineages while the variability within population was relative low at only 0.33%.

Table 4.8: AMOVA result for hierarchical genetic subdivision for percentage of variation and fixation indices (ФST, ФSC and ФCT) of cyt b of H. pogonognathus populations. Bold values indicate significant value (p < 0.05).

Among Among populations Within Group ФST ФSC ФCT groups (%) within groups (%) population (%) Paleo-drainages 0.995 0.991 0.378 37.82 61.63 0.54 Cyt b lineages 0.997 0.982 0.816 81.62 18.05 0.33

The Mantel test result shows a weak signal of isolation by distance, r = 0.2402

(p<0.05) for correlation between pairwise ФST and geographical distance (Figure 4.8) among the H. pogonognathus populations. This combination of highly significant ФST values and weak Mantel’s correlation test implies that genetic differentiation is not necessarily proportional to geographical distance across the Sundaland river basins and

111 isolation by distance is not suitable to describe the overall pattern of genetic differentiation at this spatial scale.

1.2 1.0 0.8 R² = 0.0578

ST ST 0.6 Ф 0.4 0.2 0.0 0 500 1000 1500 2000 2500 3000 Geographical Distance (km)

Figure 4.8: The ФST plotted against the shortest path following the hypothetical Paleo- drainages systems (km) of H. pogonognathus cyt b data. Trendline (black dashed line) shows the general pattern of only slight increase in genetic distance with a higher leap in geographical distance (IBD).

4.4.1(e) Population history and demographic changes

Thirteen populations (KWK, TE, TB, SK, DM, SU, SOM, PI, LB, S, PT, JBI and

RBT) of the total 25 populations could not be tested for deviation from mutation drift as no gene/haplotype variation was observed. Tajima’s D and Fu’s Fs neutrality tests (Table

4.9) for the other populations were non-significant suggesting that these populations do not deviate from mutation-drift and gene flow-drift equilibrium, i.e. no evidence for recent population expansion. Additionally, the mismatch distribution of R2 (Table 4.9) also failed to detect any significant population expansion as all values were non-significant. The

112 same was (no recent population expansion) observed for the total sample dataset where all tests exhibited small positive values except Fu’s Fs tests with large positive values.

However, all the values were non-significant.

Table 4.9: Summary of population neutrality tests and demographic analyses based on Tajima’s D, Fu’s Fs, Rasmos-Onsins & Rozas (R2) and Harpending’s raggedness index (Hri) for H. pogonognathus populations based on cyt b gene.

Locus: cyt b Population Tajima’s D Fu’s Fs R2 Hri KWK 0.000 0.000 0.000 0.000 TE 0.000 0.000 0.000 0.000

TB 0.000 0.000 0.000 0.000

BP -1.159 -0.649 0.249 0.556 PTG -1.155 -0.595 0.258 0.531 SK 0.000 0.000 0.000 0.000 PJG -1.159 -0.649 0.249 0.556 DM 0.000 0.000 0.000 0.000 SU 0.000 0.000 0.000 0.000

SOM 0.000 0.000 0.000 0.000

KS -0.394 0.436 0.141 0.101 PI 0.000 0.000 0.000 0.000 JP -0.024 -0.064 0.172 0.226 LB 0.000 0.000 0.000 0.000 S 0.000 0.000 0.000 0.000

RA -0.274 0.240 0.141 0.270

PA 0.235 0.597 0.171 0.216 PT 0.000 0.000 0.000 0.000 KJ 0.755 2.508 0.238 0.728 ST 2.009 4.590 0.256 0.763 SKR -0.476 -1.406 0.130 0.220 JBI 0.000 0.000 0.000 0.000

JMP -0.809 2.944 0.433 0.750

RBT 0.000 0.000 0.000 0.000 MRW -1.159 -0.694 0.249 0.556 Total 0.243 33.214 0.099 0.006 Bold values indicate significance at p < 0.05.

113

In contrast, the Hri results (Table 4.9) indicate lack of support for a stationary population, however, it does not preclude population expansion. Further inspection with mismatch distribution shows absence of unimodal distribution. Mismatch distribution for most of these populations (Figure 4.9) were observed to have undergone rapid population reduction towards its new equilibrium which is congruent with the large Hri value (usually refer to stationary population) detected here. Generally, the neutrality test and mismatch distribution shows no indication of population expansion in any population. Nevertheless, the mismatch distribution also suggests rapid population reduction (bottleneck) toward its new equilibrium. The EBSPs analysis (Figure 4.10) of cyt b for the total sample dataset of H. pogonognathus suggested that the population retained a constant effective population size, and only declined very recently, which is congruent with the mismatch distribution analysis.

114

BP JMP JP

KJ KS MRW

PA PJG PTG Figure 4. 9: Mismatch distribution of cyt b of twelve H. pogonognathus populations showing the expected and observed pairwise differences between the sequences with the respective frequencies under constant population size. The solid lines represent the expected distribution and the dotted lines represent the observed distribution. The dotted line shows that the left edge of distribution converges rapidly toward the new equilibrium. Multimodal was detected for three populations (JMP, KJ and ST) and total samples dataset.

115

RA SKR ST

Total samples

Figure 4.9: continue…

116

cyt b 20

5 Population sizescalar Population 0 0 0.25 0.5 0.75 1 1.25 1.5 Time (millions of years before present)

Figure 4.10: Extended Bayesian Skyline Plots (EBSPs) showing the demographic history of H. pogonognathus based on cyt b. Solid blue line is the median effective population size, the dashed lines are the upper and lower 95% HPD for those estimates.

117

4.4.2 Nuclear DNA (SCNP: Hp5 and Hp54) analysis

4.4.2(a) Nucleotide composition and genetic diversity

For nuclear DNA, not all individuals were reliably amplified, and some individuals contained missing data and ‘noise’ due to indels and heterozygosity. None of the individuals from KWK was successfully amplified for locus Hp54. Sequences were trimmed in order to minimize missing data. The sequences with PHASE results that did not reach the probability threshold of > 0.7 were discarded, thus the actual nuclear sample sizes were reduced. Most of the number of nuclear samples was doubled following

PHASE results (Table 4.10).

A total of 458 sequences and 375 sequences were obtained after the above criteria for Hp5 and Hp54 respectively. The Hp5 sequences, trimmed to 264 bp with indels at positions 49 to 51 and 57, contained mean nucleotide composition of A = 27%, T =35%,

C = 19% and G = 19% and revealed 43 variable sites, of which 41 sites were parsimony informative. Hp54 had 245 bp with indels in several positions (79, 166-167, 179-180 and

221-223), mean nucleotide composition of A = 37%, T =30%, C = 15% and G = 18%, revealed 35 variable sites, all of which were parsimony informative. The saturation plots for Hp5 (Figure 4.11) and Hp54 (Figure 4.12) illustrate no saturation of substitution.

118

Table 4.10: Sample location in abbreviation, no of sequences (n), no. of haplotype (hp), no. of polymorphic sites (#V), nucleotide diversity (π), haplotype diversity (h) and expected heterozygosity per site based on number of segregating sites (ϴs) of Hp5 and Hp54 for H. pogonognathus populations.

Locus Hp5 Hp54 Location n hp #V π h ϴs n hp #V π h ϴs KWK 26 3 5 0.0039 0.5662 0.7862 / / / / / / TE 6 1 0 0.0000 0.0000 0.0000 8 2 1 0.0018 0.4286 0.3856 TB 10 1 0 0.0000 0.0000 0.0000 7 1 0 0.0000 0.0000 0.0000 BP 20 2 1 0.0038 0.1000 0.2819 28 3 4 0.0026 0.5185 0.5139 PTG 10 1 0 0.0000 0.0000 0.0000 12 4 9 0.0115 0.7121 2.3179 SK 10 1 0 0.0000 0.0000 0.0000 26 2 6 0.0088 0.5169 1.0482 PJG 22 6 5 0.0063 0.8095 1.0973 26 6 5 0.0053 0.8000 0.7862 DM 20 2 1 0.0017 0.4421 0.2819 16 2 4 0.0022 0.5333 0.3014 SU 18 2 3 0.0016 0.2092 0.5815 8 1 0 0.0000 0.0000 0.0000 SOM 10 1 0 0.0000 0.0000 0.0000 10 1 0 0.0000 0.0000 0.0000 KS 24 3 2 0.0015 0.5942 0.2678 20 5 5 0.0050 0.7526 0.8456 PI 26 2 1 0.0000 0.4431 0.0000 18 5 4 0.0054 0.7516 1.1629 JP 6 1 0 0.0000 0.0000 0.0000 10 1 0 0.0000 0.0000 0.0000 LB 7 1 0 0.0000 0.0000 0.0000 9 1 0 0.0000 0.0000 0.0000 S 9 1 0 0.0000 0.0000 0.0000 16 2 4 0.0020 0.2333 0.6027 RA 20 4 3 0.0037 0.7211 0.8456 14 4 9 0.0131 0.6923 1.8867 PA 10 1 0 0.0000 0.0000 0.0000 20 7 10 0.0132 0.8789 2.2550 PT 18 7 10 0.0158 0.8627 2.6166 5 1 0 0.0000 0.0000 0.0000 KJ 22 7 8 0.0062 0.8009 1.9202 18 3 4 0.0021 0.4641 0.5815 ST 32 8 8 0.0082 0.8690 1.4899 12 3 5 0.0038 0.5455 0.9934 SKR 20 5 8 0.0097 0.8000 1.9731 20 6 9 0.0112 0.8421 1.9731 JBI 22 8 11 0.0124 0.8961 2.4689 18 5 13 0.0126 0.7582 2.6166 JMP 36 14 13 0.0141 0.9270 2.4115 20 4 6 0.0085 0.7105 1.6912 RBT 24 3 9 0.0094 0.4529 1.8745 20 2 1 0.0014 0.3368 0.2818 MRW 30 8 7 0.0060 0.8828 1.2621 14 3 4 0.0077 0.7033 1.2578 Total 458 @71 @43 0.0252 0.9568 6.2659 375 @48 @35 0.0251 0.9563 5.3823 / no data is available @- actual total no. of haplotype and no. of polymorphic sites as shown in Table4.11 to 4.14.

119

Figure 4.11: Nucleotide substitution saturation analysis for Hp5. Transitions (s) and transversions (v) plotted against p distance.

Figure 4.12: Nucleotide substitution saturation analysis for Hp54. Transitions (s) and transversions (v) plotted against p distance.

120

Nucleotide diversity for Hp5 (Table 4.10) was generally low across most populations, ranging from 0% to 1.58% with PT recording the highest nucleotide diversity.

Haplotype diversity, ranged from 0 to the highest value of 0.927 recorded in JMP. Several populations exhibited high genetic diversity (PJG, PT, KJ, ST, SKR, JBI, JMP and MRW).

Nine populations (TE, TB, PTG, SK, SOM, JP, LB, S, PA) showed absence of both haplotype and nucleotide variation. For Hp54, genetic variability was present in several populations across all drainage regions but several populations lacked of variability (TB,

SU, SOM, JP, LB, PT) similar to cyt b and Hp5. The nucleotide diversity ranged from 0% to 1.32% while haplotype diversity ranged from 0 to 0.8789 (Table 4.10). Three populations had the highest nucleotide diversity (RA, PA and JBI) while PA shows highest haplotype diversity. Generally, three populations (TB, SOM and LB) showed total absence of genetic variability in both markers which is also observed in cyt b while more than half of the populations exhibited moderate to high haplotype diversity for Hp54. In total sample dataset, both nDNA markers (Hp5 and Hp54) exhibited low nucleotide diversity and high haplotype diversity.

4.4.2(b) Haplotype distribution

A total of 71 haplotypes (Table 4.11 and Table 4.12) were obtained from 458 sequences for Hp5 (haplotype frequencies in Appendix G). Most of the haplotypes were unique to their own population with ten haplotypes shared by two to five populations.

Three haplotypes (Hp506, Hp515 and Hp518) were considered as common haplotype as

121 they were shared by five populations followed by Hp517, Hp532 and Hp533 shared by three populations, and Hp505, Hp511, Hp516 and Hp519 were shared by two populations.

Hp54 revealed 48 haplotypes (Table 4.13 and Table 4.14) from 375 sequences

(haplotype frequencies in Appendix H). Compared to the other two loci, more haplotypes from Hp54 were shared by at least two populations namely Hp5402, Hp5403, Hp5404,

Hp5406, Hp5408, Hp5409, Hp5413, Hp5415, Hp5423, Hp5426, Hp5428, Hp5429 and

Hp5438. Haplotype Hp5402 was the most common haplotype, shared by seven populations (TE, PTG, DM, S, RA, PA and KJ). The remaining haplotypes were unique to specific populations.

122

Table 4.11: The 71 haplotypes with 79 variable sites generated from 264 nucleotide bases of Hp5 in H. pogonognathus populations.

123

Table 4.11: continue

124

Table 4.12: Haplotype distribution of Hp5 across 25 H. pogonognathus populations.

Central-east Southeast Bangka Region Northwest, West PM Central PM Northeast, East, Southeast PM SM SM SM KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI MRW Haplotype (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Ma) (Ma) (nS) (nS) Total n 26 6 10 20 10 10 22 20 18 10 24 26 6 7 9 20 10 18 22 32 20 22 30 458 Hp501 16 16 Hp502 6 6 Hp503 4 4 Hp504 6 6 Hp505 10 7 17 Hp506 19 10 10 3 42 Hp507 1 1 Hp508 2 2 Hp509 8 8 Hp510 2 2 Hp511 5 5 10 Hp512 2 2 Hp513 14 14 Hp514 6 6 Hp515 16 4 8 6 2 36 Hp516 2 2 4 Hp517 10 14 18 42 Hp518 6 9 8 10 9 6 48 Hp519 6 7 13 Hp520 6 6 Hp521 5 5 Hp522 1 1 Hp523 5 5 Hp524 1 1 Hp525 4 4 Hp526 2 2

125

Table 4.12: continue…

Southeast Bangka Region Northwest, West PM Central PM Northeast, East, Southeast PM Central-east SM SM SM KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW Haplotype (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Ma) (Ma) (nS) (nS) (nS) (nS) Total n 26 6 10 20 10 10 22 20 18 10 24 26 6 7 9 20 10 18 22 32 20 22 36 24 30 458 Hp528 3 3 Hp529 1 1 Hp530 2 2 Hp531 2 2 Hp532 2 2 6 10 Hp533 4 6 6 16 Hp534 1 1 Hp535 2 2 Hp536 2 2 Hp537 2 2 Hp538 2 2 Hp539 2 2 Hp540 6 6 Hp541 4 4 Hp542 2 2 Hp543 4 4 Hp544 2 2 Hp545 2 2 Hp546 5 5 Hp547 2 2 Hp548 3 3 Hp549 8 8 Hp550 2 2 Hp551 2 2 Hp552 2 2 Hp553 2 2

126

Table 4.12: continue…

Southeast Bangka Region Northwest, West PM Central PM Northeast, East, Southeast PM Central-east SM SM SM KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW Haplotype (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Ma) (Ma) (nS) (nS) (nS) (nS) Total n 26 6 10 20 10 10 22 20 18 10 24 26 6 7 9 20 10 18 22 32 20 22 36 24 30 458 Hp554 2 2 Hp555 4 4 Hp556 2 2 Hp557 2 2 Hp558 2 2 Hp559 2 2 Hp560 2 2 Hp561 2 2 Hp562 2 2 Hp563 17 17 Hp564 6 6 Hp565 1 1 Hp566 4 4 Hp567 4 4 Hp568 2 2 Hp569 4 4 Hp570 2 2 Hp571 2 2 n = number of individuals. Abbreviation in parenthesis after location is hypothetical Paleo-drainage where it is situated. PM = Peninsular Malaysia, SM = Sumatra.

127

Table 4.13: The 48 haplotypes with 41 variable sites generated from 264 nucleotide bases of Hp54 for H.pogonognathus.

128

Table 4.13: continue…

129

Table 4.14: Haplotype distribution of Hp54 in 25 H. pogonognathus populations.

Southeast Bangka Region Northwest, West PM Central PM Northeast, East, Southeast PM Central-east SM SM SM TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW Haplotype (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Ma) (Ma) (nS) (nS) (nS) (nS) Total n 8 7 28 12 26 26 16 8 10 20 18 10 9 16 14 20 5 18 12 20 18 20 20 14 375 Hp5401 6 24 Hp5402 2 2 8 2 4 3 3 31 Hp5403 7 18 4 2 34 Hp5404 8 4 14 4 4 2 Hp5405 2 5 Hp5406 3 2 1 Hp5407 1 14 Hp5408 6 8 16 Hp5409 12 4 14 Hp5410 14 10 Hp5411 10 4 Hp5412 4 3 Hp5413 2 1 2 Hp5414 2 35 Hp5415 8 10 9 8 3 Hp5416 9 4 Hp5417 4 2 Hp5418 2 2 Hp5419 2 2 Hp5420 3 Hp5421 2 Hp5422 1 Hp5423 10 9 Hp5424 7 Hp5425 2

130

Table 4.14: continue…

Southeast Bangka Region Northwest, West PM Central PM Northeast, East, Southeast PM Central-east SM SM SM TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW Haplotype (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Ma) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Si) (Ma) (Ma) (nS) (nS) (nS) (nS) Total n 8 7 28 12 26 26 16 8 10 20 18 10 9 16 14 20 5 18 12 20 18 20 20 14 375 Hp5426 5 5 13 23 Hp5427 2 2 Hp5428 2 2 4 Hp5429 2 6 8 Hp5430 8 8 Hp5431 2 2 Hp5432 4 4 Hp5433 2 2 Hp5434 2 2 Hp5435 2 2 Hp5436 4 4 Hp5437 8 8 Hp5438 2 4 6 Hp5439 2 2 Hp5440 2 2 Hp5441 9 9 Hp5442 3 3 Hp5443 6 6 Hp5444 2 2 Hp5445 16 16 Hp5446 4 4 Hp5447 6 6 Hp5448 4 4 n = number of individuals. Abbreviation in parenthesis after location is hypothetical Paleo-drainage where it is situated. PM = Peninsular Malaysia, SM = Sumatra.

131

4.4.2(c) Phylogeography and evolutionary relationships among haplotypes

The NJ (Figure 4.13) and BI (Figure 4.14) trees for Hp5 locus were constructed using Kimura 2 parameter with Gamma distribution (K2P+G) model. Both constructed phylogenetic trees showed nearly same topology. Two clades were presented; the first

(Main) with all haplotypes except Hp519 (with high bootstrap value (>70) and high

Bayesian pp) and the second clade, consisting of only haplotype Hp519 which represent populations JP and LB (Kelantan). Internal nodes were generally not well resolved. For

Hp54 locus, both NJ (Figure 4.15) and BI (Figure 4.16) trees were constructed using

Tamura 3 parameter with Gamma distribution (TN92+G) model. Both trees for the Hp54 genes were not well resolved with polytomous relationships among haplotypes and also low bootstrap values clustering some internal nodes. Nevertheless, the haplotype from

Kelantan (Hp5423) was represented by a deep branch in both trees. Furthermore, haplotypes from southern Sumatra clustered together with high Bayesian pp in BI tree but low support in NJ tree though all haplotypes clustered into only one poorly resolved clade.

132

Figure 4.13: Neighbour-Joining HP5 tree among 71 haplotypes of H. pogonognathus generated through K2P+G model. Values at nodes represent bootstrap confidence level (1000 replicates). A H. kuekenthali (KJI501 OG) sequence was used as outgroup. Two major clades are indicated by black vertical bars: Main and Kelantan. The scale bar refers to genetic distance.

133

Figure 4.14: Bayesian Inference Hp5 tree generated through K2P+G model. Value at nodes represents the Bayesian posterior probabilities. A H. kuekenthali (KJI501 OG) sequence was used as outgroup. Two major clades are indicated by black vertical bars: Main and Kelantan. The scale bar refers to genetic distance.

134

Figure 4.15: Neighbour-Joining Hp54 tree among 48 haplotypes of H. pogonognathus generated through TN92+G distribution model. Values at nodes represent bootstrap confidence level (1000 replicates). A H. kuekenthali (KJI OG) sequence was used as an outgroup. The scale bar refers to genetic distance.

135

Figure 4.16: Bayesian Inference Hp54 tree generated through TN92+G model. Values at nodes represents the Bayesian posterior probabilities. A H. kuekenthali (KJI OG) sequence was used as outgroup. The scale bar refers to genetic distance.

136

The constructed MSN (Figure 4.17) of Hp5 haplotypes presented two clades where haplotype Hp519 that represented JP and LB populations separated from the other haplotypes by 43 mutation sites forming its own clade (Kelantan clade). The other clade consisted of the rest of the haplotypes from various populations. The haplotypes Hp506,

Hp515, Hp517 and Hp518 that were internally positioned were just a few mutational sites differentiated from most of the other haplotypes. Based on these characteristics, as in accordance to predictions from coalescence theory (Posada & Crandall, 2001; So, et al.,

2006a), these haplotypes were most likely the ancestral variants which is considered as the common central haplotype especially when their distributions are widespread. Under coalescent theory, tip haplotypes are generally considered to be recently evolved (Castello

& Templeton, 1994; Crandall, 1996), where these new genetic variants have not sufficient time to disperse widely across the distribution at large. Most of the haplotypes of southern

Sumatra populations from the north Sunda drainage were positioned as tip haplotypes that were linked to internal haplotypes (Hp515, Hp517 and Hp518). It appeared that many southern Sumatra haplotypes (haplotypes in red colour) represented descendants from the

Siam drainage populations of JBI (Hp515) and the southern Sumatra MRW (Hp518) haplotype. In general, the Siam drainage seemed to be the ancestral for other haplotypes.

137

Figure 4.17: Minimum Spanning Network of 71 Hp5 haplotypes obtained from 458 individuals. Crossbars on connecting line indicate the number of substitutions separating haplotypes and the numbers indicate additional mutation steps. The size of the circles is proportional to haplotype frequency. Small black dots are missing haplotypes linking the clades. (b) Map of Sundaland with Paleo- drainages systems indicated by colours.

138

The MSN (Figure 4.18) for haplotypes of Hp54 revealed three clades. One clade consisted of haplotypes representing southern Sumatra populations from north Sunda drainage (JBI, JMP, RBT). Another clade consisted of haplotype Hp5423 representing JP and LB populations (Kelantan clade) from central Siam drainage, and a clade consisting of the remaining haplotypes from other populations from Malacca and Siam drainages.

Interestingly, haplotypes Hp5446 to Hp5448 from MRW were relatively highly diverged from other populations which were also from southern Sumatra. Although three clades were presented in MSN, these were however, separated by only 2 to 7 mutation sites where

Hp5423 (Kelantan) had the highest mutation sites from the central common haplotype.

The haplotypes Hp5402, Hp5403, Hp5404 and Hp5415 were represented as the common central haplotypes as they were shared by several populations and with just a few mutational sites connected to the other haplotypes and therefore considered to be the ancestral variants.

139

Figure 4.18: Minimum Spanning Network of 48 Hp54 haplotypes obtained from 375 individuals. Crossbars on connecting line indicate the number of substitutions separating haplotypes and the numbers indicate additional mutation steps. The size of the circles is proportional to haplotype frequency. Small black dots are missing haplotypes linking the clades. (b) Map of Sundaland with Paleo- drainages systems indicated by colours.

140

4.4.2(d) Population structure

For the Hp5, pairwise genetic distance (Table 4.15) generated using K2P+G model revealed generally lower values compared to cyt b gene ranging from 0% to

5.4% (TE vs JP and LB). The pairwise distance of JP and LB populations compared to other populations ranged from 3.3% to 5.4% which nearly overlapped with comparisons of the outgroup with other populations (ranged from 3.8% to 6.5%).

Other than these two Kelantan populations, the pairwise distances among the other populations ranged from 0% to 3.5 %. In concordance with this, in both the generated tree and MSN, the JP and LB populations from Kelantan were highly diverged from other populations. This is congruent with the findings for cyt b.

Pairwise genetic distance (Table 4.16) for Hp54 generated using TN92+G model in general revealed even lower values than Hp5 ranging from 0% to 4.6%.

However, for this analysis similar levels of divergence were observed for comparisons involving (the Peninsular Malaysia and central Sumatra populations vs the southern

Sumatra populations - 1.6 to 4.6%.) and (the Kelantan populations vs other populations

- 1.7 to 4.6%). The divergence values for other comparisons ranged from 0% to 1.3%.

Interestingly, the outgroup divergences ranging from 1.3% to 4.8% overlapped with intraspecific divergence.

141

Table 4.15: Pairwise genetic distances with K2P+G model of Hp5 between H. pogonognathus populations.

KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW

KWK 0.002 TE 0.022 0.000 TB 0.013 0.009 0.000 BP 0.018 0.014 0.005 0.000 PTG 0.018 0.014 0.005 0.000 0.000 SK 0.018 0.014 0.005 0.000 0.000 0.000 PJG 0.018 0.011 0.010 0.012 0.012 0.012 0.008 DM 0.012 0.015 0.015 0.020 0.020 0.020 0.012 0.020 SU 0.002 0.022 0.013 0.018 0.018 0.018 0.018 0.011 0.020 SOM 0.001 0.023 0.014 0.019 0.019 0.019 0.019 0.011 0.001 0.000 KS 0.002 0.022 0.013 0.018 0.017 0.017 0.017 0.012 0.002 0.001 0.020 PI 0.001 0.023 0.014 0.019 0.019 0.019 0.019 0.011 0.001 0.000 0.001 0.000 JP 0.037 0.054 0.043 0.049 0.049 0.049 0.049 0.050 0.038 0.038 0.037 0.038 0.000 LB 0.037 0.054 0.043 0.049 0.049 0.049 0.049 0.050 0.038 0.038 0.037 0.038 0.000 0.000 S 0.004 0.019 0.009 0.014 0.014 0.014 0.014 0.015 0.005 0.005 0.003 0.005 0.033 0.033 0.000 RA 0.005 0.020 0.011 0.016 0.015 0.015 0.015 0.017 0.006 0.006 0.005 0.006 0.035 0.035 0.001 0.020 PA 0.004 0.019 0.009 0.014 0.014 0.014 0.014 0.015 0.005 0.005 0.003 0.005 0.033 0.033 0.000 0.001 0.000 PT 0.013 0.030 0.021 0.026 0.025 0.025 0.024 0.020 0.013 0.013 0.013 0.013 0.049 0.049 0.014 0.016 0.014 0.017 KJ 0.004 0.022 0.012 0.017 0.017 0.017 0.017 0.016 0.005 0.005 0.004 0.005 0.036 0.036 0.003 0.004 0.003 0.014 0.005 ST 0.010 0.015 0.008 0.013 0.013 0.013 0.013 0.015 0.010 0.010 0.010 0.010 0.041 0.041 0.007 0.009 0.007 0.019 0.010 0.010 SKR 0.010 0.023 0.015 0.020 0.020 0.020 0.018 0.019 0.011 0.011 0.010 0.011 0.041 0.041 0.006 0.008 0.006 0.018 0.008 0.013 0.010 JBI 0.009 0.028 0.018 0.023 0.023 0.023 0.022 0.020 0.010 0.010 0.009 0.010 0.043 0.043 0.009 0.010 0.009 0.020 0.010 0.015 0.015 0.013 JMP 0.019 0.035 0.025 0.030 0.030 0.030 0.028 0.032 0.020 0.020 0.019 0.020 0.046 0.046 0.016 0.017 0.016 0.026 0.017 0.023 0.018 0.023 0.013 RBT 0.006 0.027 0.018 0.023 0.022 0.022 0.021 0.017 0.007 0.006 0.006 0.006 0.042 0.042 0.008 0.010 0.008 0.017 0.009 0.014 0.015 0.012 0.023 0.009 MRW 0.005 0.021 0.011 0.016 0.016 0.016 0.016 0.017 0.006 0.006 0.005 0.006 0.035 0.035 0.002 0.003 0.002 0.015 0.004 0.009 0.007 0.011 0.016 0.010 0.003 OG 0.047 0.065 0.054 0.059 0.059 0.059 0.059 0.061 0.049 0.048 0.047 0.048 0.038 0.038 0.043 0.045 0.043 0.059 0.046 0.052 0.051 0.053 0.057 0.052 0.046

142

Table 4.16: Pairwise genetic distances with TN92+G distribution model of Hp54 between H. pogonognathus populations.

TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW TE 0.000 TB 0.000 0.000 BP 0.000 0.000 0.001 PTG 0.005 0.005 0.005 0.008 SK 0.007 0.007 0.007 0.012 0.007 PJG 0.007 0.007 0.007 0.011 0.010 0.005 DM 0.000 0.000 0.000 0.005 0.007 0.007 0.000 SU 0.004 0.004 0.005 0.009 0.011 0.011 0.004 0.000 SOM 0.004 0.004 0.005 0.009 0.011 0.011 0.004 0.000 0.000 KS 0.006 0.006 0.006 0.010 0.013 0.013 0.006 0.004 0.004 0.005 PI 0.005 0.005 0.005 0.010 0.010 0.011 0.005 0.002 0.002 0.005 0.004 JP 0.017 0.017 0.018 0.023 0.024 0.025 0.017 0.022 0.022 0.024 0.022 0.000 LB 0.017 0.017 0.018 0.023 0.024 0.025 0.017 0.022 0.022 0.024 0.022 0.000 0.000 S 0.000 0.000 0.000 0.005 0.007 0.007 0.000 0.004 0.004 0.006 0.005 0.017 0.017 0.000 RA 0.007 0.007 0.007 0.008 0.014 0.011 0.007 0.011 0.011 0.013 0.011 0.025 0.025 0.007 0.007 PA 0.005 0.005 0.006 0.008 0.012 0.011 0.005 0.010 0.010 0.010 0.010 0.023 0.023 0.005 0.009 0.007 PT 0.004 0.004 0.005 0.007 0.011 0.011 0.004 0.009 0.009 0.007 0.009 0.022 0.022 0.004 0.011 0.005 0.000 KJ 0.004 0.004 0.004 0.007 0.010 0.010 0.004 0.008 0.008 0.007 0.008 0.021 0.021 0.004 0.010 0.005 0.001 0.001 ST 0.001 0.001 0.002 0.006 0.008 0.008 0.001 0.006 0.006 0.007 0.006 0.019 0.019 0.001 0.008 0.006 0.004 0.004 0.003 SKR 0.006 0.006 0.006 0.010 0.013 0.011 0.006 0.010 0.010 0.012 0.011 0.024 0.024 0.006 0.011 0.011 0.010 0.010 0.007 0.008 JBI 0.022 0.022 0.023 0.021 0.027 0.020 0.022 0.027 0.027 0.029 0.027 0.041 0.041 0.022 0.018 0.024 0.027 0.026 0.024 0.026 0.013 JMP 0.034 0.034 0.034 0.031 0.041 0.036 0.034 0.039 0.039 0.041 0.039 0.046 0.046 0.034 0.027 0.034 0.039 0.038 0.036 0.037 0.021 0.009 RBT 0.025 0.025 0.026 0.022 0.033 0.028 0.025 0.030 0.030 0.032 0.030 0.044 0.044 0.025 0.018 0.026 0.030 0.029 0.027 0.028 0.013 0.015 0.001 MRW 0.016 0.016 0.017 0.019 0.023 0.018 0.016 0.021 0.021 0.023 0.021 0.035 0.035 0.016 0.018 0.020 0.021 0.020 0.018 0.019 0.031 0.039 0.033 0.008 OG 0.013 0.013 0.013 0.018 0.017 0.020 0.013 0.017 0.017 0.020 0.018 0.031 0.031 0.013 0.020 0.019 0.017 0.017 0.014 0.019 0.036 0.048 0.039 0.030

143

For Hp5, the pairwise ФST values (Table 4.17) from AMOVA calculations after

Bonferroni corrections showed moderate to high significant value for most populations.

Comparisons among several populations within Siam drainage (S, RA, PA, PT and KJ), north Sunda drainage (JBI, JMP, RBT and MRW) and two populations from Malacca drainage (ST and SKR) showed from low to moderate high values with a few of the low pairwise FST values being non-significant. However, overall the ФST comparisons plots

(Figure 4.19) of Hp5 indicate that the H. pogonognathus populations in Sundaland river basins are highly structured. For Hp54, the pairwise ФST values (Table 4.18) after

Bonferroni corrections also showed moderate to high significant value with approximately half of the populations showing moderately high values. The ФST comparisons plots

(Figure 4.20) were also in agreement in that the H. pogonognathus populations are generally moderately structured with the southern Sumatra and Kelantan populations, being highly structured from the rest (Main comprising of Malacca and Siam –excluding

Kelantan).

The hierarchical AMOVA examined at all hierarchical level for both loci Hp5 and

Hp54 showed significant genetic structure (Table 4.19). The high and significant ФCT value, 0.649 (Hp5) and 0.586 (Hp54), and low within population variation shows that most of the variance was attributed to differences among nDNA lineages (Main, southern

Sumatra, Kelantan). The high ФST value of 0.856 and 0.809 in Hp5 and Hp54 lineages respectively suggested that the H. pogonognathus populations are highly structured.

144

Table 4.17: Pairwise ФST values of Hp5 (below diagonal) between H. pogonognathus populations. Bold values indicate non-significant ФST values after Bonferroni corrections (p > 0.05). Geographical distances (above diagonal in italics) between populations of the shortest path follow the Paleo-drainages (km).

KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW

KWK 0.000 317 320 719 742 995 992 1056 2225 2324 2203 2258 2461 2504 2304 2307 2137 1505 2120 1351 1474 1794 1981 1812 2339 TE 0.875 0.000 233 750 804 1025 1019 1115 2324 2359 2168 2246 2524 2536 2388 2324 2198 1550 2100 1362 1488 1851 2023 1851 2397 TB 0.839 1.000 0.000 772 790 1101 1077 1176 2346 2375 2197 2277 2546 2590 2411 2389 2249 1566 2145 1357 1496 1903 2069 1866 2322 BP 0.889 0.974 0.936 0.000 27 1062 1014 1121 2295 2314 2174 2285 2549 2600 2375 2319 2178 1520 2164 1356 1499 1851 1958 1803 2303 PTG 0.868 1.000 1.000 -0.040 0.000 1088 1041 1150 2326 2343 2200 2310 2579 2635 2398 2350 2201 1536 2190 1389 1531 1890 1975 1841 2321 SK 0.868 1.000 1.000 -0.040 0.000 0.000 90 441 1617 1716 1478 1626 2091 2145 1715 1657 1652 912 1447 774 894 1207 1340 1190 1686 PJG 0.759 0.518 0.513 0.665 0.605 0.605 0.000 480 1601 1706 1457 1601 2080 2127 1705 1633 1631 901 1413 812 950 1212 1321 1180 1655 DM 0.819 0.898 0.911 0.938 0.932 0.932 0.586 0.000 1500 1514 1382 1444 1669 1610 1580 1481 1324 693 1304 548 662 1050 1130 1001 1477 SU 0.628 0.934 0.904 0.935 0.929 0.929 0.714 0.814 0.000 163 130 92 1365 1400 1383 1143 528 1014 648 1525 1314 2273 2417 2416 2189 SOM 0.638 1.000 1.000 0.983 1.000 1.000 0.731 0.874 0.012 0.000 166 147 1308 1341 1168 1126 530 1050 660 1554 1346 2253 2386 2200 2236 KS 0.642 0.935 0.901 0.933 0.927 0.927 0.735 0.839 0.029 0.131 0.000 126 1076 1111 1004 984 501 819 521 1384 1146 2128 2280 2141 2103 PI 0.726 1.000 1.000 0.989 1.000 1.000 0.809 0.917 0.090 0.000 0.227 0.000 1142 1186 1035 1030 583 885 570 1413 1323 2173 2328 2159 2163 JP 0.898 1.000 1.000 0.992 1.000 1.000 0.875 0.967 0.961 1.000 0.960 1.000 0.000 53 334 414 998 1035 1043 1751 1645 2302 2507 2300 2245 LB 0.901 1.000 1.000 0.993 1.000 1.000 0.879 0.969 0.963 1.000 0.961 1.000 0.000 0.000 380 455 1042 1080 1088 1700 1596 2347 2560 2390 2299 S 0.707 1.000 1.000 0.977 1.000 1.000 0.640 0.909 0.732 1.000 0.650 1.000 1.000 1.000 0.000 334 944 1022 968 1639 1454 2229 2354 2176 2155 RA 0.642 0.844 0.765 0.856 0.827 0.827 0.639 0.822 0.573 0.633 0.524 0.736 0.899 0.903 0.161 0.000 936 923 915 1663 1431 2155 2339 2121 2038 PA 0.714 1.000 1.000 0.978 1.000 1.000 0.648 0.912 0.740 1.000 0.659 1.000 1.000 1.000 0.000 0.172 0.000 819 457 1440 1263 2242 2383 2213 2095 PT 0.491 0.583 0.486 0.648 0.570 0.570 0.485 0.524 0.263 0.239 0.303 0.369 0.715 0.725 0.293 0.373 0.305 0.000 781 716 551 2142 2248 2417 2125 KJ 0.532 0.758 0.650 0.779 0.734 0.734 0.589 0.718 0.291 0.327 0.230 0.449 0.839 0.844 0.073 0.192 0.082 0.247 0.000 1343 1169 2145 2327 2176 2086 ST 0.568 0.532 0.260 0.545 0.490 0.490 0.340 0.563 0.364 0.396 0.364 0.494 0.805 0.810 0.207 0.289 0.215 0.317 0.231 0.000 727 1012 1194 980 1586 SKR 0.587 0.652 0.561 0.714 0.652 0.652 0.504 0.666 0.442 0.445 0.426 0.572 0.782 0.790 0.168 0.258 0.179 0.247 0.108 0.263 0.000 903 1003 825 1421 JBI 0.518 0.646 0.555 0.693 0.630 0.630 0.539 0.626 0.324 0.303 0.325 0.424 0.735 0.743 0.249 0.328 0.260 0.294 0.163 0.335 0.250 0.000 646 485 920 JMP 0.643 0.671 0.611 0.703 0.658 0.658 0.602 0.697 0.569 0.550 0.577 0.632 0.704 0.711 0.458 0.510 0.466 0.446 0.436 0.524 0.368 0.403 0.000 173 1089 RBT 0.550 0.722 0.649 0.758 0.709 0.709 0.615 0.655 0.326 0.297 0.345 0.412 0.790 0.796 0.402 0.405 0.413 0.321 0.240 0.412 0.348 0.092 0.454 0.000 912 MRW 0.618 0.757 0.645 0.767 0.728 0.728 0.601 0.748 0.466 0.494 0.424 0.591 0.833 0.837 0.112 0.221 0.120 0.344 0.072 0.287 0.094 0.237 0.444 0.349 0.000

145

Table 4.18: Pairwise ФST values of Hp54 (below diagonal) between H. pogonognathus populations. Bold values indicate non-significant ФST values after Bonferroni corrections (p > 0.05). Geographical distances (above diagonal in italics) between populations of the shortest path follow the Paleo- drainages systems (km).

TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW TE 0.000 233 750 804 1025 1019 1115 2324 2359 2168 2246 2524 2536 2388 2324 2198 1550 2100 1362 1488 1851 2023 1851 2397 TB 0.870 0.000 772 790 1101 1077 1176 2346 2375 2197 2277 2546 2590 2411 2389 2249 1566 2145 1357 1496 1903 2069 1866 2322 BP 0.707 0.149 0.000 27 1062 1014 1121 2295 2314 2174 2285 2549 2600 2375 2319 2178 1520 2164 1356 1499 1851 1958 1803 2303 PTG 0.394 0.377 0.374 0.000 1088 1041 1150 2326 2343 2200 2310 2579 2635 2398 2350 2201 1536 2190 1389 1531 1890 1975 1841 2321 SK 0.454 0.356 0.469 0.439 0.000 90 441 1617 1716 1478 1626 2091 2145 1715 1657 1652 912 1447 774 894 1207 1340 1190 1686 PJG 0.695 0.456 0.521 0.519 0.431 0.000 480 1601 1706 1457 1601 2080 2127 1705 1633 1631 901 1413 812 950 1212 1321 1180 1655 DM 0.438 0.651 0.460 0.285 0.386 0.626 0.000 1500 1514 1382 1444 1669 1610 1580 1481 1324 693 1304 548 662 1050 1130 1001 1477 SU 0.923 1.000 0.691 0.558 0.545 0.654 0.825 0.000 163 130 92 1365 1400 1383 1143 528 1014 648 1525 1314 2273 2417 2416 2189 SOM 0.932 1.000 0.705 0.588 0.564 0.670 0.838 0.000 0.000 166 147 1308 1341 1168 1126 530 1050 660 1554 1346 2253 2386 2200 2236 KS 0.708 0.468 0.527 0.509 0.529 0.578 0.638 0.190 0.214 0.000 126 1076 1111 1004 984 501 819 521 1384 1146 2128 2280 2141 2103 PI 0.660 0.400 0.415 0.433 0.468 0.559 0.551 0.083 0.105 0.193 0.000 1142 1186 1035 1030 583 885 570 1413 1323 2173 2328 2159 2163 JP 0.969 1.000 0.906 0.751 0.798 0.866 0.935 1.000 1.000 0.879 0.860 0.000 53 334 414 998 1035 1043 1751 1645 2302 2507 2300 2245 LB 0.967 1.000 0.903 0.741 0.793 0.862 0.932 1.000 1.000 0.875 0.855 0.000 0.000 380 455 1042 1080 1088 1700 1596 2347 2560 2390 2299 S 0.754 0.691 0.317 0.210 0.528 0.634 0.466 0.845 0.857 0.648 0.516 0.932 0.930 0.000 334 944 1022 968 1639 1454 2229 2354 2176 2155 RA 0.466 0.389 0.452 0.005 0.481 0.489 0.400 0.543 0.571 0.546 0.469 0.740 0.731 0.382 0.000 936 923 915 1663 1431 2155 2339 2121 2038 PA 0.285 0.238 0.292 0.004 0.360 0.440 0.156 0.438 0.462 0.369 0.368 0.697 0.689 0.247 0.163 0.000 819 457 1440 1263 2242 2383 2213 2095 PT 0.848 1.000 0.796 0.453 0.500 0.725 0.740 1.000 1.000 0.656 0.716 1.000 1.000 0.871 0.534 0.176 0.000 781 716 551 2142 2248 2417 2125 KJ 0.705 0.819 0.754 0.513 0.533 0.735 0.612 0.883 0.892 0.676 0.724 0.952 0.951 0.806 0.606 0.227 -0.030 0.000 1343 1169 2145 2327 2176 2086 ST 0.633 0.029 0.198 0.350 0.337 0.453 0.433 0.647 0.674 0.406 0.390 0.910 0.906 0.540 0.402 0.193 0.665 0.632 0.000 727 1012 1194 980 1586 SKR 0.388 0.183 0.283 0.265 0.361 0.368 0.276 0.433 0.457 0.430 0.368 0.743 0.735 0.350 0.294 0.224 0.451 0.501 0.164 0.000 903 1003 825 1421 JBI 0.687 0.646 0.717 0.453 0.656 0.613 0.701 0.707 0.724 0.714 0.681 0.805 0.799 0.681 0.348 0.514 0.714 0.769 0.659 0.577 0.000 646 485 920 JMP 0.826 0.814 0.847 0.685 0.795 0.808 0.838 0.840 0.850 0.828 0.813 0.872 0.868 0.834 0.616 0.693 0.840 0.867 0.810 0.734 0.485 0.000 173 1089 RBT 0.948 0.957 0.917 0.763 0.838 0.867 0.934 0.965 0.967 0.897 0.888 0.977 0.977 0.935 0.665 0.732 0.965 0.947 0.914 0.782 0.451 0.661 0.000 912 MRW 0.808 0.793 0.809 0.585 0.749 0.764 0.802 0.830 0.843 0.798 0.767 0.880 0.875 0.778 0.538 0.603 0.830 0.856 0.776 0.648 0.650 0.783 0.875 0.000

146

Figure 4.19: Pairwise ФST comparisons of Hp5 (below diagonal) between H. pogonognathus populations.

Figure 4.20: Pairwise ФST comparisons of Hp54 (below diagonal) between H. pogonognathus populations.

147

Table 4.19: AMOVA results for hierarchical genetic subdivision for percentage of variation and fixation indices (ФST, ФSC and ФCT) of Hp5 and Hp54 of H. pogonognathus populations. Bold values indicate significant value (p < 0.05).

Among Among populations Within Group ФST ФSC ФCT groups (%) within groups (%) population (%) Paleo-drainages 0.649 0.579 0.166 16.59 48.28 35.13 Hp5 lineages 0.856 0.590 0.649 64.89 20.71 14.40 Paleo-drainages 0.727 0.599 0.319 31.93 40.76 27.31 Hp54 lineages 0.809 0.540 0.586 58.57 22.35 19.08

The Mantel test results for both HP5 and Hp54 (Figure 4.21 and Figure 4.22) revealed significant but weak correlations between pairwise ФST and geographical distances among the H. pogonognathus populations (r = 0.281, p<0.05 and r = 0.345, p

<0.05 respectively). The weak Mantel correlation results (although significant) shows that isolation by distance is not the major factor to describe the overall pattern of genetic differentiation at this spatial scale.

148

1.2

1.0

0.8

0.6

ST 0.4 Ф R² = 0.0792 0.2

0.0 0 500 1000 1500 2000 2500 3000 -0.2 Geographical Distance (km)

Figure 4.21: The ФST plotted against the shortest path following the hypothetical Paleo- drainages systems (km) of H. pogonognathus Hp5 data. Trendline (Black dashed line) shows the general pattern of little increasing genetic distance with greater geographical distance (IBD).

1.2 1 0.8 0.6

ФST 0.4 0.2 R² = 0.1191 0 0 500 1000 1500 2000 2500 3000 -0.2 Geographical Distance (km)

Figure 4.22: The ФST plotted the shortest path following the hypothetical Paleo-drainages systems (km) of H. pogonognathus Hp54 data. Trendline (black dashed line) shows the general pattern of little increasing genetic distance with greater geographical distance (IBD).

149

4.4.2(e) Population history and demographic changes

No evidence for recent population expansion was observed for all populations as

Tajima’s D, Fu’s FS and R2 (except SU population) were non-significant for both Hp5 and

Hp54 (Table 4.20). Four populations (TB, SOM, JP and LB - in both loci) and six populations (TE, SK, SU, S, PA and PT – in either one of the locus) could not be tested for deviation from mutation drift as no gene/haplotype variation was observed. The calculated Hri for each population (except SKR population) showed non-significant values which indicates lack of support for a stable population. Further inspection with mismatch distribution (Figure 4.23 and Figure 4.24) showed distribution with either demographic growth and subsequent bottleneck, or populations that had undergone rapid reduction towards its new equilibrium. Generally, taken together, the non-significant neutrality test results, Hri value and mismatch distribution suggest rapid population reduction (bottleneck) towards its new equilibrium for all testable populations.

Interestingly, the total sample dataset of both loci exhibited large significant negative values in Fu’s Fs tests (Table 4.20). This suggests that population expansion occurred in the overall H. pogonognathus populations. The Fu’s Fs test is considered to be more sensitive in detecting population expansion and suggested possible past population growth. Furthermore, the mismatch distribution of total sample dataset for both loci exhibited unimodal distribution which indicates demographic growth, followed however by population reduction towards its new equilibrium.

150

Table 4.20: Summary of population neutrality tests and demographic analyses. Tajima’s D, Fu’s Fs, Rasmos-Onsins & Rozas (R2) and Harpending’s raggedness index (Hri) of Hp5 and HP54 of H. pogonognathus populations.

Locus Hp5 Hp54

Population Tajima’s D Fu’s Fs R2 Hri Tajima’s D Fu’s Fs R2 Hri KWK 0.696 3.167 0.162 0.237 / / / / TE 0.000 0.000 0.000 0.000 0.334 0.536 0.214 0.204 TB 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 BP -1.164 0.880 0.000 0.650 0.406 2.889 0.153 0.430 PTG 0.000 0.000 0.218 0.000 0.640 3.136 0.215 0.254 SK 0.000 0.000 0.000 0.000 2.598 8.305 0.259 0.768 PJG 1.424 -0.153 0.207 0.104 1.465 0.163 0.209 0.064 DM 1.026 1.169 0.221 0.209 1.529 5.163 0.267 0.787 SU -0.685 1.175 0.105 0.713 0.000 0.000 0.000 0.000 SOM 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 KS 0.776 0.533 0.196 0.195 1.103 0.502 0.199 0.098 PI 0.000 1.324 0.000 0.209 0.338 -0.605 0.162 0.052 JP 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 LB 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 S 0.000 0.000 0.000 0.000 -0.578 2.536 0.117 0.697 RA 0.405 -0.070 0.163 0.164 2.265 3.208 0.256 0.192 PA 0.000 0.000 0.000 0.000 1.279 0.893 0.194 0.029 PT 1.997 1.025 0.227 0.043 0.000 0.000 0.000 0.000 KJ -0.509 -0.964 0.113 0.059 -0.328 0.933 0.126 0.191 ST 1.201 -0.081 0.177 0.070 -0.278 1.599 0.152 0.199 SKR 0.887 1.571 0.180 0.256 1.097 1.525 0.188 0.135 JBI 1.015 0.201 0.180 0.070 0.422 3.052 0.163 0.101 JMP 1.574 -2.157 0.183 0.031 0.592 1.797 0.167 0.202 RBT 0.928 5.722 0.170 0.498 0.352 0.721 0.168 0.220 MRW 0.651 -1.067 0.156 0.034 1.541 2.352 0.231 0.179 Total -1.109 -24.367 0.045 0.011 -0.560 -12.448 0.054 0.004 / no data is available Bold values indicate significant with p < 0.05.

151

BP DM JBI

JMP KJ KS

KWK MRW PJG Figure 4.23: Mismatch distribution of Hp5 of 12 H. pogonognathus populations showing the expected and observed pairwise differences between the sequences with the respective frequencies under constant population size. The solid lines represent the expected distribution and the dotted lines represent the observed distribution. The dotted line shows that the left edge of distribution converges rapidly toward the new equilibrium.

152

PT RA RBT

SKR ST SU

Total samples Figure 4.23: continue…

153

BP DM JBI

JMP KJ KS

MRW PA PI Figure 4.24: Mismatch distribution of Hp54 of 12 H. pogonognathus populations showing the expected and observed pairwise differences between the sequences with the respective frequencies under constant population size. The solid lines represent the expected distribution and the dotted lines represent the observed distribution.

154

PJG PTG RA

RBT S SK

SKR ST TE Figure 4.24: continue…

155

Total samples Figure 4.24: continue…

156

The EBSPs analysis (Figure 4.25) of Hp5 shows that a gradual increase began around 3mya followed by rapid recent decline. For Hp54, a constant population size was detected and eventually declined. However, no obvious expansion was observed in agreement with Fu’s Fs test. Overall, both loci showed a recent occurrence of population reduction through a possible population bottleneck, congruent with the mismatch distribution analysis.

157

Hp5 14 12 10 8 6 4 2

Population sizescalar Population 0 0 2 4 6 8 10 Time (millions of years before present) (a)

Hp54 15

Population sizescalar Population 0 0 2 4 6 8 10 Time (millions of years before present) (b)

Figure 4.25: Extended Bayesian Skyline Plots (EBSPs) showing the demographic history of H. pogonognathus of (a) Hp5 and (b) Hp54. Solid blue line is the median effective population size, the dashed lines are the upper and lower 95% HPD for those estimates.

158

4.5 Discussion

4.5.1 Genetic Diversity and haplotype distribution

One of the crucial steps in developing conservation and management strategies for natural fish populations, are to assess and understand various critical information such as genetic diversity and population genetic structure of the target species. In this study, low nucleotide diversity was observed across H. pogonognathus populations.

Three populations (TB, SOM and LB) showed total absence of genetic variability in both mtDNA and nDNA markers which may be explained by several factors. These populations may have been from only ‘a single female origin’ since mtDNA is maternally inherited (Avise, 1994). Additionally, inadequate or limited number of samples could have translated into an underestimation of the actual genetic variation.

Another causal factor could be founder effect, a consequence of the recolonization of the Sunda Shelf as sea level rose after the last glaciations (Kochzius & Nuryanto, 2008) especially for the Malacca drainage; most populations from this drainage showed low haplotype diversities which could be attributed to this or to the small sample sizes.

Another factor could be population bottleneck (and supported by the neutrality tests and historical demographic patterns) due to habitat loss as a result of human activities

(Kochzius & Nuryanto, 2008) including deforestation, pollution and even overexploitation (Hauser, et al., 2002). However, it is unlikely that this forest halfbeak has been subjected to the latter due to its lack of commercial value with limited demand in the ornamental industry.

Other than that, a wide range of haplotype diversity was observed for both mtDNA and nDNA markers across the populations. High variation was present

159 particularly in those populations (KS, KJ, ST, SKR and JMP) located at adjacent drainage basin (at the extended river mouths or tips of paleo-drainages) but representing distinct paleo-drainages (north Sunda, southern Siam and Malacca drainages). The high haplotype diversity observed in the mentioned populations, parallels other studies of freshwater fishes inhabiting non-glaciated region (past glaciations era) or temperate regions (Bernatchez & Wilson, 1998; Roos, 2004) also observed high haplotype diversity. Several population studies of freshwater fish inferred by cyt b have also documented high haplotype diversity including studies of freshwater sprat Zacco platypus (Perdices, et al., 2004), mahseer Tor tor (Komal, et al., 2013), golden mahseer T. putitora (Sati, et al., 2013), Salangid icefish Neosalanx taihuensis (Zhao, et al., 2008). Similarly, studies of striped snakehead Channa striata in the same region (Sundaland river basins) inferred through cyt b (Adamson, 2010),

ND-5 gene (Tan, et al., 2012) and COI gene (Tan, et al., 2015) also documented high levels of diversity on most of the populations. Worth noting, low nucleotide within population but relatively high in total sample dataset implied that there was genetic variation among the populations which indicated non-homogeneous distributions for

H. pogonognathus populations across Sundaland.

Most of the mtDNA haplotypes were unique to specific populations except

Hpb02, Hpb07 and Hpb29. A recent bottleneck event could also have led to the loss of haplotypes in populations, with the concomitant occurrence of population-specific haplotypes. Incomplete sampling may also cause the failure in discovering haplotypes

(Liao, et al., 2010). Of the two, bottleneck event(s) is more likely, given the historical demography findings. The nuclear DNA markers, had a few more shared haplotypes among populations across drainages compared to mtDNA, four in H5 (Hp515, Hp516,

160

Hp532 and Hp533) and five haplotypes in Hp54 (Hp5402, Hp5403, Hp5404, Hp5406 and Hp5413) with a maximum of seven populations having a haplotype in common.

The haplotype sharing in mtDNA and nDNA markers signals some level of gene flow or migration (although minimal) between populations located at adjacent drainages basins but representing distinct paleo-drainages (among Malacca, north

Sunda and Siam drainages). These Paleo-drainages were believed to have connected many rivers of Sumatra, Peninsular Malaysia and Borneo. During the low sea level era, the majority of the extended river basins would had been almost fully developed and connected to each other thus could have led to extensive opportunities of faunal exchanges (McConnell, 2004). As concluded by a number of topological studies of the extended Pleistocene river basins on the Sunda shelf (Rainboth, 1996; Voris, 2000), the extended river mouths at the southern region of Malacca drainage is thought to have been joined at some point in time to the extended Siam drainage (southern region) and also the extended north Sunda drainage during that time (Voris, 2000)of the

Pleistocene inter-glacial events. Subsequently, these populations were separated by sea as the sea level rose until present day. This occurrence is concordant to the population divergence time constructed by de Bruyn et al. (2013) (Figure 4.26). According to the constructed divergence time, most of the H. pogonognathus populations started to diverge between 200kya to 1.3mya which coincided with the cyclical glaciation events during Pleistocene.

161

Figure 4.26: Divergence time of H. pogonognathus populations in Sundaland river basins based on ultrametric Bayesian mitochondrial DNA trees of COI and control region variation. Values at nodes are median ages (in millions of years; bars=95% highest posterior densities). (Adopted and modified from de Bruyn, et al., 2013).

162

Separation of populations as sea level rose had created geographical barriers that led to the restriction of gene flow. This was reflected in mtDNA marker where there was virtually almost no haplotype sharing among populations. Mitochondrial

DNA has greater sensitivity in detecting more recent population history compared to nDNA as it has a 4X faster substitution rate and smaller effective population size

(Avise 1987; 2009). Nuclear DNA markers on the other hand showed the occurrence of higher haplotype sharing. This can be explained by their higher sensitivity in detecting ancient population history with the relatively deep coalescent times (Harding, et al., 1997; Fu & Li, 1999) due to low substitution rate and larger effective population size. The haplotype sharing is attributed to the recently separated populations which have insufficient time for genetic drift to produce new haplotype particularly in nDNA

(Printzen, et al., 2003). The genotype sharing of these nDNA markers could also be due to the breakdown of refugial or genetic isolation in nuclear genome through secondary contact, and/or also facilitated by male-mediated gene flow (Barlow, et al.,

2013) during the cyclical glaciation events. Furthermore, incomplete lineage sorting, introgression and retention of ancestral polymorphisms (Beck, et al., 2013) may not be discounted. Incomplete lineage sorting is a common recognized source of genotype sharing between newly evolved and closely related taxa (Hawlitschek, et al., 2012) in agreement with the recent radiation hypothesis of H. pogonognathus. The retention of ancestral polymorphism will produce misleading level of high gene flows and phylogeny with lack of distinctiveness (Bulgin, et al., 2003). However, this does not seem a suitable explanation as the gene flows among populations was not pronounced.

The present study is in broad agreement with several previous studies which have shown haplotype sharing across drainages in the Sunda shelf region including in

163 the freshwater cyprinid Hampala macrolepidota (Ryan & Esa, 2006); mahseer T. tambroides (Esa, et al., 2008); tinfoil barb Barbonymus schwanenfeldii (Kamarudin &

Esa, 2009) and snakehead C. striata (Tan, et al., 2012; 2015).

4.5.2 Population structure and phylogeography

Characteristic molecular variants are to be expected in populations isolated due to geographical barriers over a lengthy period of time. The independently evolutionary process occurring in these isolated populations resulted in significant population structure across drainages as shown in overall high ФST value with weak ‘isolation by distance’ relationship although some populations show low ФST value in nDNA markers which might be attributed to the low substitution rates. High ФST values between populations indicate very low gene flow or absence of contemporary migration between populations. The presence of unique haplotypes in both DNA markers in this study also provides evidence of high population structuring (So, et al.,

2006a). Another factor that could account for the high population structuring observed in H. pogonognathus populations could be attributed to the specialized habitat requirement (forest stream dependent) (Collette, 2004) that limit the dispersal. Non- migratory freshwater fishes tend to appear high genetic structuring and has been well documented (Dominguez-Dominguez, et al., 2008; Michel, et al., 2008; Kamarudin &

Esa, 2009). Furthermore, the cyclical sea level and climatic oscillations during

Pleistocene were also postulated to result in the extensive diversification of numerous populations (Hewitt, 2000; April, et al., 2013) driving the H. pogonognathus populations into refugia isolation with subsequent structuring.

164

The phylogeographical signatures derived from ancestral drainage configurations will be likely reflected by the contemporary population structure of aquatic taxa (Hurwood, et al., 1998; Waters, et al., 2001; Poissant, et al., 2005). Many studies have illustrated how phylogeographical pattern have arisen as a direct result of the upthrust of mountains, division of seas or retreat of glaciers (Avise, 1994; Soltis, et al., 2006; Zemlak, et al., 2008). The lowering of sea level during Pleistocene would have re-connected paleo-drainages especially between adjacent drainage basins, facilitating H. pogonognathus dispersal across independent paleo-drainages. This is evident in the incidence of shared haplotypes (Hp515, Hp516, Hp532, Hp533, Hp5402,

Hp5403, Hp5404 and Hp5406). Based on several lines of evidence, the Siam drainage is hypothesised to be the ancestor for other drainages and the southern Sumatra haplotypes represent the descendant haplotypes. Generally, H. pogonognathus populations from the same ancestral drainage but in different contemporary landmass are closely related while those from adjacent drainage basins but representing different paleo-drainages were phylogenetically dissimilar. This is in broad agreement with previous studies of catfish H. nemurus by Dodson, et al. (1995) reported that the population connectivity was affected by the Paleo-drainages rearrangement resulting from glaciation events during Pleistocene. Fish sampled from the same ancestral drainage were genetically closely related while dissimilarity occurred among samples from nearby locations on the same landmass but in different ancestral drainages.

Similar conclusions have been reported for freshwater barb B. gonionotus (McConnell,

2004), freshwater cyprinid H. macrolepidota (Ryan & Esa, 2006), mahseer T. tambroides (Esa, et al., 2008), tinfoil barb B. schwanenfeldii (Kamarudin & Esa, 2009) and snakehead C. striata (Tan, et al., 2012; 2015).

165

Haplotype sharing patterns revealed influence of Paleo-drainage rearrangement and connectivity between independent ancestral drainages coupled with the cyclical glaciation events during Pleistocene. On the other hand, the phylogeny of

H. pogonognathus was not in agreement with Paleo-drainage delineation. The major lineages/clades across the H. pogonognathus distribution in Sundaland river basins show limited Paleo-drainage influenced delineation. The delineation events associated with the Paleo-drainages was only evident in the genetic distinctiveness of the southern

Sumatra clade from north Sunda drainage.

Based on the population divergence dating (Figure 4.26), it is postulated that the southern Sumatra clade had separated around 3.3Mya in the late Pliocene indicating that the divergence was not influenced by the Paleo-drainage events during

Pleistocene. In addition, the development of the admixed haplotype clade (Main clade), being composed of two drainages (Malacca and Siam) as inferred by both mtDNA and nDNA markers, is not in agreement with a Paleo-drainage delineation. AMOVA revealed that regional structuring of the DNA marker lineages and not Paleo-drainage lineages explained much of the total genetic variance. The phylogeographic study of three halfbeak species by de Bruyn, et al. (2013), had reported that generally the species tree that followed the Paleo-drainage delineations is preferable, however, the authors also reported that unlike the Dermogenys species, the phylogeny of

Hemirhamphodon showed limited influence of Paleo-drainage delineation, and the current study is in agreement with these earlier findings.

It must be noted that both DNA markers revealed that the Kelantan clade (JP and LB) was highly diverged from other populations including those from the same

166 ancestral drainage (Siam drainage). The divergence dating (Figure 4.26) analysis showed that the Kelantan clade has been diverged before the Pleistocene (around

5.3Mya) i.e. since late Miocene or early Pliocene to form a reciprocal monophyletic clade. The cyclical glaciation events during Pleistocene seem to have no effect on this clade in terms of gene exchange with other populations where it harboured private haplotypes separated by high number of mutations. More intensive sampling to the north of Jeram Pasu and Lata Belatan populations (southern region of Thailand) should be conducted in order to investigate the evolution of different lineages along the Siam drainage.

4.5.3 Historical demography

Historical demographic analyses able to reveal that populations of H. pogonognathus might had experienced past demographic events and population size fluctuations which could be traced in patterns of their genetic diversity (Liao, et al.,

2010). The identification of populations that have undergone ancient and recent bottlenecks is necessary for any management program, not only beneficial to the studied species but also a useful reference to closely related sympatric species. These small populations would have gone through demographic stochasticity, leading to inbreeding or fixation of deleterious alleles. Eventually the evolutionary potential will be reduced and thus lead to increment of extinction probability (So, et al., 2006a) which could only be addressed with rapid action using effective measures.

167

The neutrality test of Tajima’s D and Fu’s Fs neutrality tests suggested neutral evolution i.e no selection and changes in population size at all sample sites. Mismatch distribution analysis of R2 was also in agreement of no demographic growth for all sites. However, the raggedness index did not support a stable population. Both raggedness index and the mismatch distribution plots suggested the occurrence of rapid reduction in population size for most populations. Based on these results, it is hypothesised that the demographic history of H. pogonognathus had experienced rapid population reduction or recent population bottleneck towards its new equilibrium. As suggested earlier, the bottleneck event could be one of the reasons for the occurrence of population specific or private haplotype (Liao, et al., 2010). Another plausible factor for the demographic reduction is loss of habitat during the rise of sea level at the end of the last glacial. In addition, specialized habitat requirement such as forest stream dependence that limited dispersal, coupled with habitat destruction due to anthropogenic activities could also have led to such situation. Interestingly, the population expansion event of H. pogonognathus populations was detected by nDNA markers for the total sample dataset. This implies that genetic variability of H. pogonagnathus populations had accumulated in nuclear markers which coalesced into a longer genealogy and also reflects the ability of nDNA markers in revealing insights into older events.

Concordant with traditional demography, the Extended Bayesian Skyline Plots

(EBSP) also empirically demonstrated the occurrence of older expansions and recent population bottleneck. The EBSP of nDNA markers Hp5 (Figure 4.25a) revealed an older expansion event detected around 3mya which may reflect the spread of the H. pogonognathus Main clade and the southern Sumatra clade (as shown by population

168 divergence time in Figure 4.26). This occurrence may account for the high divergence between the two clades as observed in gene trees and pairwise distance. The H. pogonognathus populations in each clade is postulated to expand and subdivide during the Pleistocene as shown in Figure 4.26 before rapidly declining at the end of the

Pleistocene (recent population bottleneck) as also detected in ESBP of mtDNA cyt b, which is attributed mainly to the rise of sea-level and loss of associated habitats as

Sundaland flooded (Voris, 2000; Woodruff, 2010). This parallels the hypothesis by

Woodruff (2010) where in general the SE Asian freshwater taxa currently are believed to be in a refugial phase.

4.5.4 Evidence for Cryptic Species

Occurrence of cryptic diversity is very common especially in the tropical ecosystem which is believed to harbour immeasurable number of undescribed species

(Kadarusman, et al., 2012; Puckridge, et al., 2013; Hubert & Hanner, 2015). One of the objectives of this study was to examine the species status of the H. pogonognathus complex at population level as the potential for cryptic diversity of this species was revealed and discussed in Chapter 3.

MtDNA and nDNA markers were in congruence with respect to the high divergence of the Kelantan clade. These clearly support the DNA barcording results described in Chapter 3. According to Johns & Avise (1998), divergences between species among 81 confamilial fish genera through cyt b gene has an average value of

12%. In this study, the cyt b divergences between Kelantan populations and other

169 populations were higher ranging from 13.5% to 17.9%. Using the same gene much lower intraspecific divergences have been reported i.e. 4.5% for freahwater chubc

Leuciscus peloponnensis (Zardoya & Doadrio, 1999), 6.3% for L. leuciscus (Costedoat, et al., 2006) and 8.3% for sisorid catfish Glyptothorax (Chen, et al., 2007). However, there is no distinct genetic distance cut-off point for species demarcation as was discussed in Chapter 3. Based on molecular dating the Kelantan clade diverged approximately 5.3mya coinciding with the late Miocene or early Pliocene era. This prolonged isolation period would have provided sufficient time for the Kelantan clade to accumulate high genetic divergence to form a reciprocally monophyletic clade. In addition, the occurrences of unique haplotypes also reflect a long period of lineage sorting. As discussed also in Chapter 3, no obvious morphological differentiation was observed between the Kelantan clade and the other populations. It is a common observation that closely related species do not necessarily differ morphologically (Li, et al., 2009). Lack of morphological differences among recently diverged sibling species could be attributed to the lack of proper selection on morphological characters or insufficient time for morphological traits to be varied. Similar morphology between

Kelantan and other H. pogonognathus clades, though they were diverged a long time ago could be attributed to the similarities of both their diet and habitat selection

(personal observation). The high sequence divergences coupled with the findings in gene trees (this study and DNA barcording in Chapter 3) and MSN, support the hypothesis that the Kelantan clade should be proposed as new or cryptic species for the Hemirhamphodon complex in Sundaland river basins.

According to Anderson & Collette, (1991), the holotype of H. pogonognathus was found in Merawang, Bangka Island of southern Sumatra. Thus, it is worth noting

170 that the H. pogonognathus of the Main clade (Peninsular Malaysia and central Sumatra) also showed high potential of being elevated to subspecies or cryptic species level. The genetic divergence between the Main clade and southern Sumatra populations in both mtDNA and nDNA markers was also remarkably high, even though lower than that exhibited by the Kelantan clade. Interestingly, genetic divergence for this clade in

Hp54 was higher (although only slightly) than the Kelantan clade. Moreover, the divergence time of the Main clade from the southern Sumatra clade took place around

3.3Mya (supported by EBSP result of Hp5) which is more recent than the divergence time of Kelantan clade (5.3Mya). Based on these observation, it is predicted that in time the Main clade would develop into a new taxon in the Hemirhanphodon complex.

However, more detailed investigations integrating molecular and morphological attributes are required in order to accurately delineate this species complex (Goldstein

& De Salle, 2011).

4.5.5 Conservation

One of the essential steps to prevent the further loss of biodiversity is quantifying species richness. As shown by the results of this study, the H. pogonognathus populations is experiencing bottleneck and currently in refugial phase as also postulated for the SE Asian freshwater taxa (Woodruff, 2010) as a whole.

According to Kuussaari, et al. (2009), the halfbeak species is facing a high risk in localised stochastic extinction events or already below the extinction threshold and a clear status of population viability is still unknown. Various factors through natural process as well as increased anthropogenic threats may lead to the further decline and diminished population connectivy of this species.

171

Wright, et al. (2009), noted that SE Asian species are most vulnerable to climatic and other perturbations. In addition, the movement of freshwater-limited species in looking for refugia is restricted due to lack of drainage connectivity (April, et al., 2013). This study also revealed significant divergence levels among clades, even among populations which warrants separate conservation management. Any further loss of populations might result in the risk of losing potential cryptic species.

172

4.6 Conclusion

In conclusion, this study supports the effectiveness of incorporating both mtDNA and nDNA in population investigations and phylogeography for a comprehensive insight. The mtDNA with its haploid nature has a smaller effective population size and coupled with high mutation rate is effective in lineage sorting. On the other hand, the nuclear DNA with lower mutation rates, has low resolution in phylogeny. However, it has a better resolving power in assessment of genotype sharing that may have coalesced in more ancient time and that might involve male-mediated gene flow. Analyses of both mtDNA and nDNA markers have given a more in depth picture of the evolutionary process of H. pogonognathus. Knowledge on population structuring is a necessary prerequisite to management of H. pogonognathus populations. The high population structure indicated that almost every single population contributes independently to the total gene pool and thus, proper conservation management strategies involving multiple aspects are needed to sustain the richness of H. pogonognathus gene pool. Besides, the paleo-drainage events during

Pleistocene are believed to be the main influence both in the gene flow and population structuring of H. pogonognathus. The genetic survey also revealed a pattern of declining H. pogonognathus population size. In addition, the study highlights the likely presence of new or cryptic species which is agreement with the initial hypothesis of new species discovery. However, a more integrated investigation framework needs to be carried out to resolve this issue.

173

CHAPTER 5

GENETIC DIVERSITY AND PHYLOGEOGRAPHY OF H. byssus AND H.

kuekenthali IN SARAWAK RIVER BASINS (NORTHWESTBORNEO)

5.1 Introduction

There are nine recognised Hemirhamphodon species presently known to occur in Borneo with seven being endemic and occupying a very restricted distribution. Thus, the centre of speciation for this genus is postulated to be in Borneo (de Bruyn, et al.,

2013; Tan & Lim, 2013). The two investigated species in the present study, H. byssus

(Tan & Lim, 2013) (Plate 5.1) and H. kuekenthali (Steindachner, 1901) (Plate 5.2) belong to the H. pogonognathus group (sensu Anderson & Collette, 1991) which have the following characters: anal fin of male with prominent posterior projection on 4th ray, having 12-17 dorsal-fin rays and 37-44 vertebrae, and absence of prominent red stripe on body (Tan & Lim, 2013). There are a few characters as described by Tan &

Lim (2013) which can be used to distinguish the H. byssus from its closest congener

H. kuekenthali. These characters include anterior half of dorsal fin with black streaks on the inter-radial membrane in H. byssus while H. kuekenthali has black streaks in the middle (Figure 5.1); anterior half of dorsal fin with filamentous fin rays reaching up to middle of caudal fin vs small extensions or none in H. kuekenthali. Worth noting is that only the large male individuals of H. byssus from the southern until the central region (Sri Aman) of Sarawak display distinct and pronounced filamentous dorsal-fin rays. For individuals of H. byssus from Sri Aman northwards to the region south of

174

Sibu (border of Rajang River), the filamentous fin rays are not as pronounced.

However, the dorsal fin features the diagnostic black streak.

Plate 5.1: Live (above) and preserved (below) colouration of male H. byssus specimen.

Plate 5.2: Live (above) and preserved (below) colouration of male H. kuekenthali specimen.

175

Figure 5.1: Morphology of dorsal fin. (A) H. byssus male, anterior half of dorsal fin with black streaks on the inter-radial membrane. (B) H. kuekenthali male, black streaks in the middle of dorsal fin. (Adapted from Tan & Lim, 2013)

The distribution of H. byssus is known to occur from the southern to the central region of Sarawak with Rajang river acting as its northern boundary (Figure 2.8, in

Chapter 2). It inhabits hill stream, swamp forest and peat swamp. On the other hand,

H. kuekenthali distributes from Sibu northwards until Limbang including Brunei. It also inhabits swamp forest, heath forest and peat swamp. The distribution of both species is apparently allopatric, illustrating an example of allopatric speciation between closely related or sister species which are separated by large river basins or mountain ranges. Similar distribution pattern has also been shown by a few allopatric species pairs occurring in Sarawak (in south to north orientation) including cyprinid fish Rasbora kalochroma and R. kottelati (Lim, 1995); fighting fish Betta ibanorum and B. akarensis (Tan & Ng, 2004; 2005).

176

The earlier DNA barcoding (COI gene) results presented in Chapter 3 indicated that H. byssus and H. kuekenthali sampled from the Sarawak river basins were genetically diverged. However, high levels of intraspecific divergence in H. byssus and H. kuekenthali were also detected. Closer examination revealed little genetic variation within population but substantial divergence between some populations. Esa, et al. (2006) and Nguyen, (2008) in separate genetic analyses of the mahseer, Tor douronensis populations in Sarawak using mtDNA sequences reported existence of significantly different clusters. Similar distribution pattern was also reported in the freshwater cyprinid, Hampala bimaculata from Sarawak (Ryan & Esa,

2006). However, the DNA barcoding data is insufficient for validating the species status of H. byssus and H. kuekenthali which remains unclear; whether each could be considered as a species complex. A robust phylogeographic analysis using multigene markers is required to infer finer-scale genetic diversity and population distribution/structure in order to elucidate accurate species status of both species. In addition, other information such as demographic history, gene flow with combination of geographical information can provide insights into factors that have influenced the contemporary structure of H. byssus and H. kuekenthali populations across the

Sarawak river basins.

Borneo is the third largest island in the world comprising of the Malaysian states of Sabah and Sarawak, Brunei and Kalimantan, Indonesia (Figure 5.2).

According to several studies (Yap, 2002; Sulaiman & Mayden, 2012; Kottelat, 2013),

Borneo is assigned into five biogeography sub-regions including north Borneo (Sabah), northwest Borneo (part of Sarawak and Brunei), west Borneo (most southern part of

Sarawak and Kapuas region of Kalimantan), south Borneo (central Kalimantan) and

177 east Borneo (east Kalimantan). It is drained by six major rivers including the Rajang and Baram rivers in Sarawak, and the Kapuas, Barito, Mahakam and Kayan rivers in

Kalimantan. The rivers in the east coast of Borneo (Mahakam and Kayan river) and all of those in north Borneo (in Sabah) are believed to have not been affected by the fluctuations in sea levels during the Pleistocene and presumably had maintained nearly the same topologies until present day (Inger & Chin, 1962). The Kapuas river on the other hand formed part of the north Sunda drainage. The evolution of many fish groups such as the Bagridae, Balitoridae, Cobitidae, Cyprinidae, Gyrinocheilidae,

Homalopteridae, Siluridae and Pangasiidae have been linked with the historical event of Sunda drainage as observed by Roberts (1989). However, there are no drainage connection between northwest Borneo (part of Sarawak) and the remaining Sundaland.

This disjunction is believed to have led to non-homogeneous distributions of freshwater fish species/clades between northwest Borneo and surrounding regions

(Sulaiman & Mayden, 2012).

178

Figure 5.2: Map showing the main rivers and boundaries within Borneo Island. The embedded map illustrates the five biogeographical sub-regions of Borneo. (Modified from Yap, 2002 and Tan, 2006).

Borneo is well recognised for its great floral and faunal diversity, and a joint declaration was made in 2007 by the three countries within this island to protect the pristine rainforest habitat known as the “Heart of Borneo.” (Sulaiman & Mayden,

2012). Furthermore, several studies have also reported that Sarawak harbours unique fish fauna compared with adjacent regions (Kottelat, et al., 1993; Doi, 1997). Much of the fauna and flora of Borneo are currently under assault from anthropogenic activities such as deforestation, plantation development, and major habitats have been changed

179

(Sodhi & Brook, 2006; Sodhi, et al., 2007; 2010). Thus, immediate international collaboration is needed to develop crucial conservation strategies for effective management and conservation of this region. In this context, understanding how populations are structured is an important consideration. The specialized habitat requirement coupled with the low dispersal abilities of these genus (Hemirahmphodon) particularly the investigated species are assumed to be in high risk of genetic diversity and population size decline due to the anthropogenic activities pressure. Therefore, knowledge on the current levels of genetic diversity and differentiation within and between populations particularly for both H. byssus and H. kuekentahli is needed for sustainable management strategies and their long term population sustainability.

Additionally, the information could be extended to other co-habiting organisms and being relatively sessile animals they could potentially be good indicators of environmental health.

The shaping of intraspecific genetic structure has been closely linked to the historical biogeography and the ability in detecting the history of past isolation by molecular markers has been well demonstrated (Dodson, et al., 1995; Avise, 2000;

Bernatchez, 2001; Volckaert, et al., 2002; van Houdt, et al., 2003). Several studies have shown that the mitochondrial DNA (mtDNA) cytochrome b (cyt b) gene is useful in elucidating evolutionary patterns in fishes (Rüber, et al., 2007; Li, et al., 2009;

Bohlen, et al., 2011; Lim, et al., 2014). In addition, the utility of single copy nuclear polymorphic (SCNP) nuclear DNA (nDNA) markers for addressing population- genetic questions has been clearly demonstrated in empirical studies, as reviewed by

(Hare, 2001) and also for Hemirhamphodon (de Bruyn, et al., 2013). Combination of

180 the mtDNA and SCNP markers are thus considered complementary as they reveal different aspects of a complex story at different evolutionary depths.

181

5.2 Objectives

This chapter describes the study on the characterization of genetic diversity patterns and phylogeography of H. byssus and H. kuekenthali populations across the

Sarawak river basins. Mitochondrial DNA cyt b gene and two nDNA markers (SCNP:

Hp5 and Hp54) were used to interpret the genetic diversity, levels of gene flow, historical population subdivision and demographic history for both species. This study also aimed to identify units for management and conservation in H. byssus and H. kuekenthali for long term population sustainability and species conservation.

182

5.3 Materials and methods

5.3.1 Sample collection, preservation and DNA extraction

Sampling locations are shown in Figure 5.3 and Table 5.1. Fifteen individuals were collected from each sampling site except only six individuals from Lanang and an individual of H. pogonognathus was employed as outgroup in the analyses.

Specimen preparation and DNA extraction of the individuals were as described in

Chapter 3.

Figure 5.3: Sampling locations of H. byssus and H. kuekenthali in contemporary geographical landmass, Sarawak (northwest Borneo). (Modified from Nguyen, et al., 2006b). For location abbreviation see Table 5.1.

183

Table 5.1: Sampling locations, sample size (n) and code of H. byssus and H. kuekenthali. The central Sarawak region for H. byssus starts from Sri Aman until south of Rajang river while for H. kuekenthali starts form north of Rajang river until Bintulu.

Present geographical Species n Location Code region H. byssus 1. 15 Kampung Semunin Cina SC southern Sarawak 2. 15 Sungai Stuum Toman SST southern Sarawak 3. 15 Sungai Duyoh SD southern Sarawak 4. 15 Tapang Rumput TR central Sarawak 5. 15 Sungai Paku SP central Sarawak H. kuekenthali 6. 15 Nangan Lassi LS central Sarawak 7. 6 Nangan Lanang LG central Sarawak 8. 15 Bukit Kemunyang BK central Sarawak 9. 15 Mukah MK central Sarawak 10. 15 Tatau TT central Sarawak 11. 15 Sungai Kemenda KMD northern Sarawak 12. 15 Sungai Liku LK northern Sarawak 13. 15 Long Lama LL northern Sarawak 14. 15 Sungai Kejin KJI northern Sarawak 15. 15 Labi-Linei LBI northern Sarawak 16. 15 Limbang LMB northern Sarawak

5.3.2 Gene amplification and sequencing

The same set of markers (mtDNA cyt b gene and two nDNA markers (SCNP:

Hp5 and Hp54)) used for H. pogonognathus as described in Chapter 4 were chosen for

H. byssus and H.kuekenthali as both studies addressed similar issues and data sets produced would be comparable.

PCR amplification for the mitochondrial cyt b gene region and two nDNA markers (SCNP: Hp5 and Hp54) on 15 individuals per site were conducted. Primer

184 sequences are listed in Appendix C and PCR condition is detailed in Appendix D. The

PCR products were electrophoresed on 1.5% agarose gels for band characterization and purified (PCR Clean-up System, Promega, Madison, WI, USA). Sequencing was conducted by a service provider (First Base Laboratories Sdn. Bhd. Malaysia) on an

ABI3730XL Genetic Analyzer (Applied Biosystems, Foster City, CA, USA).

5.3.3 Data analyses

5.3.3(a) Data sorting and haplotype generation

Similar data analyses as described in Chapter 4 were conducted as the same set of markers were used to address the same issues. Thus, only a brief explanation of data analyses is given below, and details can be referred in Chapter 4. Sequence editing and multiple sequence alignments were performed using MEGA v6.0 (Tamura, et al.,

2013). Haplotype generation was conducted using DNA Sequence Polymorphism

(dnaSP) v5.10.01 (Librado & Rozas, 2009). For nDNA markers, haplotypes were generated using PHASE v2.1.1 (Stephens, et al., 2001; Stephens & Scheet, 2005) and the results with a probability threshold of > 0.7 (Harrigan, et al., 2008) were accepted for further analyses.

185

5.3.3(b) Genetic diversity

The datasets were tested for optimal substitution models under the Bayesian

Information Criterion (BIC: Schwarz, 1978) using the ModelTest that is incorporated in MEGA v6.0. Saturation test was performed for each locus and the saturation plots

(the number of transition and transversion observed between pairs of sequences against p-distance) were drawn using DAMBE v5.6.9 (Xia & Xie, 2001) to avoid weak interpretation. ARLEQUIN v3.5 (Excoffier & Lischer, 2010) was used to calculate three estimations of diversity measurement to describe DNA variation in each population. These included nucleotide diversity (π) (Tajima, 1983), haplotype/gene diversity (h) (Nei, 1987) and theta S (ϴs) (Watterson, 1975).

5.3.3(c) Gene tree and haplotype network construction

Gene trees were constructed to assess the relationships among haplotypes through Neighbor-Joining (NJ) and Bayesian Inference (BI) methods using MEGA v6.0 and MrBayes v3.2.2 (Ronquist, et al., 2012) respectively. An individual of H. pogonognathus was employed as outgroup to root the tree. Minimum Spanning

Network (MSN) of haplotype for each gene were constructed using the maximum parsimony-median joining (MP-MJ) method as implemented in NETWORK Version

4.2.0.1 (Bandelt, et al., 1999) to explain the relationships among the haplotypes and spatial distribution.

186

5.3.3(d) Population structure

Analysis of molecular variance (AMOVA) was performed using ARLEQUIN v3.5 to test the significance of pairwise ФST and population structure. For hierarchical

AMOVA, the populations were partitioned according to the present geographical regions (southern, central and northern Sarawak) and the DNA lineages that were recovered in the phylogenetic analyses to infer the relative contribution of variances among groups within total (ФCT) and among populations within groups (ФSC).

Geographical distances were measured with Google Earth v7.1.7.2606 (2016).

Distances representing the shortest path between two populations follow the hypothetical Paleo-drainages as shown in Figure 5.4. Calculations were performed in

ARLEQUIN v3.5 with 10,000 permutations.

187

Figure 5.4: Map showing the hypothetical Paleo-drainages in Borneo during the low sea level around 20,000 BP and the sampling sites of H. byssus and H. kuekenthali. Number at the sampling sites follows the location sequence from top to bottom in Table 5.1. (Modified from Irwanto, 2015).

5.3.3(e) Historical demographic analysis

The population demographic history was conducted for each population through neutrality tests and mismatch distribution analysis. The Tajima’s D (Tajima,

1989) and Fu’s Fs (Fu, 1997) were used to check for deviation from neutrality through

ARLEQUIN v3.5 while the R2 test (Ramos-Onsins & Rozas, 2002) which allows detection of past population demographic changes was conducted using dnaSP

188 v5.10.01. The demographic historical expansion of each population was assessed with their mismatch distribution using the same program. Harpending’s raggedness index

(Hri) (Harpending, 1994) was also used to test against the null distribution of recent population expansion with ARLEQUIN v3.5. Pattern of historical demography inferred from estimates of effective population size over time was reconstructed with

Extended Bayesian Skyline Plots (EBSPs) (Heled & Drummond, 2008) method as implemented in BEAST v1.8.0 (Drummond, et al., 2012).

189

5.4 Results

5.4.1 Nucleotide composition and genetic diversity

In both species, no nuclear mitochondrial pseudogenes (numts) were detected after the protein coding regions (cyt b gene) were successfully translated. For nDNA markers (Hp5 and Hp54), not all individuals were satisfactorily amplified, and some individuals contained missing data and ‘noise’ due to indels and heterozygosity.

Sequences were trimmed in order to minimize missing data. The number of nuclear samples for those populations that consisted of heterozygous individuals was doubled following PHASE results. However, the actual nuclear sample sizes were reduced as the sequences with PHASE results that did not reach the probability threshold of > 0.7 were discarded.

A total of 74 and 155 sequences were successfully PCR amplified for the cyt b gene region from five populations of H. byssus and eleven populations of H. kuekenthali respectively as shown in Table 5.2. The cyt b region with 884 base pair

(bp) length after editing consisted of 134 bp of ND 6, 74 bp of t-RNA-Glu and 676 bp of cyt b gene (ND 6-tRNAglu-cyt b region). The average nucleotide composition was

A = 31%; T = 30%; C = 26% and G = 13% with 88 variable sites, of which 87 were parsimony informative sites. Meanwhile, 76 and 168 of Hp5 sequences were amplified respectively in both species (Table 5.2). The Hp5 sequences were trimmed to 261 bp with indel at position 49, containing mean nucleotide composition of A = 28%, T

=34%, C = 20% and G = 18% and revealed 11 variable sites, of which all 11 sites were parsimony informative. For Hp54 markers, 92 and 164 sequences were obtained respectively, had length of 243 bp with indel at position 219, mean nucleotide

190 composition of A = 37%, T =30%, C = 15% and G = 18%, and revealed 13 variable sites, of which 11 sites were parsimony informative.

Across the five populations of H. byssus, the nucleotide diversity (π) of cyt b was generally low ranging from 0% to 0.39% with three populations showing absence of diversity (π = 0%) and the TR population with highest value of 0.39% (Table 5.2).

Haplotype diversity (h) across populations ranged from 0 to 0.4725. As expected, TR population had the highest haplotype diversity while the second highest was in SD

(0.2571) with the rest of the populations having diversity equaled to zero. Three populations (SC, SST and SP) showed absence of both haplotype and nucleotide variation. However, in the combined total sample dataset, H. byssus showed high nucleotide and haplotype diversity at 4.21% and 0.8367 respectively.

For nDNA Hp5, nucleotide diversity was also generally low across most populations, ranging from 0% to 0.52% with SD recording the highest nucleotide diversity. Haplotype diversity, ranged from 0 to the highest value of 0.6377 which is also recorded in SD. Three populations (SC, TR and SP) showed absence of both haplotype and nucleotide variation. For Hp54, genetic variability was present in all populations. The nucleotide diversity ranged from 0.08% to 1.15% while haplotype diversity ranged from 0.1895 to 0.7000. The SC population had the highest nucleotide and haplotype diversities. Apart from SC population, the other populations had low nucleotide diversity and moderate haplotype diversity. In the total sample dataset, both nDNA markers exhibited low nucleotide diversity and high haplotype diversity. The saturation plots for all markers of H. byssus (Figure 5.5) shown no saturation of nucleotide substitution.

191

Table 5.2: Sampling location in abbreviation, no. of sequences (n), no. of haplotype (hp), no. of polymorphic sites (#V), nucleotide diversity (π), haplotype diversity (h) and expected heterozygosity per site based on number of segregating sites (ϴs).

Locus cyt b Hp5 Hp54 Location n hp #v π h ϴs n hp #v π h ϴs n hp #v π h ϴs H. byssus SC 15 1 0 0.0000 0.0000 0.0000 10 1 0 0.0000 0.0000 0.0000 16 5 7 0.0115 0.7000 1.8082 SST 15 1 0 0.0000 0.0000 0.0000 22 2 1 0.0007 0.1732 0.2743 20 2 1 0.0008 0.1895 0.2819 SD 15 3 2 0.0003 0.2571 0.6151 24 3 4 0.0052 0.6377 0.8034 10 2 1 0.0023 0.5556 0.3535 TR 14 3 10 0.0039 0.4725 3.1445 10 1 0 0.0000 0.0000 0.0000 20 3 2 0.0012 0.2789 0.5637 SP 15 1 0 0.0000 0.0000 0.0000 10 1 0 0.0000 0.0000 0.0000 26 3 5 0.0025 0.4800 1.0482 Total 74 @9 @88 0.0421 0.8367 18.0531 76 @7 @11 0.014 0.7804 2.244 92 @14 @13 0.0170 0.8743 2.5522 H. kuekenthali LS 15 2 2 0.0008 0.3429 0.6151 20 2 1 0.0019 0.4421 0.2819 16 2 1 0.0016 0.4000 0.3014 LG 6 1 0 0.0000 0.0000 0.0000 12 2 1 0.0007 0.1667 0.3311 2 1 0 0.0000 0.0000 0.0000 BK 15 1 0 0.0000 0.0000 0.0000 20 2 1 0.0022 0.5211 0.2818 18 2 1 0.0005 0.1111 0.2907 MK 15 1 0 0.0000 0.0000 0.0000 14 1 0 0.0000 0.0000 0.0000 28 6 6 0.0038 0.8069 0.7709 TT 15 4 3 0.0012 0.7714 0.9226 18 2 1 0.0005 0.1111 0.2907 6 1 0 0.0000 0.0000 0.0000 KMD 15 1 0 0.0000 0.0000 0.0000 10 1 0 0.0000 0.0000 0.0000 20 2 1 0.0011 0.2684 0.2818 LK 15 3 3 0.0011 0.4476 0.9226 18 1 0 0.0000 0.0000 0.0000 18 3 2 0.0039 0.5686 0.5815 LL 15 2 1 0.0003 0.2476 0.3075 20 2 1 0.0004 0.1000 0.2818 10 1 0 0.0000 0.0000 0.0000 KJI 15 3 2 0.0004 0.3619 0.6151 10 1 0 0.0000 0.0000 0.0000 10 1 0 0.0000 0.0000 0.0000 LBI 15 3 3 0.0012 0.4571 0.9226 18 4 2 0.0024 0.5294 0.5815 18 2 1 0.0009 0.2092 0.2907 LMB 14 2 1 0.0005 0.4396 0.3145 8 1 0 0.0000 0.0000 0.0000 18 3 3 0.0026 0.3922 0.8722 Total 155 @23 @166 0.0642 0.9384 29.5509 168 @13 @15 0.0736 0.8487 2.632 164 @15 @13 0.0120 0.8225 2.2911 @- actual total no. of haplotypes and no. of polymorphic sites as shown in Tables 5.3 to 5.6.

192

(a) cyt b

(b) Hp5

Figure 5.5: Nucleotide substitution saturation analysis for (a) cyt b, (b) Hp5 and (c) Hp54 of H. byssus. Transitions (s) and transversions (v) plotted against p distance.

193

For H. kuekenthali, nucleotide diversity across the eleven populations in cyt b was low ranging from 0% to 0.12% with four of the populations (LG, BK, MK and KMD) showing absence of diversity (π = 0%) as shown in Table 5.2. Haplotype diversity across populations ranged from 0 to 0.7714. The TT population had the highest haplotype diversity while the other populations had zero to moderate haplotype diversity. Four populations (LG, BK, MK and KMD) showed absence of both haplotype and nucleotide variations. Although most of the intrapopulation nucleotide diversity was zero, the total sample dataset on the other hand showed high nucleotide diversity of 6.42%. High haplotype diversity was also detected in the total sample dataset with value of 0.9384.

The nuclear Hp5 marker also showed generally low intrapopulation nucleotide diversity, ranging from 0% to 0.24% with three populations (LS, BK and LBI) recording the highest nucleotide diversity, three populations (LG, TT and LL) with around 0.05% and 0% for the rest. Haplotype diversity, ranged from 0 to the highest value of 0.5294 recorded in LBI with most of the populations showing low or zero haplotype diversity.

Five populations (MK, KMD, LK, KJI and LMB) showed absence of both haplotype and nucleotide variation. For Hp54, nucleotide diversity ranged from 0% to 0.39% with LK having the highest diversity and four of the populations (LG, TT, LL and KJI) had zero nucleotide variation. The haplotype diversity ranged from 0 to 0.8069 with the MK populations having the highest haplotype diversity and the other populations having zero to moderately high diversity. Four populations (LG, TT, LL and KJI) showed absence of both haplotype and nucleotide variations. In the total sample dataset, both nDNA markers

(Hp5 and Hp54) for H. kuekenthali also exhibited low nucleotide diversity and high

194 haplotype diversity. The saturation plots for all markers of H. kuekenthali (Figure 5.6) illustrated absence of nucleotide substitution.

(a) cyt b

(b) Hp5

Figure 5.6: Nucleotide substitution saturation analysis for (a) cyt b, (b) Hp5 and (c) Hp54 of H. kuekenthali. Transitions (s) and transversions (v) plotted against p distance.

195

5.4.2 Haplotype distribution

The generated haplotypes and its distribution across five H. byssus populations are shown in Table 5.3 and Table 5.4 (haplotype frequency in Appendix I). A total of 9, 7 and 14 haplotypes were revealed from cyt b, Hp5 and Hp54 sequences respectively. The cyt b haplotypes consisted of high number of variable sites (91) compared to Hp5 and

HP54 with only 14 and 17 variable sites. Haplotype Hbb02 varied from the reference sequence of Hbb01 at 3 sites while Hbb06 to Hbb08 from TR population seem to be the most varied among the cyt b haplotypes from Hbb01. As shown in Table 5.4, all haplotypes of cyt b were unique to their own population with no haplotype shared among populations. Three populations (SC, SST and SP) were monomorphic for a single haplotype while both SD and TR populations harboured three haplotypes each.

For Hp5 of H. byssus, haplotypes Hb502 varied at only single site from the reference Hb501 haplotype while the other haplotypes were varied at 3 to 8 sites. Most of the haplotypes were unique to their own population with only haplotype Hb501 shared by

SC and SST populations. Haplotypes Hp54 varied at 3 to 6 sites from the reference

Hp5401. Five out of 14 haplotypes (Hb5401 to Hb5405) belonged to SC population where the variation sites/pattern seem to be different from other haplotypes (Table 5.3(c)). Only haplotype Hbb5408 was shared by two populations (SD and SP) while other haplotypes were unique to specific population.

196

Table 5.3: The generated haplotypes of three DNA markers of H. byssus. (a) Nine haplotypes with 91 variable sites of 883 nucleotide bases of cyt b. (b) Seven haplotypes with 14 variable sites of 261 nucleotide bases of Hp5 and (c) fourteen haplotypes with 17 variable sites of 243 nucleotide bases of Hp54.

(a) cyt b

(b) Hp5

197

Table 5.4: Haplotype distribution of cyt b, Hp5 and Hp54 across five H. byssus populations.

Region southern Sarawak central Sarawak Locus Haplotype SC SST SD TR SP Total cyt b N 15 15 15 14 15 74 Hbb01 15 15 Hbb02 15 15 Hbb03 13 13 Hbb04 1 1 Hbb05 1 1 Hbb06 3 3 Hbb07 10 10 Hbb08 1 1 Hbb09 15 15

hp5 n 10 22 24 10 10 76 Hb501 10 20 30 Hb502 2 2 Hb503 12 12 Hb504 4 4 Hb505 8 8 Hb506 10 10 Hb507 10 10

hp54 n 16 20 10 20 26 92 Hb5401 7 7 Hb5402 6 6 Hb5403 1 1 Hb5404 1 1 Hb5405 1 1 Hb5406 18 18 Hb5407 2 2 Hb5408 5 2 7 Hb5409 5 5 Hb5410 17 17 Hb5411 2 2 Hb5412 1 1 Hb5413 6 6 Hb5414 18 18 n = no. of individuals.

198

The generated haplotypes and its distribution for H. kuekenthali across eleven populations are shown in Table 5.5 and Table 5.6 (haplotype frequency in Appendix J). A total of 23, 13 and 15 haplotypes were revealed from cyt b, Hp5 and Hp54 sequences respectively. Similar to H. byssus, the cyt b haplotypes are composed of relatively high number of variable sites (166) compared to Hp5 and HP54 with only 64 and 16 variable sites respectively. Five different patterns of variation can be observed in Table 5.5(a).

Haplotypes Hkb02 varied from reference haplotype Hkb01 of cyt b at only a single site, as they are from same population (LS). Haplotypes (Hkb04 to Hkb09) from central

Sarawak (BK, MK and TT) had nearly the same variable sites while the haplotypes

(Hkb11 to Hkb21) from northern Sarawak (LK, LL, KJI and LBI) were also similar. On the other hand, haplotype (Hkb10) from population KMD, located between the above mentioned regions/groups had its own variation pattern/sites. In addition, haplotypes

(Hkb01 and Hkb02) from the central Sarawak populations (LS and LG) as well as those haplotypes (Hkb22 and Hkb23) from the most northern population (LMB) also had their own variation pattern/sites. As shown in Table 5.6, all haplotypes of cyt b were unique to their own population with no haplotype shared among the populations. Four populations

(LG, BK, MK and KMD) were monomorphic while the other populations consisted of two to four private haplotypes where the TT population recorded the highest number of haplotypes.

199

Table 5.5: The generated haplotypes of three DNA markers of H. kuekenthali. (a) Twenty-three haplotypes with 166 variable sites of 883 nucleotide bases of cyt b. (b) Thirteen haplotypes with 64 variable sites of 261 nucleotide bases of Hp5 and (c) Fifteen haplotypes with 16 variable sites of 243 nucleotide bases of Hp54.

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 4 (a) cyt b1 1 2 2 2 3 3 4 6 7 8 9 9 9 0 1 1 2 2 2 2 2 2 4 4 4 4 4 4 5 6 6 6 6 7 8 8 8 8 0 0 2 3 4 4 4 5 5 5 6 7 7 8 8 8 0 1 1 1 1 2 3 3 4 6 6 7 8 9 9 0 0 1 1 1 2 3 3 4 5 5 6 8 1 2 1 7 9 5 9 5 9 2 7 3 5 8 7 5 6 0 1 2 3 4 9 0 1 4 5 8 9 8 2 4 7 8 0 0 1 6 7 3 5 6 7 6 7 9 2 5 6 7 6 9 2 5 8 9 2 3 6 8 4 0 9 5 0 9 8 1 0 6 5 6 1 4 7 6 0 5 1 0 6 5 6 Hkb01 T A C T C C T T G T T T G C A A C A G T A G A C C T T G T A C C G C T G T T T C T T C T G C T T G C A C C C C C A T C A A C A A T T T T G C C G C T C T T T C T C T T Hkb02 ...... C ...... Hkb03 ...... A ...... C . A . . . T . . . A ...... C ...... C . . . Hkb04 . G T . . . C C A . . C ...... C ...... C A . . . . . C ...... A C . T T ...... C . . . Hkb05 . G T . . . C C A . . C . . . G ...... A ...... C . . . A . . . . . C ...... T ...... A C ...... C . . . Hkb06 . G T . . . C C A . . C ...... T . C ...... A . . . . . C ...... T ...... G . . . A C A ...... C C . . . . . Hkb07 . G T . . . C C A . . C ...... T . C ...... A . . . . . C ...... T ...... G . . C A C A ...... C C . . . . . Hkb08 . . T . . . C C A . . C ...... T . C ...... A . . . . . C ...... T ...... G . . C A C A ...... C C . . . . . Hkb09 . G T . . . C C A . . C ...... T . C ...... A . . . . . C ...... T ...... G . . C A C A ...... C C . . . . . Hkb10 . . T C A T C C . . C . A . . . . T A . . A ...... T . A . . . . . C . . . T T . . T ...... G T . . A C A . . . . C . C . . T . . C . Hkb11 ...... C A . . . A . . . T C A C . A . . . C . . C G T . A T . . . . C T C A T A . T C C . T . T . T T . C . . G T . . . C C A C A . . A . . T . . . T C T C C Hkb12 ...... C A . . . A . . . T C A C . A . . . C C . C G T . A T . . . . C T C A T A . T C C . . . T . T T . C . . G T . . . C C A C A . . A . . T . . . T C T C C Hkb13 ...... C A . . . A . . . T C A C . A . . . C C . C G T . A T . . . . C T C A T A A T C C . . . T . T T . C . . G T . . . C C A C A . . A . . T . . . T C T C C Hkb14 ...... C A C . . A . . . T C A C . A . . . C C . C G . . A . C . . . C T C A T A . T C C . . G . T . . . C . . G T T . . C C A C A . . A . . T . . C T C T C C Hkb15 ...... C A C . . A . . . T C A C . A . . . C C . C G . . A . C . . . C T C A T A . T C C . . . . T . . . C . . G T T . . C C A C A . . A . . T . . C T C T C C Hkb16 ...... C A C . . A . . . T T A C . A . . . C C . C G . . A . C . . . C T C A T A . T C C . . G . T . . . C . . G T T . . C C A C A . . A . . T . . C T C T C C Hkb17 ...... C A C . . A . . . T C A C . A . . . C C . C G . . A . C . . . C T C A T A . T C C . . G . T . . . C . . G T T . . C C A C A . . A . . T . . C T C T C C Hkb18 ...... C A C . . A . . . T C A C . A . . . C C . C G . . A . C . . . C T C A T A . T C C . . G . T . . . C . . G T T . . C C A C A . . A . . T . . C T C T C C Hkb19 ...... C A C . . A . . . T C A C . A . . . C C . C G . . A . . . C . C T C A T A . T C C . . G . T . . . C . . G T T . . C C A C A . . A . . T . . C T C T C C Hkb20 ...... C A C . . A . G . T C A C . A . . . C C . C G . . A . . . C . C T C A T A . T C C . . G . T . . . C . . G T T . . C C A C A . . A . . T . . C T C T C C Hkb21 ...... C A C . . A . . . T C A C . A . . . C C . C G . . A . . . C . C T C A T A . T C C . . G . T . . . C . . G T T . . C C A C A . . A . . T . . C T C T C C Hkb22 C . . . . . C C A . . . A T . . T T A C G A G . T C ...... C T C A T C . T . C A . . T T T . . T C . . G . . . C . A C A . . A T . . . . . T C T C . Hkb23 C . . . . . C C A . . . A T . . T T A C G A G . T C ...... C T C A T C . T . C A . . T T T . . T C . . G . . . C . A C A . . A T . . . . . T C T C .

Table 5.5: continue... (a ) cyt b 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 9 9 9 9 9 9 0 0 1 1 1 1 2 2 2 3 4 4 5 6 6 6 7 7 8 9 0 0 1 1 2 2 4 4 5 5 6 6 6 7 7 8 8 8 9 1 1 3 3 3 4 5 5 6 6 6 7 7 8 8 8 8 9 9 0 0 1 2 2 2 3 3 3 4 4 4 4 5 6 6 6 7 9 2 3 4 5 8 9 1 7 0 1 6 9 2 5 8 4 6 9 6 1 4 7 0 3 8 4 3 6 2 5 4 7 5 8 1 4 0 6 9 2 3 1 4 7 9 4 5 0 2 5 1 3 7 0 5 8 1 5 0 1 6 7 5 8 1 4 9 2 5 8 1 4 8 2 3 6 9 1 1 4 7 9 Hkb01 T T C T T T A T A C C T C T A C C A C G C A C A T C G A T A C C C T T T T G C C T G C T A C C G C T C T C C C T T T G C C C T C T C T T C C C A A A G C A T T C C C C Hkb02 ...... C ...... Hkb03 C C T . . . . C ...... A ...... A T . . C . . . . Hkb04 C C T . . . . C . . . . T ...... T ...... A . . . A ...... C ...... T ...... A . . . C T . . . Hkb05 C C T . . C . C . T ...... T ...... C A . . . A ...... T ...... A . . . C T . . . Hkb06 . C T ...... T . . T . . . . . T ...... G . G ...... A . T . A ...... T ...... T . . T . A . . . C T . . . Hkb07 . C T ...... T . . T . . . . . T ...... A G . G ...... A . T . A ...... T ...... T . . T . A . . . C T . . . Hkb08 . C T ...... T . . T . . . . . T ...... G . G ...... A . T . A ...... T ...... T . . T . A . . . C T . . . Hkb09 . C T ...... T . . T . . . . . T ...... G . G ...... A . T . A ...... T ...... T . . T . A . . . C T . . . Hkb10 . C . . C . G . G T T . . A . . T . T A . . . G A . . . A . T . . . C C . A . T C A . A . . T . . A T C . . . . C C . . T . . . . T . . . T T . . G A . . C C T . . . Hkb11 . C . . . C . C . T T C . A C T T . T . . C . . C . A . C . . T . C C C . A T . . A T . C T . . T A T C . T T . . C . T T . C T C . . C T . . . . . A . . C C . . T . Hkb12 . C . . . C . C . T T C . A C T T . T . . C . . C . A . C . . T . C C C . A T . . A T . C T . . T A T C . T T . . C . T T . C T C . . C T . . . . . A . . C C . . T . Hkb13 . C . . . C . C . T T C . A C T T . T . . C . . C . A . C . . T . C C C . A T . . A T . C T . . T A T C . T T . . C . T T . C T C . . C T . . . . . A . . C C . . T . Hkb14 . C T . . C . C . . T C . A T . T G T A T C T . C T A . . . T T . C C C . A . . . A T . C T . . T A T C . T T . . . . T T . C T C . . C T . . G . . A . G C C . . T . Hkb15 . C T . . C . C . . T C . A T . T G T A T C T . C T A . . . T T . C C C . A . . . A T . C T . . T A T C . T T . . . . T T . C T C . . C T . . G . . A . G C C . . T . Hkb16 . C T . . C . C . . T C . A T . T G T . T C T . C T A . . . T T T C C C . A . . . A T . C T . . T A T C . T T . . . . T T . C T C . . C T . . G . . A . G C C . . T . Hkb17 . C T . . C . C . . T C . A T . T G T . T C T . C T A . . . T T T C C C . A . . . A T . C T . . T A T C . T T . . . . T T . C T C . . C T . . G . . A . G C C . . T . Hkb18 . C T . . C . C . . T C . A T . T G T . T C T . C T A . . . T T T C C C . A . . . A T . C T . A T A T C . T T . . . . T T . C T C . . C T . . G . . A . G C C . . T . Hkb19 . C T C . C . C . . T C . A T . T . T . T C T . C T A . . . T T . C C C . A . . . A T . C T . . T A T C . T T . . C . T T . C T . . . C T . . G . . A . G C C . . T . Hkb20 . C T . . C . C . . T C . A T . T . T . T C T . C T A . . . T T . C C C . A . . . A T . C T . . T A T C . T T . . C . T T . C T . . . C T . . G . . A . G C C . T T . Hkb21 . C T . . C . C . . T C . A T . T . T . T C T . C T A . . . T T . C C C . A . . . A T . C T . . T A T C . T T . . C . T T . C T . . . C T . . G . . A . G C C . . T . Hkb22 . C T . . . . C . T T . . A . T T . T . . T . . . . A . . . . T . C . C . A T T C A T . . T . . T A T C . . T . . C A T T . C . C . C C ...... A . . C C . . . T Hkb23 . C T . . . . C . T T . . A . T T . T . . T . . . . A . . . . T . C . C . A T T C A T . . T . . T A T C . . T . . C A T T . C . C . C C . . T . . . A . . C C . . . T

200

Table 5.5: continue...

(b) Hp5

201

Table 5.6: Haplotype distribution of cyt b, Hp5 and Hp54 across eleven H. kuekenthali populations.

Region central Sarawak northern Sarawak Locus Haplotype LS LG BK MK TT KMD LK LL KJI LBI LMB Total cyt b n 15 6 15 15 15 15 15 15 15 15 14 155 Hkb01 3 3 Hkb02 12 12 Hkb03 6 6 Hkb04 1 5 15 Hkb05 15 15 Hkb06 3 3 Hkb07 5 5 Hkb08 2 2 Hkb09 5 5 Hkb10 15 15 Hkb11 11 11 Hkb12 3 3 Hkb13 1 1 Hkb14 13 13 Hkb15 2 2 Hkb16 1 1 Hkb17 12 12 Hkb18 2 2 Hkb19 2 2 Hkb20 11 11 Hkb21 2 2 Hkb22 4 4 Hkb23 10 10

202

Table 5.6: continue…

Region central Sarawak northern Sarawak Locus Haplotype LS LG BK MK TT KMD LK LL KJI LBI LMB Total Hp5 n 20 12 20 14 18 10 18 20 10 18 8 168

Hk501 6 11 9 26 Hk502 14 14 Hk503 1 11 12 Hk504 14 14 Hk505 17 17 Hk506 1 1 Hk507 10 10 Hk508 18 19 10 4 51 Hk509 1 1 Hk510 1 1 Hk511 1 1 Hk512 12 12 Hk513 8 8 hp54 n 16 2 18 28 6 20 18 10 10 18 18 164

Hk5401 12 2 17 31 Hk5402 4 4 Hk5403 1 1 Hk5404 8 6 14 Hk5405 7 7 Hk5406 2 2 Hk5407 7 7 Hk5408 2 2 Hk5409 2 2 Hk5410 17 2 14 33 Hk5411 3 11 10 10 16 50 Hk5412 5 5 Hk5413 2 2 Hk5414 2 2 Hk5415 2 2

203

For Hp5 of H. kukenthali, three haplotypes (Hk501, Hk503 and Hk508) were shared by two to four populations. As shown in Table 5.6, haplotype Hk503 were shared by two adjacent populations in central Sarawak (LG and BK) while Hk508 was shared by four adjacent populations in northern Sarawak (LK, LL, KJI and LBI). Meanwhile, haplotypes Hk505 and Hk506 from TT population seem to be the most varied compared to other haplotypes as observed in Table 5.5(b). Interestingly, the adjacent populations in central Sarawak (LG, BK and MK) and adjacent populations in northern Sarawak (KMD,

LK, LL, KJI and LBI) also shared haplotypes generated from Hp54 marker. Each mentioned group shared haplotypes Hk5401 and Hk5411 respectively. Worth noting, haplotype sharing was occurred in all H. kuekenthali populations for Hp54 marker.

5.4.3 Phylogeography and evolutionary relationships among haplotypes

The constructed NJ cyt b trees of H. byssus and H. kuekenthali employed Tamura-

Nei with Gamma distribution (TN93+G) model while the BI cyt b tree was constructed with the best model across partitions (Table 5.7) following the results of the

PartitionFinder v1.1.1.

Table 5.7: Best model across partitions after PartitionFinder analysis for BI cyt b tree of H. byssus and H.kuekenthali.

Partition Best Model Subset Partitions Subset Sites 1 HKY+G nd_6, tglu 1-134, 135-208 2 K80+G cytb_1 209-884\3 3 F81 cytb_2 210-884\3 4 HKY+G cytb_3 211-884\3

204

Discrepancy in tree topologies between the constructed cyt b NJ and BI trees

(Figure 5.7) was observed. This might be attributed to the different algorithms and models employed in tree construction. The NJ tree revealed two major clades where surprisingly the haplotypes of H. kuekenthali from central Sarawak and the haplotypes of H. byssus formed one major clade while the haplotypes of H. kuekenthali from northern Sarawak formed another clade. On the other hand, the constructed BI tree showed high ability in resolving the interspecific relationship and also the intraspecific relationships with high

Bayesian posterior probabilities (pp). All haplotypes in BI tree were grouped according to the respective species and populations. The BI tree was preferable as it seems likely the plausible correct tree and it is also congruent with the COI tree in Chapter 3. Thus, the BI cyt b tree was employed for further explanation.

For H. byssus, three clades (Clade A to C) were revealed with Clade A, B and C consisting of haplotypes from populations of (SC and SST), SD and (SP and TR) respectively. The haplotypes Hbb03 to Hbb05 in southern Sarawak (Clade B) were clustered with the haplotypes from central Sarawak (Hbb06 to Hbb09) (Clade C), however, this relationship was only supported with moderate Bayesian pp. Meanwhile, in H. kuekenthali, the haplotypes were also diverged into three clades (Clade D to F). Most of the H. kuekenthali haplotypes were grouped into two major clades, Clade E and Clade F which paralleled the present geographical regions, central Sarawak and northern Sarawak respectively, however, with moderate Bayesian pp. Interestingly, the relationship between the haplotype Hbk10 of KMD population from central Sarawak with haplotypes of both

H. byssus and H. kuekenthali was unclear based on the constructed cyt b BI tree which presented polytomic relationship. Further subdivision within each clade with high

205

Bayesian pp was also observed where both clades were further bifurcated into two sub- regions/subpopulations. In Clade E of central region, LS and LG populations, Hkb01 to

Hkb03 formed one clade while BK, MK and TT populations (Hkb04 to Hkb09) formed anther clade. For Clade F, the most northern population of LMB (Hkb22 and Hkb23) formed its own clade while the rest of the populations formed another clade. Further examination at the shallowest level of the tree revealed that all of the haplotypes also formed clades according to their own populations.

206

Figure 5.7: Neighbor-Joining and Bayesian Inference trees among cyt b haplotypes of H. byssus and H. kuekenthali. Values at nodes represent bootstrap confidence level and Bayesian posterior probabilities. A single H. pogonognathus (BP01 OG) sequence was used as outgroup. The major clades were indicated by coloured vertical bars. The scale bar refers to genetic distance.

The constructed Hp5 NJ and BI trees (Figure 5.8) also showed different topologies which might be attributed to the different algorithms and models employed in tree construction. Both trees were unable to resolve interspecific relationship with the

207 generation of a single clade (Clade A) that consists of admixed haplotypes from both species and also polytomy relationship was shown in the BI tree. This clade (Clade A) was formed by the haplotypes (Hb501 and Hb502) of H. byssus populations from southern

Sarawak and a long branch clade of haplotypes (Hk505 and Hk506) together with haplotype Hk503 of H. kuekenthali populations from central Sarawak with high Bayesian pp but low bootstrap value. In addition, haplotypes Hb503, Hb504 and Hb505) formed a discrete clade while Hk508, Hk509 and Hk512 formed another one. However, there was no trend corresponding to geographical structuring in both trees.

Discrepancy of tree topologies was also observed between NJ and BI trees of Hp54

(Figure 5.9). The NJ tree compared to BI tree was better in resolving the interspecific relationship but however was not well supported. For the BI tree, the interspecific relationship was poorly resolved with occurrence of polytomy. From a population perspective, clades that formed in H. byssus corresponded to its own population except for haplotype Hb5401 from SC population which was grouped together with haplotypes from SST population. For H. kuekenthali, a few clades were formed which corresponded to the adjacent populations such as a clade that consisted of haplotypes from KMD, LK,

LL, KJI and LBI, and another clade with haplotypes from LS, LG and BK. Most of the clades were supported with high Bayesian pp but moderate bootstrap value. Comparison between two nDNA markers (Hp5 vs Hp54) revealed that the Hp54 marker was more efficient in resolving at the shallowest level (between populations) of relationships.

208

Figure 5.8: Neighbor-Joining and Bayesian Inference trees among Hp5 haplotypes of H. byssus and H. kuekenthali. Values at nodes represent bootstrap confidence level and Bayesian posterior probabilities. Clade (A) that consists of admixed haplotypes from both species was indicated by black vertical bar. A single H. pogonognathus (BP01 OG) sequence was used as outgroup. The scale bar refers to genetic distance.

209

Figure 5.9: Neighbor-Joining and Bayesian Inference trees among Hp54 haplotypes of H. byssus and H. kuekenthali. Values at nodes represent bootstrap confidence level and Bayesian posterior probabilities. A single H. pogonognathus (BP01 OG) sequence was used as outgroup. The scale bar represents 0.2%(NJ) and 5%(BI) of substitution divergence.

210

The evolutionary relationships among the generated haplotypes were also illustrated in the Minimum Spanning Network (MSN). In H. byssus, five clades were observed in MSN of cyt b which corresponded to each population (Figure 5.10(a)). No haplotype sharing occurred among the five populations where each haplotype was assigned into its own population. Other than SC and SST populations, the other three populations of SD from southern region of Sarawak, TR and SP (both from northern region) were not closely related with high divergence of 45 substitutions. Although the

SD population and the other two closely related populations (SC and SST) were located in southern region of Sarawak, however, the former diverged from the other two populations with 40 substitutions. In addition, the TR and SP populations which is in northern region were also not closely related with divergence of 27 substitutions.

Haplotype sharing on the other hand occurred for both nDNA markers. For Hp5, haplotype Hb501 was shared by the two closely related populations (SC and SST) while, the remaining haplotypes were assigned into its own population to form four clades, however, with low mutation sites. Interestingly, the divergence between the haplotypes

(Hb506 and Hb507) from northern populations (TR and SP) and the haplotype Hb501 of southern region populations (SC and SST) with 3 to 4 substitutions was lower compared to the divergence between the haplotypes (Hb503 and Hb505) of SD population and the haplotype Hb501 which is from the same region with 7 substitutions.

For Hp54, haplotype sharing occurred between SD and SP populations for haplotype Hb5408. Similar to the cyt b and Hp5, most of the haplotypes of Hp54 were assigned into its own population to form clades with low substitutions among each other.

211

The haplotype Hb5401 from SC appeared as tip haplotypes which may represent a descendant to the haplotype Hb5406 of SST population. Under coalescent theory, tip haplotypes is generally considered to be recently evolved (Castello & Templeton, 1994;

Crandall, 1996), where these new genetic variants have not had sufficient time to disperse widely across the distribution at large. Overall, the relationships among H. byssus populations (except SC and ST) did not appear to be closely related as there was little evidence of haplotype sharing among populations. In addition, both mtDNA and nDNA markers revealed that the SD population was not appropriate to be assigned into southern

Sarawak region as it was observed more diverged from the SC and SST populations compare to the SP and TR populations particularly in cyt b and Hp5. Moreover, the Hp54 revealed that the SD population was more closely related to central Sarawak populations as it shared haplotype with SP population.

212

Figure 5.10: Haplotype Minimum Spanning Network of (a) cyt b, (b) Hp5 and (c) Hp54 of H. byssus. Number in red on the connecting line indicated the number of substitutions separating haplotypes. The size of the circles is proportional to haplotype frequency. Small black dots are missing haplotypes linking the clades.

213

For H. kuekenthali, the constructed MSN (Figure 5.11) revealed that none of the cyt b haplotype was shared. On the other hand, haplotype sharing was observed in nDNA markers. The haplotypes (Hk501, Hk503 and Hk509) of Hp5, and (Hk5401, Hk5404,

Hk5410 and Hk5411) of Hp54 were shared by several populations. These haplotypes were also internally positioned where just a few mutational sites differentiated from most of the other haplotypes. Based on these characteristics, as in accordance to predictions from coalescence theory (Posada & Crandall, 2001; So, et al., 2006a), these haplotypes are most likely the ancestral variants based on central position and widespread distribution. In general, the constructed MSN of H. kuekenthali is able to demarcate the haplotypes into two regions which is similar to the a prior assigned geographical region (central and northern Sarawak). This was exhibited in both mtDNA and nDNA markers (cyt b and

Hp5) where the haplotypes in central region (LS, LG, BK, MK and TT) was diverged from the northern region (KMD, LK, LL, KJI, LBI and LMB) by around 50 and 22 substitutions respectively. In addition, the populations between these two regions namely

KMD represented by haplotype Hkb10 in cyt b and TT represented by haplotypes Hk506 and Hk506 in Hp5 were diverged from adjacent populations by around 56 and 34 substitutions respectively. Furthermore, both nDNA markers supported the assigning of the KMD population into northern Sarawak region even though the cyt b showed that the haplotype of KMD was linked with TT population from central Sarawak, however, with very high substitutions. For Hp54, although each haplotype was separated by only one to two substitutions, the haplotype sharing patterns (haplotypes shared between adjacent populations within region) still exhibited regional distribution of haplotype.

214

Figure 5.11: Minimum Spanning Network of (a) cyt b, (b) Hp5 and (c) Hp54 haplotypes of H. kuekenthali. Number in red on the connecting line indicated the number of substitutions separating haplotypes. The size of the circles is proportional to haplotype frequency. Small black dots are missing haplotypes linking the clades.

215

Figure 5.11: continue……

5.4.4 Population structure

In H. byssus, pairwise ФST comparisons of cyt b were highly significant (p < 0.05) after Bonferroni corrections (Table 5.8). The ФST comparisons plots (Figure 5.12(a)) also indicate that the H. byssus populations were highly structured. This is congruent with the haplotype distribution being population specific. For Hp5, pairwise ФST values and ФST comparisons plots (Figure 5.12(b)) also showed high significant value for most of the populations. However, SC and SST populations were not differentiated. The Hp54 marker also generated significant values, but with lower estimates than the other two markers by having moderate pairwise ФST values for most of the pairwise comparisons. The ФST comparison plots (Figure 5.12(c)) confirmed the structured population of H. byssus. In

216 general, the H. byssus populations were highly structured as supported by the haplotype distribution, constructed gene trees and pairwise ФST values, although a low degree of haplotype sharing occurred in the nDNA markers.

Table 5.8: Pairwise ФST values of cyt b, Hp5 and Hp54 (below diagonal) between H. byssus populations. Geographical distance (above diagonal in italics) between populations is based on the shortest path following the hypothetical Paleo-drainage systems (km).

Region southern Sarawak central Sarawak Locus Location SC SST SD TR SP cyt b SC 0.000 45.360 77.340 320.670 315.360 SST 1.000 0.000 32.980 280.860 270.780 SD 0.997 0.997 0.000 311.450 312.690 TR 0.972 0.973 0.966 0.000 260.630 SP 1.000 1.000 0.997 0.943 0.000 Hp5 SC 0.000 45.360 77.340 320.670 315.360 SST 0.000 0.000 32.980 280.860 270.780 SD 0.853 0.878 0.000 311.450 312.690 TR 1.000 0.943 0.785 0.000 260.630 SP 1.000 0.971 0.853 1.000 0.000 Hp54 SC 0.000 45.360 77.340 320.670 315.360 SST 0.574 0.000 32.980 280.860 270.780 SD 0.364 0.67467 0.000 311.450 312.690 TR 0.525 0.766 0.616 0.000 260.630 SP 0.423 0.652 0.472 0.612 0.000

Bold values indicate non-significant ФST values after Bonferroni corrections (p > 0.05).

217

Figure 5.12: Pairwise ФST comparisons of (a) cyt b, (b) Hp5 and (c) Hp54 (below diagonal) between H. byssus populations.

218

The hierarchical AMOVA (Table 5.9) examined in all hierarchical levels of both cyt b and Hp54 markers showed that most of the variance was attributed to differences among the DNA lineages (70.1% and 50.4%). However, these results were not supported by the Hp5 marker and also the non-significance of ФCT values in all markers. The hierarchical analyses revealed that molecular variance was attributed to differences among populations (ФST) and among populations within groups (ФSC), where both fixation indexes in all comparisons were significant. The high significant ФST and ФSC values coupled with non-significant ФCT values for both partitioning (geographical regions and

DNA lineages) suggested that the H. byssus populations are highly structured regardless of the partitioning groups. Worth noting is that the variation within population was relative low in cyt b at ~1.3% but moderate to high in Hp5 and Hp54 at ~10% and ~25% respectively which is probably attributed to heterozygous alleles due to existence of indels in nuclear markers.

The Mantel test of cyt b showed non-significant negative values (r = 0.405, p >

0.05) (Figure 5.13(a)) indicating non-association between genetic distance and geographical distance among the H. byssus populations. For nDNA markers, Hp5 exhibited significant values with r = 0.568, p < 0.05. On the other hand, Hp54 exhibited weak and non-significant values with r =0.256, p > 0.05. In general, the combination of highly significant ФST values and weak Mantel’s correlation test though significant in Hp5 marker implies that genetic differentiation is not necessarily associated to the geographical distance and ‘isolation by distance’ model is not suitable to describe the overall pattern of genetic differentiation of H. byssus at this spatial scale.

219

Table 5.9: AMOVA results for hierarchical genetic subdivision for percentage of variation and fixation indices (ФST, ФSC and ФCT) of cyt b, Hp5 and Hp54 of H. byssus populations.

Among Among populations Within groups within groups population Locus Groups ФST ФSC ФCT (%) (%) (%) cyt b Geographical 0.988 0.976 0.485 48.54 50.22 1.24 region cyt b lineages 0.986 0.955 0.701 70.12 28.51 1.37

Hp5 Geographical 0.905 0.858 0.329 32.90 57.60 9.50 region Hp5 lineages 0.904 0.830 0.434 43.35 47.01 9.64

Hp54 Geographical 0.757 0.701 0.187 18.71 56.97 24.32 region Hp54 lineages 0.760 0.515 0.504 50.44 25.53 24.03

Bold values indicate significant value (p < 0.05).

220

1.020

1.000 R² = 0.1642

ST ST 0.980 Ф 0.960

0.940 0 100 200 300 400 Geographical Distance (km)

(a) cyt b

1.200 1.000 0.800

ST 0.600 Ф 0.400 0.200 R² = 0.3235 0.000 0 100 200 300 400 Geographical Distance (km)

(b) Hp5

1.000 0.800

0.600 ST

Ф 0.400 R² = 0.0657 0.200 0.000 0 100 200 300 400 Geographical Distance (km)

Figure 5.13: The ФST plotted against the shortest path following the hypothetical Paleo- drainage systems (km) of H. byssus. (a) cyt b, (b) Hp5 and (c) Hp54 data.

221

Based on AMOVA, all ФST comparisons of cyt b in H. kuekenthali were highly significant (p < 0.05) after Bonferroni corrections (Table 5.10) and the ФST comparisons plots (Figure 5.14(a)) indicated that the H. kuekenthali populations are highly structured, supported by the lack of haplotype sharing among populations. Pairwise ФST values for

Hp5 also showed high significant value for almost all population pairwise comparisons.

However, three comparisons showed low and non-significant values involving adjacent populations (LG and BK), (LK, LL and KJI) which appeared to have haplotype sharing among populations. The ФST comparisons plots (Figure 5.14(a)) of Hp5 also clearly exhibited that most of the H. kuekenthali populations are highly structured. For Hp54, high and significant pairwise ФST values were presented in most of the populations. The adjacent populations such as (LS, LG and BK), (LK, LL, KJI and LBI) on the other hand showed low and non-significant values. The ФST comparisons plots (Figure 5.14(c)) are also in agreement with this population structure patterns. In addition, Figure 5.14(c) also highlighted that the central Sarawak populations (LS, LG and BK) have highly limited gene flow with the populations in northern Sarawak. In general, the H. kuekenthali populations seem highly structured as supported by the haplotype distribution, constructed gene trees and pairwise ФST values. However, nDNA makers showed some degree of gene flow amongst adjacent populations (LG and BK; LK, LL, KJI and LBI).

222

Table 5.10: Pairwise ФST values of cyt b, Hp5 and Hp54 (below diagonal) between H. kuekenthali populations. Geographical distances (above diagonal in italics) between populations based on the shortest path following the hypothetical Paleo-drainage systems (km).

Region central Sarawak northern Sarawak Locus Location LS LG BK MK TT KMD LK LL KJI LBI LMB cyt b LS 0.000 63.790 439.900 923.460 968.480 999.780 911.120 997.560 1008.640 854.350 1060.980 LG 0.963 0.000 379.900 857.990 905.470 946.870 853.500 931.270 948.900 783.230 1002.590 BK 0.987 1.000 0.000 740.870 700.860 888.020 831.740 919.560 934.860 778.830 964.250 MK 0.989 1.000 1.000 0.000 51.980 676.890 612.380 699.780 712.380 661.150 828.060 TT 0.975 0.977 0.979 0.981 0.000 720.740 666.920 749.090 761.280 700.790 880.910 KMD 0.995 1.000 1.000 1.000 0.991 0.000 440.540 511.270 537.690 457.570 630.180 LK 0.991 0.992 0.995 0.995 0.989 0.995 0.000 78.290 91.350 74.060 400.670 LL 0.995 0.998 0.999 0.999 0.993 0.999 0.978 0.000 15.380 155.180 479.700 KJI 0.995 0.997 0.998 0.998 0.993 0.998 0.976 0.866 0.000 170.880 499.120 LBI 0.991 0.991 0.994 0.994 0.989 0.994 0.963 0.914 0.907 0.000 400.000 LMB 0.993 0.996 0.997 0.997 0.991 0.997 0.985 0.994 0.993 0.98734 0.000 Hp5 LS 0.000 63.790 439.900 923.460 968.480 999.780 911.120 997.560 1008.640 854.350 1060.980 LG 0.582 0.000 379.900 857.990 905.470 946.870 853.500 931.270 948.900 783.230 1002.590 BK 0.615 0.329 0.000 740.870 700.860 888.020 831.740 919.560 934.860 778.830 964.250 MK 0.849 0.929 0.806 0.000 51.980 676.890 612.380 699.780 712.380 661.150 828.060 TT 0.971 0.984 0.961 0.992 0.000 720.740 666.920 749.090 761.280 700.790 880.910 KMD 0.831 0.916 0.783 1.000 0.992 0.000 440.540 511.270 537.690 457.570 630.180 LK 0.864 0.939 0.825 1.000 0.994 1.000 0.000 78.290 91.350 74.060 400.670 LL 0.848 0.894 0.809 0.974 0.990 0.967 0.000 0.000 15.380 155.180 479.700 KJI 0.831 0.916 0.784 1.000 0.992 1.000 0.000 0.000 0.000 170.880 499.120 LBI 0.781 0.760 0.747 0.876 0.968 0.847 0.619 0.595 0.547 0.000 400.000 LMB 0.819 0.907 0.770 1.000 0.991 1.000 1.000 0.965 1.000 0.836 0.000

223

Table 5.10: continue…

Locus Region central Sarawak northern Sarawak Location LS LG BK MK TT KMD LK LL KJI LBI LMB Hp54 LS 0.000 63.790 439.900 923.460 968.480 999.780 911.120 997.560 1008.640 854.350 1060.980 LG 0.000 0.000 379.900 857.990 905.470 946.870 853.500 931.270 948.900 783.230 1002.590 BK 0.172 0.000 0.000 740.870 700.860 888.020 831.740 919.560 934.860 778.830 964.250 MK 0.764 0.705 0.793 0.000 51.980 676.890 612.380 699.780 712.380 661.150 828.060 TT 0.911 1.000 0.972 0.178 0.000 720.740 666.920 749.090 761.280 700.790 880.910 KMD 0.865 0.885 0.913 0.324 0.825 0.000 440.540 511.270 537.690 457.570 630.180 LK 0.786 0.717 0.825 0.442 0.656 0.307 0.000 78.290 91.350 74.060 400.670 LL 0.925 1.000 0.977 0.652 1.000 0.796 0.230 0.000 15.380 155.180 479.700 KJI 0.925 1.000 0.977 0.652 1.000 0.796 0.230 0.000 0.000 170.880 499.120 LBI 0.912 0.938 0.950 0.671 0.925 0.750 0.270 0.012 0.012 0.000 400.000 LMB 0.801 0.761 0.847 0.283 0.669 0.077 0.363 0.710 0.710 0.711 0.000 Bold values indicate insignificant ФST values after Bonferroni corrections (p > 0.05).

224

Figure 5.14: Pairwise ФST comparisons of (a) cyt b, (b) Hp5 and (c) Hp54 (below diagonal) between H. kuekenthali populations.

225

Figure 5.14: continue…

The hierarchical AMOVA examined in all hierarchical levels for all three markers revealed significant genetic structure (Table 5.11). The high and significant ФCT values

(0.668 to 0.783) showed that most of the variance was attributed to differences among the

DNA lineages. For cyt b, variation among groups following geographical region partitioning also revealed highly significant ФCT values (0.600) which is nearly same with the ФCT values (0.668) following DNA lineages partitioning. This is likely due to the assignment of KMD being the only differentiated population between these two groups.

The high variation among groups (DNA lineages) and ФST values (0.800 to 0.995) with relative low variability within population (0.53% to 4.7%, except Hp54 with 20%)

226 suggested that the H. kuekenthali populations are highly structured between the populations and also between the DNA lineages.

Table 5.11: AMOVA results for hierarchical genetic subdivision for percentage of variation and fixation indices (ФST, ФSC and ФCT) of cyt b, Hp5 and Hp54 of H. kuekenthali populations.

Among Among populations Within groups within groups population Locus Group ФST ФSC ФCT (%) (%) (%) cyt b Geographical 0.995 0.987 0.600 60.01 39.48 0.51 region cyt b lineages 0.995 0.984 0.668 66.76 32.71 0.53 Hp5 Geographical 0.941 0.924 0.218 21.84 72.25 5.91 region Hp5 lineages 0.953 0.781 0.783 78.34 16.92 4.74 Hp54 Geographical 0.791 0.683 0.342 34.16 44.97 20.86 region Hp54 lineages 0.800 0.399 0.668 66.79 13.24 19.97 Bold values indicate significant value (p < 0.05).

The Mantel test results of all DNA markers (Figure 5.15) showed a signal of isolation by distance which exhibited significant correlation values of r = 0.5870, r = 0.558, r = 0.794 for cyt b, Hp5 and Hp54 respectively. These results indicated that there is correlation between pairwise ФST and geographical distance among the H. kuekenthali populations. Thus, ‘isolation by distance’ model of population differentiation at this spatial scale can be applied for H. kuekenthali.

227

1.050

1.000

ST 0.950 Ф R² = 0.3448 0.900

0.850 0 200 400 600 800 1000 1200 Geographical Distance (km)

(a) cyt b

1.200 1.000 0.800

ST 0.600 Ф 0.400 R² = 0.3115 0.200 0.000 0 200 400 600 800 1000 1200 Geographical Distance (km)

(b) Hp5

1.200 1.000 0.800

ST 0.600 Ф 0.400 R² = 0.6307 0.200 0.000 0 200 400 600 800 1000 1200 Geographical Distance (km)

Figure 5.15: The ФST plotted against the shortest path following the hypothetical Paleo- drainage systems (km) of H. kuekenthali. (a) cyt b, (b) Hp5 and (c) Hp54 data.

228

5.4.5 Population Genetic divergence

The cyt b pairwise distance for H. byssus and H. kuekenthali populations generated using Tamura-Nei (TN93)+G model is shown in Table 5.12. Pairwise distance within H. byssus populations ranged from 0% to 0.4% with the TR population having the highest value while the rest of populations possessed no variation (0%). The between population comparisons ranged from 0.3% (SC vs SST) to 7.4% (SST vs SP). The populations from central Sarawak (TR and SP) were highly diverged from the southern populations with distance of 5.6 % to 7.4%. Interestingly, the SD population from southern region was also highly diverged from the other two populations (SC and SST) of the same region with

4.8% to 5.2% genetic distance. Apart from the SD population, the divergences among adjacent populations were relatively low at only 0.3% to 0.4%.

For H. kuekenthali, within population divergence was generally low with highest divergence only reaching 0.1% and between populations divergence ranging from 0.3%

(KJI vs LL) to 13.2% (KMD vs KJI). As in H. byssus, high divergence corresponding to regional distribution was also exhibited in H. kuekenthali. The populations from central region was highly diverged from the northern region populations ranging from 7.4 % to

13.1%. Interestingly, the KMD population from northern region had lower divergence values with populations from central region (7.4% to 8.3%) compared to the populations from the same region (northern region) (9.5% to 13.2%). Additionally, the LMB population in the most northern part diverged from the other northern region population at 5.8% to 7.7%.

229

Table 5.12: Pairwise genetic distances with Tamura-Nei+G model of cyt b between H. byssus and H. kuekenthali populations.

Species H. byssus H. kukenthali Region southern central central northern Location SC SST SD TR SP LS LG BK MK TT KMD LK LL KJI LBI LMB SC 0.000 SST 0.003 0.000 SD 0.048 0.052 0.000 TR 0.072 0.073 0.063 0.004 SP 0.072 0.074 0.056 0.034 0.000 LS 0.086 0.091 0.076 0.100 0.101 0.001 LG 0.091 0.096 0.072 0.100 0.094 0.016 0.000 BK 0.087 0.092 0.084 0.104 0.100 0.032 0.025 0.000 MK 0.099 0.104 0.082 0.105 0.102 0.035 0.026 0.018 0.000 TT 0.090 0.095 0.071 0.107 0.099 0.042 0.041 0.030 0.033 0.001 KMD 0.087 0.089 0.083 0.115 0.116 0.083 0.088 0.081 0.082 0.074 0.000 LK 0.103 0.109 0.109 0.141 0.141 0.119 0.116 0.121 0.122 0.123 0.119 0.001 LL 0.117 0.124 0.108 0.148 0.139 0.125 0.119 0.119 0.124 0.130 0.129 0.034 0.000 KJI 0.119 0.125 0.109 0.150 0.140 0.126 0.120 0.120 0.125 0.131 0.132 0.034 0.003 0.000 LBI 0.108 0.114 0.103 0.144 0.135 0.126 0.119 0.120 0.124 0.130 0.128 0.034 0.009 0.009 0.001 LMB 0.104 0.107 0.094 0.126 0.129 0.107 0.101 0.102 0.106 0.107 0.095 0.058 0.077 0.077 0.077 0.001 OG 0.154 0.156 0.164 0.182 0.190 0.167 0.164 0.151 0.161 0.180 0.178 0.171 0.179 0.180 0.177 0.182 OG = Out group (H. pogonognathus)

230

The divergence between two species was ranged from 7.1% to 15.0%. Worth noting, the divergence between the H. kuekenthali populations from central region (LS,

LG, BK and MK) and H. byssus populations (7.1% to 11.6%, highlighted in red) was lower than the divergence between populations from central and northern region of H. kuekenthali (10.1% to 13.2%, highlighted in blue). This interesting population genetic divergence pattern is also illustrated in the constructed NJ cyt b gene tree in Figure 5.5(a).

Pairwise distance of Hp5 (Table 5.13) generated using Tamura-3-parameter

(TN92) model for both species revealed generally lower divergence values compared to cyt b gene ranging from 0% to 5.7% (TT and LBI). Within H. byssus populations, pairwise distance ranged from 0% to 0.6% with the SD population having the highest value.

Comparisons between H. byssus populations ranged from 0% to 2.6% (SST vs SD). Again, the SD population from southern region had high divergence with the other two populations (SC and SST) from the same region at 2.5% to 2.6%. Interestingly, these divergence values were even higher than the values for most of the interspecific divergence (H. byssus vs H. kuekenthali, ranged from 0% to 5.4% highlighted in red)

Similar to cyt b, the SC and SST populations were closely related with only 0.1% divergence.

231

Table 5.13: Pairwise genetic distances with Tamura-3-parameter model of Hp5 between H. byssus and H. kuekenthali populations.

Species H. byssus H. kuekenthali Region southern central central northern Location SC SST SD TR SP LS LG BK MK TT KMD LK LL KJI LBI LMB SC 0.000 SST 0.000 0.001 SD 0.025 0.026 0.006 TR 0.008 0.008 0.017 0.000 SP 0.016 0.016 0.025 0.016 0.000 LS 0.011 0.012 0.022 0.003 0.020 0.002 LG 0.008 0.009 0.019 0.000 0.017 0.003 0.001 BK 0.006 0.007 0.021 0.002 0.015 0.005 0.002 0.002 MK 0.013 0.013 0.023 0.004 0.021 0.007 0.005 0.007 0.000 TT 0.049 0.049 0.061 0.049 0.054 0.052 0.048 0.046 0.049 0.001 KMD 0.016 0.016 0.024 0.008 0.024 0.007 0.005 0.007 0.008 0.054 0.000 LK 0.016 0.016 0.025 0.008 0.024 0.007 0.005 0.007 0.008 0.054 0.008 0.000 LL 0.016 0.016 0.025 0.008 0.024 0.007 0.005 0.007 0.009 0.054 0.008 0.000 0.000 KJI 0.016 0.016 0.025 0.008 0.024 0.007 0.005 0.007 0.008 0.054 0.008 0.000 0.000 0.000 LBI 0.018 0.018 0.028 0.010 0.026 0.010 0.007 0.009 0.011 0.057 0.010 0.003 0.003 0.003 0.002 LMB 0.016 0.016 0.025 0.008 0.024 0.007 0.005 0.007 0.008 0.054 0.008 0.008 0.008 0.008 0.010 0.000 OG 0.048 0.048 0.058 0.039 0.056 0.042 0.039 0.041 0.043 0.091 0.048 0.048 0.048 0.048 0.047 0.048 OG = Outgroup (H. pogonognathus)

232

For H. kuekenthali, within population divergence was generally low with the highest divergence of 0.2% occurring in three populations. Between populations divergence ranging from 0% to 5.7% (TT vs LBI). As observed in cyt b, some populations of H. kuekenthali from central region had low interspecific divergence with H. byssus particularly TR population with divergence ranging from 0% (LG vs TR) to 0.4% (MK vs

TR). Interestingly, the TT population from the northern region had relatively much higher divergence values (around 5%) with populations from both species. This divergence value is even much higher than the interspecific divergence.

Pairwise population genetic distance (Table 5.14) of Hp54 generated using TN+G model for both species in general revealed a slightly higher values than Hp5 but still lower than cyt b ranged from 0% to 2.4%. Within H. byssus populations, genetic divergence ranged from 0.1% to 1.2% with the SC population having the highest value. Comparisons between H. byssus populations ranged from 0.5% to 2.4%. Surprisingly, the SC population had high within population divergence (1.2%). The SC population was also highly diverged from other populations with value ranged from 1.6% to 2.4% and even overlapped with the interspecific (H. byssus vs H. kuekenthali) divergence ranged from

1.0% to 2.4%. In addition, the SD population was closer to the other two populations (TR and SP) from central region than with its intra-region populations as observed in other two markers.

For H. kuekenthali, within population divergence was generally low ranging from

0% to 0.4%. Between populations divergence ranged from 0% to 1.4%. Interestingly, two populations of from central region (MK and TT) were closer to the northern region

233 population. The divergence value between these two populations and populations from the same region ranged from 1.2% to 1.4% while on the other hand lower values ranging from 0.4 to 0.9% were observed when compared to the northern region populations.

234

Table 5.14: Pairwise genetic distances with Tamura-3-parameter+G model of Hp54 between H. byssus and H. kuekenthali populations.

Species H. byssus H. kuekenthali Region southern central central northern Location SC SST SD TR SP LS LG BK MK TT KMD LK LL KJI LBI LMB SC 0.012 SST 0.016 0.001 SD 0.016 0.013 0.002 TR 0.021 0.005 0.009 0.001 SP 0.024 0.009 0.013 0.005 0.002 LS 0.020 0.023 0.018 0.019 0.022 0.002 LG 0.019 0.021 0.017 0.018 0.021 0.001 0.000 BK 0.019 0.022 0.017 0.018 0.021 0.001 0.000 0.000 MK 0.022 0.015 0.019 0.011 0.015 0.013 0.012 0.012 0.004 TT 0.023 0.017 0.021 0.013 0.017 0.014 0.013 0.013 0.003 0.000 KMD 0.020 0.014 0.018 0.010 0.014 0.010 0.009 0.009 0.004 0.005 0.001 LK 0.023 0.017 0.021 0.013 0.017 0.013 0.012 0.012 0.007 0.008 0.004 0.004 LL 0.023 0.017 0.021 0.013 0.017 0.014 0.013 0.013 0.007 0.008 0.004 0.003 0.000 KJI 0.023 0.017 0.021 0.013 0.017 0.014 0.013 0.013 0.007 0.008 0.004 0.003 0.000 0.000 LBI 0.024 0.018 0.022 0.014 0.018 0.014 0.013 0.013 0.008 0.009 0.004 0.003 0.000 0.000 0.001 LMB 0.020 0.014 0.018 0.011 0.014 0.011 0.010 0.010 0.005 0.006 0.002 0.005 0.006 0.006 0.006 0.003 OG 0.037 0.026 0.038 0.031 0.034 0.040 0.038 0.039 0.032 0.034 0.031 0.034 0.034 0.034 0.035 0.031 OG = Outgroup (H. pogonognathus)

235

5.4.6 Population history and demographic changes

Neutrality tests (Tajima’s D and Fu’s Fs) for population size changes of H. byssus populations based on three markers in generally exhibited no deviation from mutation- drift and gene flow-drift equilibrium, i.e. no evidence for recent population expansion where almost all of the tested values are non-significant (Table 5.15). Two populations

(SC and SP) could not be tested for deviation from mutation drift in cyt b and Hp5 markers as no gene/haplotype variation was observed. In cyt b, the SD population exhibited small significant negative for both tests while TR population showed small non-significant positive values. For Hp5, SST population exhibited low negative values while SD population exhibited low positive values for both tests, however, all are non-significant.

Two populations (SC and SD) and three populations (SST, TR and SP) showed non- significant low negative and positive values respectively in Hp54. For total sample dataset of H. byssus, cyt b exhibited large positive values in Fu’s Fs test while other markers had either low negative or positive values on both tests with all of the values being non- significant.

236

Table 5.15: Summary of population neutrality tests and demographic analyses based on Tajima’s D, Fu’s Fs, Rasmos-Onsins & Rozas (R2) and Harpending’s raggedness index (Hri) for H. byssus and H. kuekenthali populations.

Locus cyt b Hp5 Hp54

Population Tajima’s D Fu’s Fs R2 Hri Tajima’s D Fu’s Fs R2 Hri Tajima’s D Fu’s Fs R2 Hri

H. byssus SC 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.783 1.671 0.222 0.235 SST 0.000 0.000 0.000 0.000 -0.641 -0.176 0.087 0.457 -0.592 -0.097 0.095 0.422 SD -1.491 -1.546 0.170 0.302 1.659 2.790 0.184 0.622 1.464 1.096 0.278 0.321 TR 0.325 4.457 0.165 0.384 0.000 0.000 0.000 0.000 -1.141 -1.206 0.119 0.272 SP 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 -1.166 1.379 0.074 0.165 Total 2.986 38.711 0.195 0.010 1.553 3.580 0.160 0.113 0.056 -1.058 0.099 0.013

H. kuekenthali LS 0.302 1.795 0.171 0.667 1.026 1.169 0.221 0.209 0.650 0.872 0.200 0.200 LG 0.000 0.000 0.000 0.000 -1.141 -0.476 0.276 0.472 0.000 0.000 0.000 0.000 BK 0.000 0.000 0.000 0.000 1.531 1.467 0.261 0.273 -1.165 -0.794 0.229 0.617 MK 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.425 0.457 0.151 0.092 TT 0.464 -0.194 0.178 0.181 -1.165 -0.794 0.229 0.617 0.000 0.000 0.000 0.000 KMD 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 -0.086 0.381 0.134 0.287 LK 0.157 0.865 0.163 0.411 0.000 0.000 0.000 0.000 1.455 0.949 0.232 0.201 LL -0.399 0.133 0.124 0.316 -1.164 -0.879 0.218 0.650 0.000 0.000 0.000 0.000 KJI -1.002 -0.918 0.139 0.192 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 LBI 0.526 1.109 0.181 0.328 0.220 -1.139 0.159 0.116 -0.529 -0.011 0.105 0.382 LMB 0.842 0.944 0.220 0.208 0.000 0.000 0.000 0.000 -0.778 0.141 0.105 0.206 Total 2.672 51.238 0.170 0.021 0.170 0.479 0.093 0.088 -0.721 -2.359 0.063 0.042

237

The mismatch distribution of R2 based on three markers for H. byssus also failed to detect significant population expansion as all values were non-significant except for two populations (SST in Hp5 and SP in Hp54). Although, significant values were detected in these two populations, overall the signal of population expansion was weak. In contrast, the Hri results indicated lack of support for a stationary population, however, it did not preclude population expansion. Further inspection with mismatch distribution of the three markers showed absence of unimodal distribution. Mismatch distribution for most of the populations (Figure 5.16) were observed to have undergone rapid population reduction towards its new equilibrium. For total sample dataset, multimodal was detected in cyt b and Hp5 while unimodal in HP54. Nevertheless, all three markers also suggested rapid population reduction towards its new equilibrium. Generally, the neutrality tests and mismatch distribution showed no indication of population expansion in H. byssus populations except for Hp54. In contrast, these results suggested rapid population reduction (bottleneck) toward its new equilibrium.

238

(a) cyt b SD TR

(b) Hp5 SST SD

Figure 5.16: Mismatch distribution of H. byssus populations based on three markers (cyt b, Hp5 and Hp54) showing the expected and observed pairwise differences between the sequences with the respective frequency under constant population size. The solid lines represent the expected distribution and the dotted lines represent the observed distribution. The dotted line shows that the left edge of distribution converges rapidly toward the new equilibrium in most populations.

239

Total samples cyt b Hp5 Hp54

Figure 5.16: continue…

240

The EBSPs analysis (Figure 5.17) of cyt b for the total sample dataset of H. byssus suggested that the populations retained a constant effective population size, which however rapidly declined very recently. This result is congruent with rapid population size reduction as observed in the mismatch distribution. The existence of large positive values of Fu’s Fs test though non-significant also supported the population decline. However, this signal of population decline might probably be attributed to the large genetic divergence among populations for cyt b (Eytan & Hellberg, 2010). For the nDNA markers, a constant effective population size with recent decline of population as illustrated in Hp5 was observed. On the other hand, a slightly (not obvious) of recent population expansion was exhibited in Hp54. Overall, both markers (cyt b and Hp5) showed a recent occurrence of population reduction which is congruent with the mismatch distribution analysis though

Hp54 showed a weak signal of recent population expansion.

241

H. byssus cyt b H. byssus Hp5 12 1

10 0.8 8 0.6 6 0.4 4

2 0.2 Population sizescalar Population 0 sizescalar Population 0 0 0.5 1 1.5 2 2.5 3 3.5 0 0.5 1 1.5 2 Time (millions of years before present) Time (millions of years before present)

H. byssus Hp54 5

1 Population sizescalar Population 0 0 1 2 3 4 5 Time (millions of years before present)

Figure 5.17: Extended Bayesian Skyline Plots (EBSPs) showing the demographic history of H. byssus based on cyt b, Hp5 and Hp54. Solid blue line is the median effective population size, the dashed lines are the upper and lower 95% HPD for those estimates.

242

As in H. byssus, no evidence of recent population expansion occurred in H. kuekenthali populations as Tajima’s D, Fu’s FS and R2 (except LBI and LMB populations) values were non-significant in all three markers (Table 5.15). Several populations could not be tested for deviation from mutation drift as no gene/haplotype variation was observed. Small with non-significant negative and positive values of Tajima’s D and Fu’s

Fs tests were found across the populations in all three markers. The calculated Hri showed non-significant values which indicated lack of support for a stable population. Further inspection with mismatch distribution (Figure 5.18) showed that the H. kuekenthali populations had undergone rapid reduction towards its new equilibrium. For total sample dataset, multimodal was detected in cyt b and Hp5 while unimodal in HP54. Nevertheless, all three markers also suggested rapid population reduction towards its new equilibrium.

Generally, the neutrality tests and mismatch distribution showed no indication of population expansion in H. kuekentahli populations and suggested rapid population reduction toward its new equilibrium.

The EBSPs analysis (Figure 5.19) of cyt b for the total samples dataset of H. kuekenthali suggested that the population retained a constant effective population size, however, rapidly declined very recently. As explained in the result of H. byssus, signal of population decline might also be attributed to the large genetic divergence among populations which was congruent with the large positive values of Fu’s Fs test though it is non-significant. For the nDNA markers, a constant effective population size with recent decline of population as illustrated in Hp5 and a non-significant recent population expansion in Hp54. Overall, as observed in H. byssus, signal of recently population

243 reduction was found which is congruent with the mismatch distribution analysis with very weak signal of population expansion observed in Hp54 as is previous analysis.

244

(a) cyt b

LS LK LL

KJI LBI LMB

TT Figure 5.18: Mismatch distribution of H. kuekenthali populations based on three markers (cyt b, Hp5 and Hp54) showing the expected and observed pairwise differences between the sequences with the respective frequency under constant population size. The solid lines represent the expected distribution and the dotted lines represent the observed distribution. The dotted line shows that the left edge of distribution converges rapidly toward the new equilibrium in most populations.

245

(b) Hp5

LS LG BK

LL LBI

Figure 5.18: continue…

246

KMD LK LBI

LMB

Total samples cyt b Hp5 Hp54

Figure 5.18: continue…

247

H. kuekenthali cyt b H. kuekenthali Hp5 10 4 3.5 8 3 6 2.5 2 4 1.5 1 2

0.5 Population sizescalar Population Population sizescalar Population 0 0 0 1 2 3 4 5 0 2 4 6 8 10 Time (millions of years before present) Time (millions of years before present)

H. kuekenthali Hp54 10 8 6 4 2 0

Population sizescalar Population 0 1 2 3 4 5 6 7 Time (millions of years before present)

Figure 5.19: Extended Bayesian Skyline Plots (EBSPs) showing the demographic history of H. kuekenthali based on cyt b, Hp5 and Hp54. Solid blue line is the median effective population size, the dashed lines are the upper and lower 95% HPD for those estimates.

248

5.5 Discussion

5.5.1 Genetic Diversity

Understanding the genetic diversity and population genetic structure of the target species is a crucial step in developing effective management strategies and conservation. In general, low level of within population variation and high level of between populations variation were found in H. byssus and H. kuekenthali. This was reflected by the low genetic variation and low number of polymorphic loci as observed in most of the populations, however, with a high significant differences in most of the pairwise ФST comparisons. Similar pattern of genetic variation based on cyt b have also been reported in the freshwater cyprinids, H. sabana and H. bimaculata (Ryan &

Esa, 2006), and mahseer T. douronensis based on mtDNA16S rRNA (Nguyen, et al.,

2006b) which is also found in Sarawak. Additionally, low nucleotide and haplotype diversity within population but relatively high in total sample datasets were observed.

This occurrence implied that there was some degree of genetic variation among the populations which also support non-homogeneous distributions for both species.

The absence or low genetic variability within population as reflected by mtDNA and nDNA markers for both species may be explained by several factors.

Firstly, the population is from ‘a single female origin’ as mtDNA is maternally inherited (Avise, 1994). Secondly, recolonization as sea level rose after the last glaciations could have led to the founder effect (Kochzius & Nuryanto, 2008). Thirdly, is event of population bottleneck (Newman & Pilson, 1997) (and supported by the neutrality tests and historical demographic patterns) due habitat loss as the sea level rose after last glaciation or contemporary anthropogenic activities (Kochzius &

249

Nuryanto, 2008) including deforestation, pollution that have disturbed/destroyed their habitats and even overexploitation (Hauser, et al., 2002). However, as this forest halfbeak lack commercial value with limited demand in the ornamental industry, it is unlikely that it has been subjected to overexploitation. Fourthly, the actual genetic variation could have been underestimated due to the inadequate or limited number of samples. Nevertheless, moderate to high haplotype diversity particularly in nDNA markers were also observed in some populations of both species (SD and SC in H. byssus; TT, BK, LBI, MK and LK in H.kuekenthali) where this might probably be attributed to heterozygous alleles due to existence of indels.

5.5.2 Population structure and phylogeography

Based on the mtDNA cyt b analyses, three major lineages corresponding to regional distribution of H. byssus were revealed in phylogenetic analyses and constructed MSN. Populations from southern region were bifurcated into two clades where populations SC and SST formed one clade with SD in another clade, while populations from central region (TR and SP) formed its own clade. Similar but not exactly the same regional distribution was also reported for freshwater fishes in

Sarawak including mahseer species T. tombroides and T. douronensis (Nguyen, et al.,

2006b; Nguyen, et al., 2008; Esa, et al., 2008) and freshwater cyprinids H. bimaculata and H. macrolepidota (Ryan & Esa, 2006).

Although only three clades were formed out of five populations, significant population structure across H. byssus populations in Sarawak with overall high ФST

250 values were observed. Additionally, the hierarchical AMOVA also revealed that molecular variance was attributed to differences among populations (ФST) and among populations within groups (ФSC). The high ФST values between populations implied lack of contemporary migration or very low gene flow occurred between populations.

The observed population subdivision/substructure in the H. byssus suggested that the differentiation was most likely more related to habitat fragmentation or independence of river systems that separately enter the sea than regional demarcation. The limited dispersal of H. byssus populations could be attributed to the specialized habitat requirement (forest stream dependent) as observed in Hemirhamphodon species

(Collette, 2004) and thus leading to the high population structuring. It has also been well documented that non-migratory freshwater fishes tend to appear highly genetically structured (Dominguez-Dominguez, et al., 2008; Michel, et al., 2008;

Kamarudin & Esa, 2009). The presence of private haplotypes in all populations revealed by cyt b also provides evidence of high population structuring (So, et al.,

2006b). The occurrence of population-specific haplotypes in populations could be attributed to a recent bottleneck event subsequently leading to the loss of haplotype variety (Newman & Pilson, 1997) which was supported by the neutrality tests and historical demographic patterns. Failure in discovering additional haplotypes (if they actually exist) may also be due to insufficient sampling (Liao, et al., 2010). Based on the historical demographic findings, it is more likely due to bottleneck event(s).

On the other hand, some level of gene flow or migration between populations had occurred as indicated by haplotype sharing among H. byssus populations as observed in nDNA markers. Haplotype sharing between SC and SST population was

251 not surprising as both populations were connected via the same river system. However, the SD and SP populations which belonged to two different unconnected river systems also shared a haplotype, Hb5408. It is postulated that these two rivers had historical connection when the sea levels was low during the Pleistocene, which had huge influences in the dispersal and gene flow of many freshwater species in Sundaland including in Sarawak (McConnell, 2004). Based on the population divergence dating by de Bruyn, et al. 2013 (Figure 5.20), population subdivision of H. byssus was estimated to have occurred during the Pleistocene epoch, approximately 0.4 to 2.2 Mya.

Thus, it is suggested that gene flow might have occurred when both rivers were interconnected during the cyclical inter-glaciation periods and subsequently population subdivision occurred due to loss of connection as the sea level rose until present day level. This is also in broad agreement with previous population studies of mahseer, T. douronensis in Sarawak (Nguyen, et al., 2006b; Nguyen, et al., 2008) which reported that apparent genetic affinity was observed between populations that belonged to different river systems but have been historically interconnected. Many studies have reported that the up thrust of mountains, division of seas or retreat of glaciers have had great influence in shaping the phylogeographical pattern (Avise,

1994; Soltis, et al., 2006; Zemlak, et al., 2008). For aquatic taxa, several studies

(Hurwood, et al., 1998; Waters, et al., 2001; Poissant, et al., 2005) have revealed that contemporary population structure were likely derived from ancestral drainage configurations.

252

Figure 5.20: Divergence time of H. byssus and H. kuekenthali populations based on ultrametric Bayesian mitochondrial DNA trees of COI and control region variation. Values at nodes are median ages (in millions of years; bars=95% highest posterior densities). (Adopted and modified from de Bruyn, et al., 2013).

253

Population structuring was also observed in H. kuekenthali. The constructed cyt b tree revealed three H. kuekenthali clades with two major clades following the present geographical regions, central Sarawak and northern Sarawak respectively and a clade of single population located in between the two regions. This is supported by hierarchical AMOVA where partitioning followed either DNA lineages or geographical regions, both revealed nearly the same level of significantly high ФCT values. Other than that, high ФST value also suggested that the H. kuekenthali populations are highly structured which indicate very low gene flow between populations. Additionally, levels of gene flow among H. kuekenthali populations in

Sarawak seems to be related partially to the magnitude of geographical distance

(Paleo-river pathways) as the Mantel tests conformed to the classical expectations of

‘isolation by distance’ model (Wright, 1943; Kimura & H., 1964; Slatkin, 1993). High

ФST values between populations. As mentioned in H. byssus and H. pogonognathus

(in Chapter 4), this population structuring could be attributed to the limited dispersal ability of Hemirhamphodon species as specialized habitat is required (forest stream dependent) (Collette, 2004). This is also supported by the presence of private cyt b haplotypes in all populations. As explained, the occurrence of population-specific haplotypes in populations could also be attributed to a recent bottleneck event or insufficient sampling.

On the other hand, haplotype sharing among H. kuekenthali populations was also observed in nDNA markers indicating some level of gene flow between populations.

However, in contrast to H. byssus, the haplotype sharing in nDNA markers of H. kuekenthali occurred among adjacent populations in the central region (LS, LG and

254

BK), in northern region (LK, LL, KJI and LBI), and in between the two regions (MK and TT). This haplotype sharing pattern was not totally unexpected as these adjacent populations, although located in different contemporary river systems was believed to have been connected through the same Paleo-river system during the Pleistocene epoch. Based on the population divergence dating (Figure 5.20), population subdivision of H. kuekenthali was estimated to have occurred around 0.2 to 2.6 Mya which coincided with the glaciation events during Pleistocene. As explained previously, gene flow might have occurred during the cyclical inter-glaciation periods and subsequently genetic homogeneity with concomitant geographic separation as in present.

In general, Paleo-drainage rearrangement and connectivity between independent river systems have been influential in both restriction of gene flow and haplotype sharing. The mtDNA marker with 4X faster substitution rate and smaller effective population size (Avise 1987, 2009) compared to nDNA has greater sensitivity in detecting more recent population history where no haplotype sharing among populations was detected. In contrast, haplotype sharing appeared in nDNA markers coupled with a clade of admixed population haplotypes of both species and polytomic relationship was also revealed in the constructed nDNA gene trees. The characteristically low substitution rate and larger effective population size of nDNA make it more sensitive in detecting ancient population history with the relatively deep coalescent times (Harding, et al., 1997; Fu & Li, 1999). The recently separated populations which have insufficient time for genetic drift to produce new haplotypes is more likely attributed to haplotype sharing of nDNA (Printzen, et al., 2003).

255

5.5.3 Historical demography

Historical demographic analyses had revealed the pattern of past demographic events in H. byssus populations. The neutrality tests of Tajima’s D and Fu’s Fs of the three markers suggested no significant changes in population size, however, a signal of population reduction was detected by cyt b in total sample datasets where the Fu’s

Fs appear to be large positive values although non-significant. Mismatch distribution analysis of R2 was also revealed no demographic growth for all populations. On the other hand, the raggedness index did not support a stable population. Further analysis through the mismatch distribution plots also suggested rapid reduction in population size for most of the populations as well as total sample datasets. It is thus hypothesised that the demographic history of H. byssus had experienced rapid population reduction or population bottleneck in the recent past after a long period of stability. This is supported by the occurrence of population specific or private haplotype in all populations (Liao, et al., 2010). Other than having private haplotypes, the existence of high ФST values also did not support a population growth scenario (Esselstyn & Brown,

2009). Loss of habitat during the rise of sea level at the end of the last glacial might be a plausible factor for the demographic reduction. In addition, habitat destruction due to anthropogenic activities and specialized habitat requirement that limited dispersal could also have led to such situation. The Extended Bayesian Skyline Plots (EBSP) also suggseted the occurrence of population bottleneck in cyt b and Hp5 though Hp54 detected non-significant population expansion which is concordant with the hypothesis that in general the SE Asian freshwater taxa currently are believed to be in a refugial phase (Woodruff, 2010).

256

For H. kuekenthali, no significant changes in population size at all sample sites were also revealed by the neutrality tests of Tajima’s D and Fu’s Fs based on the three markers. As in H. byssus, a signal of population reduction was detected by cyt b in total sample dataset where the Fu’s Fs appeared to be large, positive values although non-significant. Apart from that, no demographic growth was revealed for all populations by mismatch distribution of R2. The raggedness index on the other hand did not indicate stationary population size. Examination of the mismatch distribution plots revealed that most of the population as well as the total sample dataset exhibited rapid population reduction towards its new equilibrium. The EBSP demonstrated the occurrence of a very recent population bottleneck event in cyt b while nDNA markers showed constant population size for a long period in Hp5 with non-significant population expansion in Hp54. Thus, based on the combination of all demographic history tests, the H. kuekenthali populations are postulated to have experienced rapid population reduction similar to H. byssus. Generally, historical demographic analyses in this study suggested rapid reduction of population size towards its new equilibrium for both species.

5.5.4 Taxonomic implications

The phylogenetic analysis and pairwise distance estimates raised two major concerns regarding the taxonomic status of H. byssus and H. kuekenthali. Firstly, the large pairwise differences among populations for both species should be noted. In H. byssus, three reciprocally monophyletic clades with correspondence to geographical region were observed in cyt b. This is however, less obvious in nDNA markers but still

257 traceable. These three clades in cyt b tree were supported by moderate to high bootstrap values and Bayesian pp. Interestingly, the populations from central region was highly diverged from other populations at ~7.2% while the SD population was diverged from the intra-region populations with values of ~5%. These values are relatively high compared to the divergence values among other populations of H. byssus which ranged from 0.3% to 3.4%. Moreover, the divergence value of populations from central region slightly overlapped with the interspecific divergence value between both species which ranged from 7.1% to 11.6%.

Intraspecific divergences using the same gene have been reported ranging from

0.1 to 1.9% for freshwater cyprinid H. macrolepidota in Sarawak. As observed in

Chapter 4, the divergence among populations of H. pogonognathus reached to 2.6% when the atypical populations (Kelantan and Main Clade) were excluded. In this study, the among population divergences ranged from 0.9% to 4.2% in H. kuekenthali when the atypical populations were also excluded. Additionally, limited or no gene flow occurred among the H. byssus populations as indicated by the presence of private haplotypes among the populations and the high ФST values. Based on the above observations, it is thus hypothesized that the high level of genetic divergence of H. byssus particularly the populations from central region could be a sign of crypticness.

The same suggestion was also reported for the freshwater cyprinid, H. bimaculata in

Sarawak where the two H. bimaculata lineages from different regions (southern and northern) were diverged at 9.2% based on cyt b (Ryan & Esa, 2006). In another population genetics study involving the mahseer, T. douronensis in Sarawak, the authors also postulated presence of two major lineages of T. douronensis, namely the

‘northeastern’ and the ‘southwestern’ (Nguyen, et al., 2006b; Nguyen, et al., 2008).

258

Similar patterning was also observed in H. kuekenthali with three reciprocally monophyletic clades produced which also corresponded to regional distribution, however, with moderate Bayesian pp. Surprisingly, one of the clade consists of populations from the central region which was more closely related to H. byssus. This was reflected by the pairwise distance of cyt b (highlighted in blue in Table 5.12) and the constructed NJ cyt b tree. As mentioned earlier, the H. byssus and H. kuekenthali were allopatrically diverged. According to the divergence time (Figure 5.20), these two species last coalesced at around 3.9 Mya. Thus, it is believed that there was insufficient time for the H. kuekenthali from the central region to accumulate high genetic divergence to form a reciprocally monophyletic compared to the other species which had earlier TMRCA (time to most recent common ancestor).

The second concern is the ambiguous status of some populations by having distinct divergence values as bolded in Table 5.12 to Table 5.14. These populations included SC and SD of H. byssus, MK, KMD and LMB of H. kuekenthali. This situation might be attributed to the individual population undergoing independent evolutionary process such as selection, genetic drift and fixation which may be influenced by specific climatic or environmental conditions, prolonged isolation of populations and demographic events such as repeated extinction and recolonization events by small founding populations (Nguyen, et al., 2006b). Furthermore, instability in regional grouping based on different markers was also observed for these populations. For instance, the SD population was closer to the southern region population based on cyt b, however, it was closer to the central populations when based on nDNA markers. The plausible explanation for this is likely attributed to the

259 coalescent stochasticity as different markers can have different substitution rates and also TMRCA (Eytan & Hellberg, 2010).

5.5.5 Conservation

Identification of population groups with independent evolutionary histories is essential for biodiversity management to prevent the further loss of biodiversity. The population structure detected in this study suggested that both H. byssus and H. kuekenthali consisted of three clusters and should be recognized as separate units for management as they represent different evolutionary distinct lineages (Avise, 1994;

Moritz, 1994; Fraser & Bernatchez, 2001). Furthermore, the historical demographic tests also showed that both species had experienced rapid population size reduction in the recent past. The increasing anthropogenic threats coupled with various natural processes may lead to the further decline and diminished population connectivy of this species. Thus, conservation management should be applied for each recognised management unit even among populations. Any further loss of populations might result in the risk of losing potential cryptic species.

5.6 Conclusion

In conclusion, the combination of both mtDNA and nDNA markers have provided a comprehensive (more complete) picture of the evolutionary processes and relationships within and between H. byssus and H. kuekenthali. The mtDNA is effective in lineage sorting with smaller effective population size coupled with high

260 mutation rate. On the other hand, the nuclear DNA with lower mutation rates has a better resolving power in assessment of genotype sharing that may have coalesced in more ancient time. This study also revealed that almost every single population contributed independently to the total gene pool as indicated by the high population structure. Additionally, the cyclical glaciation events during Pleistocene seemed to have influenced the population structuring of both species. Furthermore, a reappraisal of the species status of both species is needed with integrated investigation framework as the study highlights the likely presence of cryptic species. This is because accuracy of taxonomic status of targeted species is a crucial key in conservation strategy. In addition, the genetic survey also revealed a pattern of declining population size for both species. Based on the above findings, proper conservation management strategies involving multiple aspects are needed to sustain the richness of H. byssus and H. kuekenthali gene pools.

261

CHAPTER 6

GENERAL DISCUSSION AND CONCLUSIONS

This study detailed results on the genetic investigation of the genus

Hemirhamphodon and how its phylogeography and diversity have been shaped in the background of past geological and climatic changes and contemporary influences.

Generally, the degree of phylogeographical structuring of species within and among drainages can definitively determined when the barriers have been remained stable for a period of time. Populations that inhabit independent or historically isolated drainage basins are assumed to have high phylogenetic divergence (Stelbrink, et al., 2012). It is common for freshwater fish populations to be structured spatially in nature (Gyllensten,

1985; Avise, 1992; Ward, et al., 1994) as the freshwater environment always have various forms of barriers that restrict dispersal. These barriers include drainage divides, lakes, waterfalls, dams and rapids among others (Monaghan, et al., 2001; Yamamoto, et al., 2004; Crispo, et al., 2006). Information on the faunal and floral patterns or genetic variation are required in understanding the biotic evolution in a particular region (or single island) or just as importantly the related processes (event-based approaches) involved (Stelbrink, 2014).

This project highlights several important aspects of the taxonomy, phylogenetics and geographical distribution of the genus Hemirhamphodon within

Malaysia and neighbouring landmasses based on several molecular approaches and markers. The DNA barcode analysis was utilised to conduct a diversity assessment of the genus and proved to be useful in revealing potential cryptic diversity in the genus

Hemirhamphodon. Barcode gap analysis in the dataset revealed that no barcode gap

262 was present in H. pogonognathus, H. kuekenthali and H. byssus, which indicated the probable existence of more than one species within each of these taxa. The existence of high cryptic diversity was apparent from the results of the genetic distance and COI gene trees. It is proposed that the H. pogonognathus of the Kelantan group is potentially a new species record. The occurrence of cryptic diversity through DNA barcodes has been frequently observed in the tropics (Kadarusman, et al., 2012;

Puckridge, et al., 2013) due to the inherent high diversity of fauna in the tropics coupled with the taxonomic impediment and limited expertise (Hubert & Hanner,

2015).

In the second part of the project, a combination of both mtDNA and nDNA markers was utilised in assessing population genetics of the three Hemirhamphodon species. These markers effectively complemented each other, enabling a comprehensive (more complete) picture of the evolutionary processes and intraspecific relationships to be generated. The mtDNA is more effective in lineage sorting due to its smaller effective population size coupled with high mutation rate. On the other hand, the nuclear DNA with lower mutation rates has a better resolving power in assessment of genotype sharing that may have coalesced in more ancient time.

The combination of life-history characteristics and complex historical Paleo- drainage rearrangements was reflected in the population structuring of the three investigated species. Although the Paleo-drainage systems of Sundaland seemed to have limited influence in driving Hemirhamphodon speciation, the geological history such as the cyclical glaciation event during the Pleistocene epoch coupled with the

Paleo-drainage rearrangements have greatly influenced the genetic diversity of the

263 three Hemirhamphodon species populations. During the rise in sea level, the founder effect due to recolonization had led to the low genetic variation within the population.

Furthermore, loss of habitat as sea level rose could contribute to the population bottleneck, resulting in population-specific haplotypes (fixation) through genetic drift and subsequently loss of genetic variation. Additionally, rise in sea level created geographical barriers, isolating populations and further restricted gene flow among populations. Over a prolonged period, independent evolutionary process occurred in these isolated populations resulting in significant population structure across drainages as shown in the overall high ФST value. According to Hurwood, et al., (2008), population structure is commonly observed in freshwater aquatic taxa. On the other hand, when the sea level was lower, the populations located at adjacent drainage basins but representing different Paleo-drainages or river systems (H. byssus and H. kuekenthali) were believed to be re-connected. Consequently, this had promoted gene flow or migration to occur among these populations as haplotype sharing was observed.

This study also revealed through historical demographic analyses that most of the populations of the three Hemirhamphodon species had experienced rapid population size reduction. This parallels the hypothesis by Woodruff (2010) where in general the SE Asian freshwater taxa currently are in a refugial phase. Based on the pairwise distance and population structure results, most of the Hemirhamphodon populations should be treated as separate management unit in the context of conservation. Thus, conservation strategies and management plans should be implemented on each of the Hemirhamphodon populations for long term population sustainability and species conservation (Moritz, 2002). This is a challenge in conservation strategies especially for a widely distributed species such as H.

264 pogonognathus as this taxon might also exhibit non-uniform ecological environment in addition to the genetic differentiation over its range. Thus, different strategies for conservation of Hemirhamphodon diversity needs to be developed since a single conservation strategy may not be suitable for all populations. Moreover, large parts of tropical forest habitats in Peninsular Malaysia, Borneo and Sumatra are under serious threats of destruction due to rapid economic development (Sodhi, et al., 2004). These areas are the most biodiverse (Lambert & Collar, 2002) but are now being rapidly transformed into palm oil plantations or urban development. Therefore, greater awareness programmes should be initiated in addition to highlight to the relevant authorities to arrest the situation from worsening.

There are two basic and complementary conservation strategies, in-situ and ex- situ. In-situ conservation strategies could be implemented with declaration of more forest reserves, natural parks even biosphere reserves which may provide less expensive protection for the wild relatives than ex-situ measures. At the same time, this strategy also provides protection for the whole biodiversity in that area. The ex- situ conservation can be either focusing on individuals or populations through establishments of aquaria, or in vitro research such as cryopreserved of sperm, embryos and as any tissues containing DNA if necessary. Nevertheless, ex-situ conservation should be considered as complementary to their in-situ conservation.

However, in a critical situation such as limited native individuals availability or unavoidable threats, ex-situ conservation becomes the main approach to ensure their long-term conservation and sustainability.

265

Finally, a reappraisal of the species status of the three studied species (H. pogonognathus, H. byssus and H. kuekenthali) should be conducted with an integrated investigation framework as this study has highlighted the likely presence of cryptic species. This is because accuracy of taxonomic status of targeted species is a crucial key in strategising conservation policies. In addition, the need for conservation is urgent not only for these highly diverse freshwater taxa but also for the diverse floral and faunal of this most threatened and vulnerable region of the tropics (Sodhi, et al.,

2004; Woodruff, 2010).

266

REFERENCES

Adamson, E. A. S., 2010. Influence of historical landscapes, drainage evolution and ecologic traits of genetic diversity in Southeast Asian freshwater snakehead. PhD thesis, Queensland University of Technology.

Adamson, E. A. S., Hurwood, D. A., Baker, A. M. & Mather, P. B., 2009. Population subdivision in Siamese mud carp Henicorhynchus siamensis in the Mekong River basin: implications for management. Journal of Fish Biology , Volume 75, pp. 1371-1392.

Adamson, E. A. S., Hurwood, D. A. & Mather, P. B., 2012. Insights into historical drainage evolution based on the phylogeography of the chevron snakehead fish (Channa striata) in the Mekong Basin. Freshwater Biology, Volume 57, pp. 2211-2229.

Aljanabi, S. M. & Martinez, I., 1997. Universal and rapid salt-extraction of high quality genomic DNA for PCR-based techniques. Nucleic Acids Research, Volume 25, pp. 4692-4693.

Anderson, W. D. III & Collette, B. B., 1991. Revision of the freshwater viviparous halfbeaks of the genus Hemirhamphodon (Teleostei:Hemiramphidae). Ichthyological Explorations of Freshwaters, 2(2), pp. 151-176.

April, J., Hanner, R. H., Dion-Côté, A. M. & Bernatchez, L., 2013. Glacial cycles as an allopatric speciation pump in north-eastern American freshwater fishes. Molecular Ecology, Volume 22, pp. 409-422.

Avise, J. C., 1989. Gene trees and organismal histories: A phylogenetic approach to population biology. Evolution, Volume 43, pp. 1192-1208.

Avise, J. C., 1998. The history and purview of phylogeography: a personal reflection. Molecular Ecology , Volume 7, pp. 371-379.

Avise, J. C., 2009. Phylogeography: retrospect and prospect. Journal of Biogeography, Volume 36, pp. 3-15.

267

Avise, J. C., 1992. Molecular population structure and the biogeographic history of a regional fauna: a case history with lessons for conservation biology. Oikos, Volume 63, pp. 62-67.

Avise, J. C., 1994. Molecular markers, natural history and evolution. New York: Chapman and Hall .

Avise, J. C., 2000. Phylogeography. The history and formation of species. Cambridge, MA.: Harvard University Press.

Avise, J. C., Arnold, J., Ball, R. M., Bermingham, E., Lamb, T., Neigel, J. E., Reeb, C. A. & Saunders, N. C., 1987. Intraspecific Phylogeography: the mitochondrial DNA bridge between population genetics and systematics. Annual Review of Ecology and Systematics, Volume 18, pp. 489-522.

Bandelt, H., Forster, P. & Rohl, A., 1999. Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution 16: 37-48., Volume 16, pp. 37-48.

Barber, A. J. & Crow, M. J., 2009. The structure of Sumatra and its implications for the tectonic assembly of Southeast Asia and the destruction of Paleotethys. Island Arc, Volume 18, pp. 3-20.

Barber, A. J., Crow, M. J. & De Smet, M. E. M., 2005. Tectonic evolution. In: A. J. Barber, M. J. Crow & J. S. Milsom, eds. Sumatra: Geology, Resources and Tectonic Evolution. London: Geological Society, pp. 234-259.

Barlow, A., Baker, K., Hendry, C. R., Peppin, Phelps, T., Tolley, K. A., Wüster, C. E. & Wüster, W., 2013. Phylogeography of the widespread African puff adder (Bitis arietans) reveals multiple Pleistocene refugia in southern Africa. Molecular Ecology, Volume 22, pp. 1134-1157.

Becker, S., Hanner, R. & Steinke, D., 2011. Five years of FISH-BOL: brief status report. Mitochondrial DNA, 22(Suppl. 1), pp. 3-9.

Beck, S., Carvalho, G. R., Barlow, A., Rüber, L., Tan, H. H., Nugroho, E., Wowor, D., Azizah, M. N. S., Ng, P. K. L. & de Bruyn, M., 2013. Multi-locus genetic data

268

elucidates the phylogeography of the Southeast Asian freshwater killifish, Aplocheilus panchax. In possession of the authors, Bangor University.

Beheregaray, L. B., 2008. Twenty years of phylogeography: the state of the field and the challenges for the Southern Hemisphere. Molecular Ecology , Volume 17, pp. 3754-3774.

Bernatchez, L., 2001. The Evolutionary history of brown trout (Salmo trutta L.) inferred from phylogeographic, nested clade, and mismatch analyses of mitochondrial DNA variation. Evolution , Volume 55, pp. 351-379.

Bernatchez, L. & Wilson, C. C., 1998. Comparative phylogeography of Neartic and Paleartic fishes. Molecular Ecology, Volume 74, p. 431–452.

Bickford, D., Lohman, D. J., Sodhi, N. S., Ng, P. K. L., Meier, R., Winker, K., Ingram, K. K. & Das, I., 2007. Cryptic species as a window on diversity and conservation. Trends in Ecology and Evolution, Volume 22, pp. 148-155.

Bintanja, R. & van de Wal, R. S. W., 2008. North American ice-sheet dynamics and the onset of 100,000-year glacial cycles. Nature , Volume 454, pp. 869-872.

Blaxter, M. L., 2004. The promise of a DNA taxonomy. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, Volume 359, pp. 669-679.

Bleeker, P., 1852. Bijdrage tot de kennis der ichthyologische fauna van Blitong (Billiton), met beschrijving van eenige nieuwe soorten van zoetwatervisschen. Natuurkd. Tijdschr. Neder. Indië v., Volume 3, pp. 87-100.

Bleeker, P., 1853. Nalezingen op de ichthyologische fauna van het eiland Banka. Natuurkd. Tijdschr. Neder. Indië v., Volume 5, pp. 175-194.

Bleeker, P., 1866-1872. Atlas ichthyologique des Indie Orientales Néêrlandaises. Tome VI. Pleuronectes, Scombrésoces, Clupées, Clupésoces, Chauliodontes, Saurides. Amsterdam : Frédéric Muller.

269

Bohlen, J., Šlechtová, V., Tan, H. H. & Britz, R., 2011. Phylogeny of the Southeast Asian freshwater fish genus Pangio (Cypriniformes; Cobitidae). Molecular Phylogenetics and Evolution, 61(3), pp. 854-865.

Bork, D. & Mayland, H. J., 1998. Seltene Schönheiten im Süßwasseraquarium. Bornheim: Birgit Schmettkamp Verlag.

Brembach, M., 1978. Ein neuer Halbschnäbler aus Kalimantan (Süd Borneo). Vorläufige Beschreibung von Hemirhamphodon chrysopunctatus spec. nov.. Aquarium Aqua Terra v., 12(110), pp. 340-344.

Brito, P. & Edwards, S. V., 2009. Multilocus phylogeography and phylogenetics using sequence-based markers. Genetics , Volume 135, pp. 439-455.

Bromham, L. & Penny, D., 2003. The modern molecular clock. Nature Reviews Genetics, Volume 4, pp. 216-224.

Bulgin, N. L., Gibbs, H. L., Vickery, P. & Baker, A. J., 2003. Ancestral polymorphisms in genetic markers obscure detection of evolutionarily distinct populations in the endangered Florida grasshopper sparrow (Ammodramus savannarum floridanus). Molecular Ecology, Volume 12, p. 831–844.

Cannon, C. H., Morley, R. J. & Bush, A. B. G., 2009. The current refugial rainforests of Sundaland are unrepresentative of their biogeographic past and highly vulnerable to disturbance. Proceedings of National Academy of Sciences USA, Volume 106, pp. 11188-11193.

Carling, P. A., 2009. The geology of the Lower Mekong River. In: I. Campbell, ed. The Mekong: Biophysical and environment of an international river basin. New York: Elsevier Academic Press, pp. 13-28.

Carstens, B. C. & Knowles, L. L., 2007. Estimating species phylogeny from gene-tree probabilities despite incomplete lineage sorting: An example from melanoplus grasshoppers. Systematic Biology , 56(3), pp. 400-411.

Carvalho, G. R. & Hauser, L., 1994. Molecular genetics and the stock concept in fisheries. Reviews in Fish Biology and Fisheries, Volume 4, pp. 326-350.

270

Cassens, I., Mardulyn, P. & Milinkovitch, M. C., 2005. Evaluating intraspecific "network" construction methods using simulated sequence data: Do existing algorithms outperform the global maximum parsimony approach?. Systematic Biology , Volume 54, pp. 363-372.

Castello, J. & Templeton, A. R., 1994. Root probabilities for intraspecific gene trees under neutral coalescent theory. Molecular Phylogenetics and Evolution , Volume 3, pp. 102-113.

Chen, W.S., 2016. Land-sea changes in Taiwan since the Last Glacial Maximum, [online]. Available at: http://www.natgeomedia.com/column/external/44459 [Accessed 1 August 2016].

Chen, W. T., Ma, X. H., Shen, Y. J., Mao, Y. T. & He, S. P., 2015. The fish diversity in the upper reaches of the Salween River, Nujiang River, revealed by DNA barcoding. Scientific Reports, Volume 5, p. 17437.

Chen, X. L., Chiang, T. Y., Lin, H. D. & Zheng, H. S., 2007. Mitochondrial DNA phylogeography of Glyptothorax fokiensis and Glyptothorax hainanensis in Asia. Journal of Fish Biology, 70(Supplement A), pp. 75-93.

Choleva, L., Musilova, Z., Kohoutova-Sediva, A., Paces, J., Rab, P. & Janko, K., 2014. Distinguishing between incomplete lineage sorting and genomic introgressions: Complete fixation of allospecific mitochondrial DNA in a sexually reproducing fish (Cobitis; Teleostei), despite clonal reproduction of hybrids. PLos ONE, 9(6), p. e80641.

Collette, B. B., 2004. Family Hemiramphidae Gill 1859 - halfbeaks. California Academy of Sciences, Annotated Checklists of Fishes, Volume 22, p. 35.

Collins, R. A., Armstrong, K. F. & Meier, R., 2012. Barcoding and border biosecurity: identifying cyprinid fishes in the aquarium trade. PLoS ONE, Volume 7, p. e28381.

Corner, E. J. H., 1960. The Malayan flora. In: R. Purchon, ed. Proceedings of the Centenary and Bicentenary Congress of Biology. Singapore: University Malaya Press, pp. 21-24.

271

Costedoat, C., Chappaz, R., Barascud, B., Guillard, O. & Gilles, A., 2006. Heterogeneous colonization pattern of European cyprinids, as highlighted by the dace complex (Teleostei: Cyprinidae). Molecular Phylogenetics and Evolution, Volume 41, pp. 127-148.

Coyne, J. A. & Orr, H. A., 2004. Speciation. Sunderland (MA): Sinauer Associates.

Crandall, K. A., 1996. Multiple interspecies transmissions of human and simian T-cell leukemia/lymphoma virus Type I sequences. Molecular Biology and Evolution , Volume 13, pp. 115-131.

Crispo, E., Bentzen, P., Reznick, D. N., Kinnison, M. T. & Hendry, A. P., 2006. The relative influence of natural selection and geography on gene flow in guppies. Molecular Ecology , Volume 15, pp. 49-62.

Curran, L. M., Trigg, S. N., McDonald, A. K., Astiani, D., Hardiono, Y. M., Siregar, P., Caniago, I. & Kasischke, E., 2004. Lowland forest loss in protected areas of Indonesian Borneo. Science , Volume 303, pp. 1000-1003.

Dam, R. A., van der Kaars, S. & Kershaw, A. P., 2001. Quaternary environmental change in the Indonesian region. Palaeogeography, Palaeoclimatology, Palaeoecology, Volume 171, pp. 91-95.

Damm, S., Schierwater, B. & Hadrys, H., 2010. An integrative approach to species discovery in odonates: from character-based DNA barcoding to ecology. Molecular Ecology, Volume 19, pp. 3881-3893.

Davis, M. B. & Shaw, R. G., 2001. Range shifts and adaptive responses to Quaternary climate change. Science, Volume 292, pp. 673-679.

de Bruyn, M., 2005. Inferring biogeography from the evolutionary history of the giant freshwater prawn (Macrobrachium rosenbergii). PhD Thesis, Queensland University of Technology.

de Bruyn, M., Grail, W. & Carvalho, G. R., 2010. Anonymous nuclear markers for halfbeak fishes of the genus Hemirhamphodon. Conservation Genetic Resources, Volume 3, pp. 155-157.

272 de Bruyn, M., Nugroho, E., Mokarrom, M. H., Wilson, J. C. & Mather, P. B., 2005. Phylogeographic evidence for the existence of an ancient biogeographic barrier: the Isthmus of Kra Seaway. Heredity, Volume 94, pp. 370-378.

de Bruyn, M., Rüber, L., Nylinder, S., Stelbrink, B., Lovejoy, N. R., Lavoué, S., Tan, H. H., Nugroho, E., Wowor, D., Ng, P. K. L., Siti Azizah, M. N., von Rintelen, T., Hall, R. & Carvalho, G. R., 2013. Paleo-drainage basin connectivity predicts evolutionary relationships across three Southeast Asian biodiversity hotspots. Systematic Biology, 62(3), pp. 398-410.

de Bruyn, M., von Rintelen, T., von Rintelen, K., Mather, P. B. & Carvalho, G. R., 2012. Molecular biogeography and phylogeography of the freshwater fauna of the Indo-Australian Archipelago. In: D. J. Gower, et al. eds. Biotic Evolution and Environmental Change in Southeast Asia. Cambridge: Systematics Association, Vol., Cambridge University Press, pp. 316-347.

de Bruyn, M., Wilson, J. C. & Mather, P. B., 2004. Reconciling geography and genealogy: phylogeography of giant freshwater prawns from the Lake Carpentaria region. Molecular Ecology, Volume 13, pp. 3515-3526.

de Queiroz, K. & Good, D. A., 1997. Phenetic clustering in biology: A critique. Quaterly Review of Biology, Volume 72, pp. 3-30.

Dodson, J. J., Colombani, F. & Ng, P. K. L., 1995. Phylogeographic structure in mitochondrial DNA of a South-east Asian freshwater fish, Hemibagrus nemurus (Siluroidei; Bagridae) and Pleistocene sea-level changes on the Sunda shelf. Molecular Ecology, Volume 4, pp. 331-346.

Doi, A., 1997. A review of taxonomic studies of cypriniform fishes in Southeast Asia. Japanese Journal of Ichthyology, Volume 44, pp. 1-33.

Dolman, G. & Moritz, C., 2006. A multilocus perspective on refugial isolation and divergence in rainforest skinks (Carlia). Evolution, Volume 60, pp. 573-582.

Dominguez-Dominguez, O., Alda, F., de Leon, P. G., Garcia-Garitagoitia, J. L. & Doadrio, I., 2008. Evolutionary history of the endangered fish Zoogoneticus quitzeoensis (Bean, 1898) (Cyprinodontiformes: Goodeidae) using a sequential

273

approach to phylogeography based on mitochondrial and nuclear DNA data. BMC Evolutionary Biology, Volume 8, p. 161.

Dorn, A. & Greven, H., 2007. Some observations on courtship and mating of Hemirhamphodon tengah Anderson & Collette, 1991 (Zenarchopteridae). Bulletin of Fish Biology, 9(1/2), pp. 99-104.

Drummond, A. J., Ho, S. Y. W., Phillips, M. J. & Rambaut, A., 2006. Relaxed phylogenetics and dating with confidence.. PLoS Biol, p. 4:e88.

Drummond, A. J., Suchard, M. A., Xie, D. & Rambaut, A., 2012. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution, Volume 29, p. 1969–1973.

Duangjai Boonkusol & Wuttipong Tongbai, 2016. Genetic variation of striped snakehead fish (Channa striata) in river basin of central Thailand inferred from mtDNA COI gene sequences analysis. Journal of Biological Sciences, Volume 16, pp. 37-43.

Dudgeon, D., Arthington, A. H., Gessner, M. O., Zen-Ichiro, K., Knowler, D. J., Le´veˆque, C., Naiman, R. J., Prieur-Richard, A., Soto, D., Stiassny, M. L. J. & Sullivan, C. A., 2006. Freshwater biodiversity: importance, threats, status and conservation challenges. Biological Reviews, 81(2), pp. 163-182.

Dwivedi, B. & Gadagkar, S., 2009. Phylogenetic inference under varying proportions of indel-induced alignment gaps. BMC Evolutionary Biology , Volume 9, p. 211.

Edgar, R. C, 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research, 32(5), pp. 1792-1797.

Emerson, B. C., Paradis, E. & Thébaud, C., 2001. Revealing the demographic histories of species using DNA sequences. TREE, Volume 16, pp. 707-716.

Esa, Y., Siraj, S. S., Daud, S. K., Rahim, K. A. A., Mohd, T. A., Japning, J. R. R. & Tan, S. G., 2006. Mitochondrial DNA Diversity of Tor douronensis Valenciennes (Cyprinidae) in Malaysian Borneo. Pertanika Journal of Tropical Agriculture Sciences, 29((1&2)), pp. 47-55.

274

Esa, Y., Siraj, S. S., Daud, S. K., Rahim, K. A. A., Mohd, T. A., Japning, J. R. R. & Tan, S. G., 2008. Mitochondrial DNA diversity of Tor tambroides Valenciennes (Cyprinidae) from five natural populations in Malaysia. Zoological Studies, 47(3), pp. 360-367.

Esselstyn, J. A., Oliveros, C. H., Moyle, R. G., Peterson, A. T., McGuire, J. A. & Brown, R. M., 2010. Integrating phylogenetic and taxonomic evidence illuminates complex biogeographic patterns along Huxley’s modification of Wallace’s line. Journal of Biogeography, Volume 37, pp. 2054-2066.

Esselstyn, J. A. & Brown, R., 2009. The role of repeated sea-level fluctuations in the generation of shrew (Soricidae: Crocidura) diversity in the Philippine Archipelago. Molecular Phylogenetics and Evolution, Volume 53, pp. 171-181.

Excoffier, L. & Lischer, H. L., 2010. Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources, Volume 10, pp. 564-567.

Excoffier, L., Smouse, P. E. & Quattro, J. M., 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: Application to human mitochondrial DNA restriction data. Genetics , Volume 131, pp. 479-491.

Eytan, R. I. & Hellberg, M. E., 2010. Nuclear and mitochondrial sequence data reveal and conceal different demographic histories and population genetic processes in caribbean reef fishes. Evolution, 64(12), pp. 3380-3397.

Farias, I. P., Orti, G., Sampaio, I., Schneider, H. & Meyer, A., 2001. The cytochrome b gene as a phylogenetic marker: the limits of resolution for analyzing relationships among cichlid fishes. Journal of Molecular Evolution , Volume 53, pp. 89-103.

Felsenstein, J., 2005. Accuracy of coalescent likelihood estimates: Do we need more sites, more sequences, or more loci? Molecular Biology and Evolution, Volume 23, pp. 691-700.

Feulner, P. G. D., Kirschbaum, F., Schugardt, C., Ketmaier, V. & Tiedemann, R., 2006. Electrophysiological and molecular genetic evidence for sympatrically occurring cryptic species in African weakly electric fishes (Teleostei:

275

Mormyridae: Campylomormyrus). Molecular Phylogenetics and Evolution, Volume 39, pp. 198-208.

Flot, J. F., 2010. seqphase: a web tool for interconverting phase input/output files and fasta sequence alignments. Molecular Ecology Resources, Volume 10, pp. 162- 166.

Francis, C. M., Borisenko, A. V. & Ivanova, N. V., 2010. The role of DNA barcodes in understanding and conservation of mammal diversity in Southeast Asia. PLoS ONE, Volume 5, p. e12575.

Fraser, D. J. & Bernatchez, L., 2001. Adaptive evolutionary conservation: Towards a unified concept for defining conservation units. Molecular Ecology, Volume 10, pp. 2741-2752.

Fu, Y. X., 1997. Statistical tests for neutrality of mutations against population growth, hitchhiking and background selection. Genetics , Volume 147, pp. 915-925.

Fu, Y. X. & Li, W. H., 1999. Coalescing into the 21st century: an overview and prospects of coalescent theory. Theoretical Population Biology, Volume 56, p. 1–10.

Fuchs, J., Ericson, P. G. P. & Pasquet, E., 2008. Mitochondrial phylogeographic structure of the white-browed piculet (Sasia ochracea): cryptic genetic differentiation and endemism in Indochina. Journal of Biogeography, Volume 35, pp. 565-575.

Gale, L. R., Harrison, S. A., Ward, T. J., O’Donnell, K., Milus, E. A., Gale, S. W. & Kistler, H. C., 2011. Nivalenol-type populations of Fusarium graminearum and F. asiaticum are prevalent on wheat in Southern Louisiana. Phytopathology, 101(1), pp. 124-134.

Glaubrecht, M., Feh ér, Z. & von Rintelen, T., 2006. Brooding in Corbicula madagascariensis (Bivalvia, Corbiculidae) and the repeated evolution of viviparity in corbiculids. Zoologica Scripta , pp. 641-65435.

Glaubrecht, M., von Rintelen, T. & Korniushin, A. V., 2003. Toward a systematic revision of brooding freshwater Corbiculidae in Southeast Asia (Bivalvia,

276

veneroida): on shell morphology, anatomy and molecular phylogenetics of endemic taxa from islands in Indonesia. Malacologia , Volume 45, pp. 1-40.

Godfray, H. C. J., 2006. To boldy sequence. Trends in Ecology and Evolution, Volume 21, pp. 503-504.

Godinho, R., Crespo, E. G. & Ferrand, N., 2008. The limits of mtDNA phylogeography: complex patterns of population history in a highly structured Iberian lizard are only revealed by the use of nuclear markers. Molecular Ecology, Volume 17, pp. 4670-4683.

Goldstein, P. Z. & De Salle, R., 2011. Integrating DNA barcode data and taxonomic practice: determination, discovery and description. BioEssays, Volume 33, p. 135–147.

Gómez, A., Wright, P. J., Lunt, D. H., Cancino, J. M., Carvalho, G. R. & Hughes, R.N., 2007. Mating trials validate the use of DNA barcoding to reveal cryptic speciation of a marine bryozoan taxon. Proceedings of the Royal Society London Series B , 274(1607), pp. 199-207.

Google Earth v7.1.7.2606, [online].Available at: http://www.google.com/earth/index.html [Accessed 15 Jun 2016].

Gorog, A. J., Sinaga, M. H. & Engstrom, M. D., 2004. Vicariance or dispersal? Historical biogeography of three Sunda shelf murine rodents (Maxomys surifer, Leopoldamys sabanus and Maxomys whiteheadi ). Biological Journal of Linnean Society, Volume 81, pp. 91-109.

Grell, W., 1998. Pflege und Zucht eines eierlegenden Halbschnabelhechtes. Aquaristik aktuell, 6(11/12), pp. 24-26.

Gusma˜o, J., Lazoski, C. & Sole′-Cava, A. M., 2000. A new species of Penaeus crustacea: Penaeidae) revealed by allozyme and cytochrome oxidase I analysis. Marine Biology, Volume 137, pp. 435-446.

Gyllensten, U., 1985. The genetic structure of fish: differences in the intraspecific distribution of biochemical genetic variation between marine, anadromous, and freshwater species. Journal of Fish Biology, Volume 26, pp. 691-699.

277

Habel, J. C., Rodder, D., Schmitt, T. & Neve, G., 2011. Global warming will affect the genetic diversity and uniqueness of Lycaena helle populations. Global Change Biology, Volume 17, pp. 194-205.

Hajibabaei, M., Janzen, D. H., Burns, J. M., Hallwachs, W. & Hebert, P.D.N., 2006. DNA barcodes distinguish species of tropical Lepidoptera. Proceedings of the National Academy of Sciences of the United States of America, Volume 968-971, p. 103.

Hall, R., 1996. Reconstructing Cenozoic SE Asia. In: R. Hall & D. Blundell, eds. Tectonic evolution of Southeast Asia. London: Geological Society, pp. 153-184.

Hall, R., 1998. The plate tectonics of Cenozoic SE Asia and the distribution of land and sea. In: R. Hall & J. D. Holloway, eds. Biogeography and geological evolution of SE Asia. Leiden: Backhuys, pp. 99-131.

Hall, R., 2001. Cenozoic reconstructions of SE Asia and the SW Pacific: changing patterns of land and sea. In: I. Metcalfe, J. Smith, M. Morwood & I. Davidson, eds. Faunal and floral migrations and evolution in SE Asia–Australasia. A.A. Balkema: (Swets & Zeitlinger , pp. 35-36.

Hall, R., 2002. Cenozoic geological and plate tectonic evolution of SE Asia and the SW Pacific: computer-based reconstructions, model and animations. Journal of Asian Earth Sciences, Volume 20, p. 353–431.

Hall, R., 2009. Southeast Asia’s changing palaeogeography. Blumea, Volume 54, p. 148–161.

Hall, R., 2011. Australia-SE Asia collision: plate tectonics and crustal flow. In: R. Hall, M. Cottam & M. Wilson, eds. The Southeast Asian gateway: history and tectonics of Australia-Asia collision. London: The Geological Society of London, pp. 75-109.

Hall, R., 2012a. Late Jurassic-Cenozoic reconstructions of the Indonesian region and the Indian Ocean. Tectonophysics, Volume 570-571, pp. 1-41.

278

Hall, R., 2012b. Sundaland and Wallacea: geology, plate tectonics and palaeogeography. In: D. Gower, et al. eds. Biotic evolution environmental change in Southeast Asia. Cambridge: Cambridge University Press, pp. 32-78.

Hamm, C. A., 2011. Development of polymorphic anonymous nuclear DNA markers for the endangered Mitchell’s satyr butterfly, Neonympha mitchellii mitchellii (Lepidoptera: Nymphalidae). Conservation Genet Resources,Volume 4(1), pp. 127-128.

Harding, R. M., Fullerton, S. M., Griffiths, R. C., Bond, J., Cox, M. J., Schneider, J. A., Moulin, D. S. & Clegg, J. B., 1997. Archaic African and Asian lineages in the genetic ancestry of modern humans. The American Journal of Human Genetics, Volume 60, pp. 772–789.

Hare, M. P., 2001. Prospects for nuclear gene phylogeography. Trends in Ecology and Evolution, Volume 16, pp. 700-706.

Harley, M. M. & Morley, R. J., 1995. Ultrastructural studies of some fossil and extant palm pollen, and the reconstruction of the biogeographical history of subtribes Iguanurinae and Calaminae. Review of Palaeobotany Palynology, Volume 85, pp. 153-182.

Harpending, H. C., 1994. Signature of ancient population growth in a low resolution DNA mismatch distribution. Human Biology , Volume 66, pp. 591-600.

Harrigan, R. J., Mazza, M. E. & Sorenson, M. D., 2008. Computation vs. cloning: evaluation of two methods for haplotype determination. Molecular Ecology Resources, Volume 8, pp. 1239-1248.

Hartl, A., 1983. Halbhechte und Halbschnabelhechte Harmlose Namensverwandte unseres Hechtes aus Sudostasien (Schluss). Aquar. Terrar. Zeitschnft., 36(10), pp. 368-371.

Hauser, L., Adcock, G. J., Smith, P. J., Bernal Ramirez, J. H. & Carvalho, G.R., 2002. Loss of microsatellite diversity and low effective population size in an overexploited population of New Zealand snapper (Pagrus auratus). Proceedings of the National Academy of Sciences of the United States of America , 99(18), pp. 11742-11747.

279

Hawlitschek, O., Hendrich, L., Espeland, M., Toussaint, Emmanuel F. A., Genner, M. J. & Balke, M., 2012. Pleistocene climate change promoted rapid diversification of aquatic invertebrates in Southeast Australia. BMC Evolutionary Biology, Volume 12, p. 142.

Haywood, A. M., Dowsett, H. J., Valdes, P.J., Lunt, D.J., Francis, J. E. & Sellwood, B. W., 2009. Introduction. Pliocene climate, processes and problems. Philosophical Transactions of the Royal Society A , Volume 367, pp. 3-17.

Heaney, L. R., 2000. Dynamic disequilibrium: a long-term, large-scale perspective on the equilibrium model of island biogeography. Global Ecology and Biogeography, Volume 9, pp. 59-74.

Hebert, P. D. N., Cywinska, A., Ball, S. L. & deWaard, J. R., 2003a. Biological identification through DNA barcodes. Proceedings of the Royal Society of London, Series B: Biological Sciences, Volume 270, pp. 313-322.

Hebert, P. D. N. & Gregory, T. R., 2005. The promise of DNA barcoding for taxonomy. Systematic Biology , Volume 54, pp. 852-859.

Hebert, P. D. N., Penton, E. H., Burns, J. M., Janzen, D. H. & Hallwachs, W., 2004. Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfly Astraptes fulgerator. Proceedings of the National Academy of Sciences of the United States of America, Volume 101, pp. 812-817.

Hebert, P. D. N., Ratnasingham, S. & deWaard, J. R., 2003b. Barcoding animal life: cytochrome c oxidase subunit 1 divergences among closely related species. Proceedings of the Royal Society of London, Series B: Biological Sciences, 270(Suppl), pp. s96-s99.

Heled, J. & Drummond, A. J., 2008. Bayesian inference of population size history from multiple loci. BMC Evolutionary Biology, Volume 9, p. 289.

Hewitt, G. M., 2000. The genetic legacy of the Quaternary ice ages. Nature, Volume 405, pp. 907-913.

Hewitt, G. M., 1999. Post-glacial re-colonization of European biota. Biological Journal of the Linnean Society, Volume 68, pp. 87-112.

280

Hey, J., 2010. Isolation with migration models for more than two populations. Molecular Biology and Evolution, Volume 27, pp. 905-920.

Hickerson, M. J., Carstens, B. C., Cavender-Bares, J., Crandall, K. A., Graham, C. H., Johnson, J. B., Rissler, L., Victoriano, P. F. & Yoder, A. D., 2010. Phylogeography's past, present, and future: 10 years after. Molecular Phylogenetics and Evolution, Volume 54, pp. 291-301.

Holmes, B. H., Steinke, D. & Ward, R. D., 2009. Identification of shark and ray fins using DNA barcoding. Fisheries Research, Volume 95, pp. 280-288.

Ho, S. Y. W., Saarma, U. & Barnett, R., 2008. The effect of inappropriate calibration: three case studies in molecular ecology. PLoS ONE, Volume 3, p. e1615.

Huang, Z. S. & Zhang, D. X., 2008. Haplotype reconstruction for scnp DNA: a consensus vote approach with extensive sequence data from populations of the migratory locust (Locusta migratoria). Molecular Ecology , Volume 17, pp. 1930-1947.

Hubert, N., Delrieu-Trottin, E., Irisson, J. O., Meyer, C. & Planes, S., 2010. Identifying early stages of coral reef fishes through DNA barcoding : A test case with the families Acanthuridae and Holocentridae. Molecular Phylogenetic Evolution, Volume 55, pp. 1195-1203.

Hubert, N. & Hanner, R., 2015. DNA Barcoding, species delineation and taxonomy: a historical perspective. DNA Barcode, Volume 3, pp. 44-58.

Hubert, N., Hanner, R., Holm, E., Mandrak, N. E., Taylor, E. & Burridge, M., 2008. Identifying Canadian freshwater fishes through DNA barcodes. PLoS ONE, 3(6), p. e2490.

Hubert, N., Kadarusman, Wibowo, A., Busson, F., Caruso, D., Sulandari, S., Nafiqoh, N., Pouyaud, L., Rüber, L., Avarre, J., Herder, F., Hanner, R., Keith, P. & Hadiaty, R. K., 2015. DNA Barcoding Indonesian freshwater fishes:challenges and prospects. DNA Barcodes , Volume 3, pp. 144-169.

Hubert, N., Meyer, C. P., Bruggemann, H. J., Guerin, F., Komeno, R. J. L. & Espiau, B., 2012. Cryptic diversity in Indo-Pacific coral-reef fishes revealed by DNA

281

barcoding provides new support to the Centre-of-Overlap hypothesis. PLoS One, 7(3), p. e28987.

Hudson, R. R., Slatkin, M. & Maddison, M. P., 1992. Estimation of level of gene flow from DNA sequence data. Genetics, Volume 132, pp. 583-589.

Hurwood, D. A., Adamson, E. A. S. & Mather, P. B., 2008. Evidence for strong genetic structure in an economically important, highly vagile cyprinid (Henicorhynchus lobatus) in the Mekong River Basin. Ecology of Freshwater Fish , Volume 17, pp. 272-283.

Hurwood, D. A., Hughes & M., J., 1998. Phylogeography of the freshwater fish, Mogurnda adspersa, in streams of northeast Queensland, Australia: evidence for altered drainage patterns. Molecular Ecology, Volume 7, pp. 1507-1517.

Huxley, T. H., 1868. On the classification and distribution of the Alectoromorphae and Heteromorphae. Proceeding of the Zoological Society of London, pp. 294-319.

Inger, R. F. & Chin, P. K., 1962. The freshwater fishes of North Borneo. Fieldiana: Zoology, Volume 45, pp. 1-268.

Inger, R. F. & Voris, H. K., 2001. The biogeographical relations of the frogs and snakes of Sundaland. Journal of Biogeography,, Volume 28, pp. 863-891.

Irwanto, D., 2015. Last Glacial Priod of Sundaland. [Image]. [online] Available at: http://atlantisjavasea.blogspot.my/p/sundaland.html [Accessed 17 August 2016].

Ivanova, N. V., Zemlak, T. S., Hanner, R. H. & Hebert, P. D. N., 2007. Universal primer cocktails for fish DNA barcoding. Molecular Ecology Notes, Volume 7, pp. 544-548.

Jamsari, A. F. J., Muchlisin, Z. A., Musri, M. & Siti Azizah, M. N., 2010. Remarkably low genetic variation but high population differentiation in the climbing perch, Anabas testudineus (Anabantidae), based on the mtDNA control region. Genetics and Molecular Research, 9(3), pp. 1836-1843.

282

Jamsari, A. F. J., Tan, M. P. & Siti Azizah, M. N., 2011. Genetic structure of the snakehead murrel, Channa striata (channidae) based on the cytochrome c oxidase subunit I gene: influence of historical and geomorphological factors. Genetics and Molecular Biology., Volume 34, pp. 152-160.

Janzen, D. H., Hajibabaei, M., Burns, J. M., Hallwachs, W., Remigio, E. & Hebert, P. D. N., 2005. Wedding biodiversity inventory of a large and complex Lepidoptera fauna with DNA barcoding. Philosophical Transactions of the Royal Society of London, SeriesB, Biological Sciences, Volume 360, pp. 1835-1845.

Jennings, W. B. & Edwards, S. V., 2005. Speciational history of Australian grass finches (Poephila) inferred from thirty gene trees. Evolution, Volume 59, p. 2033–2047.

Johns, G. C. & Avise, J. C., 1998. A comparative summary of genetic distances in the vertebrates from the mitochondrial cytochrome b gene. Molecular Biology and Evolution , Volume 15, pp. 1481-1490.

Kadarusman, Hubert, N., Hadiaty, R. K., Sudarto, Paradis, E. & Pouyaud, L., 2012. Cryptic diversity in Indo-Australian rainbow fishes revealed by DNA barcoding: implications for conservation in a biodiversity hotspot candidate. PLoS ONE, 7(7), p. e40627.

Kamarudin, K. R. & Esa, Y., 2009. Phylogeny and phylogeography of Barbonymus schwanenfeldii (Cyprinidae) from Malaysia inferred using partial cytochrome b mtDNA gene. Journal of Tropical Biology and Conservation, Volume 5, pp. 1- 13.

Katherine, A., 2006. A Field Guide to the Fishes of Kuching Rivers, Sarawak Malaysia. Borneo: Natural History Publication.

Kim, S., Eo, H. S., Koo, H., Choi, J. K. & Kim, W., 2010. DNA Barcode based molecular identification system for fish species. Molecules and Cells, Volume 30, pp. 507-512.

Kimura, M. & Weiss, G. H., 1964. The Stepping Stone Model of population structure and the decrease of genetic correlation with distance. Genetics , Volume 19, pp. 561-576.

283

Kingman, J. F. C., 1982a. The coalescent. Stochastic Processes and their Applications , Volume 13, pp. 235-248.

Kingman, J. F. C., 1982b. On the genealogy of large population. In: J. Gani & E. Hannan, eds. Essays in statistical science. London, UK.: Applied Probability Trust, pp. 27-43.

Kingman, J. F. C., 2000. Origins of the coalescent. Genetics, Volume 156, pp. 1461- 1463.

Knowles, L. L., 2009. Statistical phylogeography. Annual Review of Ecology, Evolution and Systematics, Volume 40, pp. 593-612.

Knowlton, N., 1993. Sibling species in the sea. Annual Review of Ecology, Evolution and Systematics, Volume 24, pp. 189-216.

Kocher, T. D. & Carleton, K. L., 1997. Base substitution in fish mitochondrial DNA: patterns and rates. In: T. D. Kocher & C. A. Stepien, eds. Molecular Systematics of Fishes. California: Academic Press, , pp. 13-24.

Kocher, T. D. & Stepien, C. A., 1997. Molecular systematics of the fishes. San Deigo, California: Academic Press.

Kochzius, M. & Nuryanto, A., 2008. Strong genetic population structure in the boring giant clam, Tridacna crocea, across the Indo-Malay archipelago: Implications related to evolutionary processes and connectivity. Molecular Ecology , Volume 17, pp. 3775-3787.

Kodandaramaiah, U., 2009. Vagility: The neglected component in historical biogeography. Evolutionary Biology, Volume 36, pp. 327-335.

Koh, L. & Sodhi, N., 2010. Conserving Southeast Asia’s imperiled biodiversity: scientific, management, and policy challenges. Biodiversity and Conservation, Volume 19, pp. 913-917.

284

Ko, H. L., Wang, Y. T., Chiu, T. S., Lee, M. A., Leu, M. Y., Chang, K. Z., Chen, W. Y. & Shao, K. T., 2013 . Evaluating the accuracy of morphological identification of larval fishes by applying DNA barcoding. PLoS One, Volume 8, p. e53451.

Komal, S. P., Lakra, W. S., Bhatt, J. P., Goswami, M. & Malakar, A. K., 2013. Population structure of Tor tor inferred from mitochondrial gene cytochrome b. Mitochondrial DNA, 24(3), pp. 290-296.

Kottelat, M., 2002. Aquatic systems: neglected biodiversity. In: E. Wikramanayake, et al. eds. Terrestrial ecoregions of the Indo-Pacific. Washington: Island Press, pp. 30-35.

Kottelat, M., 2013. The fishes of the inland waters of Southeast Asia: A catalog and core bibliography of the fishes known to occur in freshwaters, mangroves and estuaries. Raffles Bulletin of Zoology, Volume Supplement 27, pp. 1-663.

Kottelat, M., Whitten, T., Kartikasari, S. N. & Wirjoatmodjo, S., 1993. Freshwater fishes of western Indonesia and Sulawesi. Jakarta, Periplus Editions, p. 221.

Kuhner, M. K., 2009. Coalescent genealogy samplers: windows into population history. Trends in Ecology and Evolution, Volume 24, pp. 86-93.

Kuussaari, M., Bommarco, R., Heikkinen, R. K., Helm, A., Krauss, J., Lindborg, R., Öckinger, E., Pärtel, M., Pino, J., Rodà, F., Stefanescu, C., Teder, T., Zobel, M. & Steffan-Dewenter, I., 2009. Extinction debt: a challenge for biodiversity conservation. Trends Ecology and Evolution, Volume 24, p. 564–571.

Lambert, F. R. & Collar, N. J., 2002. The future for Sundaic lowland forest birds: longterm effects of commercial logging and fragmentation. Forktail , Volume 18, pp. 127-146.

Lanfear, R., Calcott, B., Ho, S. Y. W. & Guindon, S., 2012. PartitionFinder: Combined selection of partitioning schemes and substitution models for phylogenetic analyses. Molecular Biology and Evolution, 29(6), pp. 1695-1701.

Lara, A., Jose´ Luis Ponce De Leo´, N., Rodri´Guez, R., Casane, D., Te´, G., Bernatchez, L. & Garci´A-Machado, E., 2010. DNA barcoding of Cuban

285

freshwater fishes: evidence for cryptic species and taxonomic conflicts.. Molecular Ecology Resources, Volume 10, pp. 421-430.

Lee, P. & Sulaiman, Z., 2015. Genetic identification and structure of Clarias batrachus (Linnaeus, 1758) from Southeast Asia using a mitochondrial DNA marker. Zootaxa, 3962(1), pp. 182-190.

León-Romero, Y., Omar, M. & Soto-Galera, E., 2012. DNA barcoding reveals taxonomic conflicts in the Herichthys bartoni species group (Pisces: Cichlidae). Molecular Ecology Resources, Volume 12, pp. 1021-1026.

Li, G. Y., Wang, X. Z., Zhao, Y. H., Zhang, J., Zhang, C. G. & He, S. P., 2009. Speciation and phylogeography of Opsariichthys bidens (Pisces: Cypriniformes: Cyprinidae) in China: Analysis of the cytochrome b gene of mtDNA from diverse populations. Zoological Studies, 48(4), pp. 569-583.

Li, J., Wang, X., Kong, X., Zhao, K., He, S. & Mayden, R. L., 2008. Variation patterns of the mitochondrial 16S rRNA gene with secondary structure constraints and their application to phylogeny of cyprinine fishes (Teleostei: Cypriniformes). Molecular Phylogenetics and Evolution, Volume 47, pp. 472-487.

Liao, P. C., Kuo, D. C., Lin, C. C., Ho, K. C., Lin, T. P. & Hwang, S. Y., 2010. Historical spatial range expansion and a very recent bottleneck of Cinnamomum kanehirae Hay. (Lauraceae) in Taiwan inferred from nuclear genes. BMC Evolutionary Biology , Volume 10, p. 124.

Librado, P. & Rozas, J., 2009. DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bioinformatics, Volume 25, pp. 1451-1452.

Lim, H. C., Abu Talib, A., Ahmad Adnan, N. & Siti Azizah, M. N., 2014. Cytochrome b gene reveals panmixia among Japanese Threadfin Bream, Nemipterus japonicus (Bloch, 1791) populations along the coasts of Peninsular Malaysia and provides evidence of a cryptic species. Mitochondrial DNA, Volume Early Online, pp. 1-10.

Lim, H. C., Rahman, M. A., Lim, S. L. H., Moyle, R. G. & Sheldon, F. H., 2011. Revisiting Wallace’s haunt: coalescent simulations and comparative niche

286

modeling reveal historical mechanisms that promoted avian population divergence in the Malay Archipelago. Evolution , Volume 65, pp. 321-334.

Lim, K. K. P., 1995. Rasbora kottelati, a new species of cyprinid fish from North- western Borneo. Raffles Bulletin of Zoology, Volume 43, pp. 65-75.

Lisiecki, L. E. & Raymo, M. E., 2005. A Pliocene-Pleistocene stack of 57 globally distributed benthic δ18O records. Paleoceanography, Volume 20, p. 1003.

Lockley, A. K. & Bardsley, R. G., 2000. DNA-based methods for food authentication. Trends Food Science and Technology, Volume 11, pp. 67-77.

Lohman, D. J., de Bruyn, M., Page, T., von Rintelen, K., Hall, R., Ng, P. K. L., Shih, H. T., Carvalho, G. R. & von Rintelen, T., 2011. Biogeography of the Indo- Australian Archipelago. Annual Review of Ecology, Evolution and Systematics, Volume 42, pp. 205-226.

Lovejoy, N. R., 2000. Reinterpreting recapitulation: Systematics of needle fishes and their allies (Teleostei: Beloniformes). Evolution, 54(4), pp. 1349-1362.

Lowenstein, J. H., Burger, J., Jeitner, C. W., Amato, G., Kolokotronis, S. O. & Gochfeld, M., 2010. DNA barcodes reveal species-specific mercury levels in tuna sushi that pose a health risk to consumers. Biology Letters, Volume 6, pp. 692-695.

Lundberg, J. G., 1993. African–South American freshwater fish clades and continental drift: problems with a paradigm. In: P. Goldblatt, ed. Biotic Relationships Between Africa and South America . New Haven, CT: Yale University Press, pp. 156-198.

Luo, S. J., Kim, J. H., Johnson, W. E., van derWalt, J. & Martenson, J., 2004. Phylogeography and genetic ancestry of tigers (Panthera tigris). PLoS Biol, Volume 2, pp. 2275-2293.

Manel, S., Schwartz, M. K., Luikart, G. & Taberlet, P., 2003. Landscape genetics: Combining landscape ecology and population genetics. Trends of Ecology and Evolution, 18:189–197., Volume 18, pp. 189-197.

287

Manni, F., Gu´erard, E. & Heyer, E., 2004. Geographic patterns of (genetic, morphologic, linguistic) variation: How barriers can be detected using Monmonier’s algorithm. Human Biology, 76:173–190., Volume 76, pp. 173-190.

Mantel, N., 1967. The detection of disease clustering and a generalized regression approach.. Cancer Research, Volume 27, p. 209–220.

Martins, C., Wasko, A. P., Oliveira, C. & Foresti, F., 2003. Mitochrondrial DNA variation in wild populations of Leporinus elongatus from the Parana River basin. Genetics and Molecular Biology, 26(1), pp. 33-38.

Maruyama, T. & Fuerst, P. A., 1985. Population bottlenecks and non-equilibrium models in population genetics. II. Number of alleles in a small population that was formed by a recent bottleneck. Genetics , Volume 111, pp. 675-689.

Mat Jaafar, T. N. A., Taylor, M. I., Mohd Nor, S. A., de Bruyn, M. & Carvalho, G.R., 2012. DNA barcoding reveals cryptic diversity within commercially exploited Indo-Malay Carangidae (Teleosteii: Perciformes). PLoS ONE, 7(11), p. e49623.

Mayr, E., 1944. Wallace’s line in the light of recent zoogeographic studies. Quarterly Review of Biology, Volume 19, pp. 1-14.

McConnell, S. K. J., 2004. Mapping aquatic faunal exchanges across the Sunda shelf, South-East Asia, using distributional and genetic data sets from the cyprinid fish Barbodes gonionotus (Bleeker, 1850). Journal of Natural History, 38(5), pp. 651-670.

Meier, R., Kwong, S., Vaidya, G. & Ng, P. K. L., 2006. DNA barcoding and taxonomy in Diptera: a tale of high intraspecific variability and low identification success. Systematic Biology, Volume 55, pp. 715-728.

Meier, R., Zhang, G. & Ali, F., 2008. The use of mean instead of smallest interspecific distances exaggerates the size of the ‘barcoding gap’ and leads to misidentification. Systematic Biology, Volume 57, pp. 809-813.

Metcalfe, I., 1984. Stratigraphy, palaeontology and palaeogeography of the Carboniferous of Southeast Asia. Memoirs of the Geological Society of France, Volume 147, pp. 107-118.

288

Metcalfe, I., 2002. Permian tectonic framework and palaeogeography of SE Asia. Journal of Asia Earth Sciences, Volume 20, pp. 551-566.

Metcalfe, I., 2005. Asia: South-East. In: R. C. Selley, L. R. M. Cocks & I. R. Plimer, eds. Encyclopediaof Geology. Oxford: Elsevier, pp. 169-198.

Metcalfe, I., 2006. Palaeozoic and Mesozoic tectonic evolution and palaeogeography of East Asian crustal fragments: the Korean Peninsula in context. Gondwana Research, Volume 9, pp. 24-46.

Metcalfe, I., 2009. Late Palaeozoic and Mesozoic tectonic and palaeogeographic evolution of SE Asia. In: E. Buffetaut, G. Cuny, J. Le Loeuff & V. Suteethorn, eds. Late Palaeozoic and Mesozoic Ecosystems in SE Asia. London: Geological Society, pp. Special Publications,315, 7-23.

Metcalfe, I., 2011a. Palaeozoic-Mesozoic history of SE Asia. In: R. Hall, M. Cottam & M. Wilson, eds. The Southeast Asian Gateway:history and tectonics of Australia-Asia collision . London: Geological Society, pp. 7-35.

Meyer, C. P. & Paulay, G., 2005. DNA barcoding: error rates based on comprehensive sampling. PLoS Biology, 3(12), p. e422.

Michaux, B., 1991. Distribution patterns and tectonic development in Indonesia: Wallace reinterpreted. Australia Systematic Botany, Volume 4, pp. 25-36.

Michel, C., Hicks, B. J., Sto¨lting, K. N., Clarke, A. C., Stevens, M. I., Ray, T., Axel, M. & van den Heuvel, M. R., 2008. Distinct migratory and non-migratory ecotypes of an endemic New Zealand eleotrid (Gobiomorphus cotidianus)- implications for incipient speciation in island freshwater fish species. BMC Evolutionary Biology, 8(49).

Miller, M. P., 2005. Alleles in space (AIS): Computer software for the joint analysis of interindividual spatial and genetic information. Journal of Heredity, Volume 96, pp. 722-724.

289

Miya, M. & Nishida, M., 2000. Use of mitogenomic information in teleostean molecular phylogenetics: a tree-based exploration under the maximum- parsimony optimality criterion. Molecular Phylogenetic Evolution, 17(3), pp. 437-455.

Mohd Lutfi, A., Mansor, M. I. & Siti Azizah, M. N., 2010. The establishment of DNA barcode for Malaysian Arrids. Bandung, Indonesia, Proceedings of the First Joint Symposium of ITB-USM.

Monaghan, M. T., Spaak, P., Robinson, C. T. & Ward, J. V., 2001. Genetic differentiation of Baetis alpinus Pictet (Ephemeroptera: Baetidae) in fragmented alpine streams. Heredity, Volume 86, pp. 395-403.

Moritz, C., 1994. Defining “Evolutionarily Significant Units” for conservation. Trends in Ecology and Evolution , 9(10), pp. 373-375.

Moritz, C., 2002. Strategies to protect biological diversity and the processes that sustain it. Systematic Biology, 51(2), p. 238–254.

Morley, R. J., 2000. Origin and evolution of tropical rain forest. Chichester: John Wiley and Sons.

Moss, S. J. & Wilson, M. E. J., 1988. Biogeographic implications from the Tertiary palaeogeographic evolution of Sulawesi and Borneo. In: J. D. Holloway & R. Hall, eds. Biogeography and geological evolution of SE Asia. Leiden: Backhuys, pp. 133-164.

Moyer, G. R., Winemiller, K. O., McPhee, M. V. & Turner, T. F., 2005. Historical demography, selection, and coalescence of mitochondrial and nuclear genes in Prochilodus species of northern South America. Evolution, Volume 59, pp. 596- 610.

Myers, N., Mittermeier, R. A., Mittermeier, C. G., da Fonseca, G. A. B. & Kent, J., 2000. Biodiversity hotspots for conservation priorities. Nature, Volume 403, pp. 853-858.

Naish, T. R. & Wilson, G. S., 2009. Constraints on the amplitude of mid-Pliocene (3.6–2.4 Ma) eustatic sea-level fluctuations from the New Zealand shallow-

290

marine sediment record. Philosophical Transaction of the Royal Society A , pp. 169-187.

Nei, M., 1987. Molecular Evolutionary Genetics. New York, USA: Columbia University Press.

Nelson, K. & Soulé, M., 1987. Genetical conservation of exploited fishes. In: N. Ryman & F.Utter, eds. Population genetics & fishery management. Seattle: University Washington Press, pp. 345-368.

Newman, D. & Pilson, D., 1997. Increased probability of extinction due to decreased genetic effective population size: experimental populations of Clarkia pulchella. Evolution , Volume 51, pp. 354-362.

Nguyen, T. T. T., 2008. Population structure in the highly fragmented range of Tor douronensis (Cyprinidae) in Sarawak, Malaysia revealed by microsatellite DNA markers. Freshwater Biology, Volume 53, pp. 924-934.

Nguyen, T. T. T., Hurwood, D., Mather, P., Na-Nakorn, U., Kamonrat, W. & Bartley, D., 2006a. Manual on applications of molecular tools in aquaculture and inland fisheries management, Part 2: Laboratory protocols and data analysis. Bangkok: NACA monograph No2.

Nguyen, T. T. T., Ingram, B. A., Sungan, S., Gooley, G., Sim, S. Y., Tinggi, D. & De Silva, S. S., 2006. Mitochodnrial DNA diversity of broostockl of two indigenous mahseer species, Tor tambroides and T. douronensis (Cyprinidae) cultured in Sarawak, Malaysia. Aquaculture, Volume 253, pp. 259-269.

Nguyen, T. T. T., Na-Nakorn, U., Sukmanomon, S. & Ziming, C., 2008. A study on phylogeny and biogeography of mahseer species (Pisces, Cyprinidae) using sequences of three mitochondrial DNA gene regions. Molecular Phylogenetics and Evolution, Volume 48, pp. 1223-1231.

Nijman, V., 2010. An overview of international wildlife trade from Southeast Asia. Biodiversity and Conservation, 19(4), pp. 1101-1114.

Noor Adelyna, M. A., 2013. DNA barcoding of the family Scombridae in Malaysia and phylogeography and population structure of the Indian mackerel,

291

Rastrelliger kanagurta for sustainable fisheries. PhD Thesis. Universiti Sains Malaysia.

Noor Adelyna, M. A., Tam, B. M., Phumee, P., Muchlisin, Z. A., Tamadoni, S., Peter, B. M. & Siti Azizah, M. N., 2015. High connectivity in Rastrelliger kanagurta: Influence of historical signatures and migratory behaviour inferred from mtDNA cytochrome b. PLoS ONE, 10(3), p. e0119749.

Packer, L., Gibbs, J., Sheffield, C. & Hanner, R., 2009. DNA barcoding and the mediocrity of morphology. Molecular Ecology Resources, Volume 9, pp. 42-50.

Padial, J. M., Miralles, A., De la Riva, I. & Vences, M., 2010. The integrative future of taxonomy. Frontiers in Zoology, Volume 7, p. 16.

Parr, C. S., Guralnick, R., Cellinese, N. & Page, R. D. M., 2012. Evolutionary informatics: unifying knowledge about the diversity of life. Trends in Ecology & Evolution, Volume 27, pp. 94-103.

Pasi, K. S., Lakra, W. S., Bhatt, J. P., Goswami, M. & Malakar, A. K., 2013. Population structure of Tor tor inferred from mitochondrial gene cytochrome b. Mitochondrial DNA, 24(3), pp. 290-296.

Peh, K. S. H., 2010. Invasive species in Southeast Asia: the knowledge so far. Biodiversity and Conservation, 19(4), pp. 1083-1099.

Perdices, A., Cunha, C. & Coelho, M. M., 2004. Phylogenetic structure of Zacco platypus (Teleostei, Cyprinidae) populations on the upper and middle Chang Jiang (=Yangtse) drainage inferred from cytochrome b sequences. Molecular Phylogenetics and Evolution, Volume 31, pp. 192-203.

Piscesilim ([email protected])(15th November 2012). Differentiation between H. kuekenthali and H. pogonognathus. Email to HC, Lim ([email protected]).

Planes, S., Doherty, P. J. & Bernardi, G., 2001. Strong genetic divergence among populations of a marine fish with limited dispersal, Acanthochromis polyacanthus, within the Great Barrier Reef and the Coral Sea. Evolution , Volume 55, pp. 2263-2273.

292

Poissant, J., Knight, T. & Ferguson, M. M., 2005. Nonequilibrium conditions following landscape rearrangement: the relative contribution of past and current hydrological landscapes on the genetic structure of a stream dwelling fish. Molecular Ecology, Volume 14, pp. 1321-1331.

Posada, D. & Crandall, K. A., 2001. Intraspecific gene genealogies: trees grafting into networks. Trends in Ecology and Evolution, Volume 16, pp. 37-45.

Posadas, P., Crisci, J. V. & Katinas, L., 2006. Historical biogeography: A review of its basic concepts and critical issues. Journal of Arid Environments, Volume 66, pp. 389-403.

Pouyaud, L., Sudarto, T. & Teugels, G., 2003. The different colour varieties of the Asian arowana Scleropages formosus (Osteoglossidae) are distinct species: Morphologic and genetic evidences. Cybium, Volume 27, pp. 287-305.

Printzen, C., Ekman, S. & Tonsberg, T., 2003. Phylogeography of Cavernularia hultenii: evidence of slow genetic drift in a widely disjunct lichen. Molecular Ecology, Volume 12, pp. 1473-1486.

Puckridge, M., Andreakis, N., Appleyard, S.A. & Ward, R. D., 2013. Cryptic diversity in flathead fishes (Scorpaeniformes: Platycephalidae) across the Indo-West Pacific uncovered by DNA barcoding. Molecular Ecology Resources, Volume 13, pp. 32-42.

Puillandre, N., Lambert, A., Brouillet, S. & Achaz, G., 2012. ABGD, automated barcode gap discovery for primary species delimitation. Molecular Ecology, Volume 21, pp. 1864-1877.

Pulquério, M. J. F. & Nichols, R. A., 2007. Dates from the molecular clock: how wrong can we be? Trends in Ecology & Evolution, Volume 22, pp. 180-184.

Rainboth, W. J., 1996. Fishes of the Cambodian Mekong. FAO species identification field guide for fishery purposes. Rome: Food and Agriculture Association of the United Nations.

Rambaut, A., 2006. FigTree. v1.3.1. [online] . Available at: http://tree.bio.ed.ac.uk/software/figtree.

293

Rambaut, A. & Drummond, A., 2007. Tracer v1.4.. [online] . Available at: http://beast.bio.ed.ac.uk/Tracer

Ramos-Onsins, S. E. & Rozas, J., 2002. Statistical properties of new neutrality tests against population growth. Molecular Biology and Evolution , Volume 19, pp. 2092-2100.

Rand, D. M., 1996. Neutrality tests of molecular markers and the connection between DNA polymorphism, demography, and conservation biology. Conservation Biology, Volume 10, pp. 665-671.

Randi, E., 2000. Mitochondrial DNA. In: A. Baker, ed. Molecular Methods in Ecology. Oxford: Blackwell Science Ltd, pp. 136-167.

Ratnasingham, S. & Hebert, P. D. N., 2007. Bold: The barcode of life data system (http://www.Barcodinglife.Org). Molecular Ecology Notes, Volume 7, pp. 355- 364.

Ren, Q. P., Fan, Z., Zhou, X. M., Jiang, G. F., Wang, Y. T. & Liu, Y. X., 2013. Identification and characterization of anonymous nuclear markers for the double-striped cockroach, Blattella bisignata. Bulletin of Entomological Research , Volume 103, pp. 29-35.

Ribeiro, A. D. O., Caires, R. A., Mariguela, T. C., Pereira, L. H. G., Hanner, R. & Oliveira, C., 2012. DNA barcodes identify marine fishes of São Paulo State, Brazil. Molecular Ecology Resources, Volume 12, pp. 1012-1020.

Ridder-Numan, J. W. A., 1996. Historical biogeography of Southeast Asian genus Spatholobus (Legum-Papilionoideae) and its allies. Blumea, Volume 0, pp. 1- 144.

Roberts, T. R., 1989. The freshwater fishes of Western Borneo (Kalimantan Barat, Indonesia). San Fancisco: California Academy of Sciences.

Robinson, E. A., Blagoev, G. A., Hebert, P. D. N. & Adamowicz, S. J., 2009. Prospects for using DNA barcoding to identify spiders in species-rich genera. Zookeys, Volume 16, pp. 27-46.

294

Rock, J. Costa, F. O., Walker, D. I., North, A. W. & Hutchinson, W. F., 2008. DNA barcodes of fish of the Scotia Sea, Antarctica indicate priority groups for taxonomic and systematics focus. Antarctic Science, Volume 20, pp. 253-262.

Rogers, A. R. & Harpending, H., 1992. Population growth makes waves in the distribution of pairwise genetic differences. Molecular Biology and Evolution , Volume 9, pp. 552-569.

Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D. L., Darling, A. & Höhna, S., 2012. MrBayes 3.2: Efficient Bayesian phylogenetic inference and model choice across a large model space. Systematic Biology, 61(3), pp. 539-542.

Roos, H., 2004. Genetic diversity in the anabantids Sandelia capensis and S. bainsii: a phylogeography and phylogenetic investigation. MSc thesis, University of Pretoria.

Rosenblum, E. B., Belfiore, N. M. & Moritz, C., 2007. Anonymous nuclear markers for the eastern fence lizard, Sceloporus undulates. Molecular Ecology Notes, Volume 7, pp. 113-116.

Rosso, J. J., Mabragana, E., Castro, M. G. & de Astarloa, J. M., 2012. DNA barcoding Neotropical fishes: recent advances from the Pampa Plain, Argentina. Molecular Ecology Resources, Volume 12, pp. 999-1011.

Rüber, L., Kottelat, M., Tan, H. H., Ng, P. K. L. & Britz, R., 2007. Evolution of miniaturization and the phylogenetic position of Paedocypris, comprising the world's smallest vertebrate. BMC Evolutionary Biology, 7(38).

Ruocco, N. L., Lucifora, L. O., Díaz de Astarloa, J. M., Mabragaña, E. & Delpiani, S. M., 2012. Morphology and DNA barcoding reveal a new species of eagle ray from the Southwestern Atlantic: Myliobatis ridens sp. nov. (Chondrichthyes: Myliobatiformes: Myliobatidae). Zoological Studies, 51(6), pp. 862-873.

Ryan, J. R. & Esa, Y. B., 2006. Phylogenetic analysis of Hampala fishes (subfamily Cyprinidae) in Malaysia inferred from partial mitochondrial cytochrome b DNA sequences. Zoological Sciences, 23(10), pp. 893-901.

295

Sati, J., Kumar, R., Sahoo, P. K., Patiyal, R. S., Ali, S. & Barat, A., 2013. Genetic characterization of Golden mahseer (Tor putitora) populations using mitochondrial DNA markers. Mitochondrial DNA, Volume Early Online, pp. 1- 7.

Schneider, S. & Excoffier, L., 1999. Estimation of past demographic parameters from the distribution of pairwise differences when the mutation rates very among sites: application to human mitochondrial DNA. Genetics, Volume 152, pp. 1079- 1089.

Schultheis, A. S., Weigt, L. A. & Hendricks, A. C., 2002. Gene flow, dispersal, and nested clade analysis among population of the stonefly Peltoperla tarteri in the southern Appalachians. Marine Ecology, Volume 11, pp. 317-327.

Schwarz, G. E., 1978. Estimating the dimension of a model. Annals of Statistics, Volume 6, pp. 461-464.

Şengör, A. M. C., 1984. The Cimmeride orogenic system and the tectonics of Eurasia. Geological Society of America Special Paper, p. 195.

Shih, H. T., Yeo, D. C. J. & Ng, P. K. L., 2009. The collision of the Indian plate with Asia: molecular evidence for its impact on the phylogeny of freshwater crabs (Brachyura, Potamidae). Journal of Biogeography , Volume 36, pp. 703-719.

Simmons, M. P. & Ochoterena, H., 2000. Gaps as characters in sequence-based phylogenetic analyses. Systematic Biology, Volume 49, pp. 369-381.

Simpson, G. G., 1977. Too many lines; the limits of the Oriental and Australian zoogeographic regions. Proceedings of the American Philosophical Society, Volume 121, pp. 107-120.

Slatkin, M., 1989. Detecting small amounts of gene flow from phylogenies alleles. Genetics, Volume 121, pp. 609-612.

Slatkin, M., 1993. Isolation by distance in equilibrium and non-equlibrium populations. Evolution , Volume 29, pp. 264-279.

296

Slatkin, M. & Hudson, R. R., 1991. Pairwise comparisons of mitochondrial DNA sequences in stable and exponentially growing populations. Genetics , Volume 129, pp. 555-562.

Slechtova, V., Bohlen, J., Freyhof, J. & Rab, P., 2006. Molecular phylogeny of the Southeast Asian freshwater fish family Botiidae (Teleostei: Cobitoidea) and the origin of polyploidy in their evolution. Molecular Phylogenetics and Evolution, Volume 39, pp. 529-541.

Smith, P. J., Steinke, D., McMillan, P. J., Stewart, A. L., McVeagh, S. M., Diaz de Astarloa, J. M., Welsford, D. & Ward, R. D., 2011. DNA barcoding highlights a cryptic species of grenadier Macrourus in the Southern Ocean. Journal of Fish Biology, 78(1), pp. 355-365.

Smouse, P. E., 1998. To tree or not to tree. Molecular Ecology, Volume 7, pp. 399- 412.

So, N., Maes, G. E. & Volckaert, F. A. M., 2006a. Intra-annual genetic variation in the downstream larval drift of sutchi catfish (Pangasianodon hypophthalmus) in the Mekong river. Biological Journal of the Linnean Society, Volume 89, pp. 719- 728.

So, N., van Houdt, J. & Volckaert, F. A. M., 2006b. Genetic diversity and population history of the migratory catfishes Pangasianodon hypophthalmus and Pangasius bocourti in the Cambodian Mekong River. Fisheries Science , Volume 72, pp. 469-276.

Sodhi, N. S. & Brook, B. W., 2006. Southeast Asian Biodiversity in Crisis. Cambridge, UK: Cambridge UniversityvPress.

Sodhi, N. S., Brook, B. W. & Bradshaw, A. C. J., 2007. Tropical conservation biology. Oxford: Blackwell.

Sodhi, N. S., Koh, L. P, Brook, B. W. & Ng, P. K. L., 2004. Southeast Asian biodiversity: an impending disaster. Trends in Ecology and Evolution, Volume 19, pp. 654-660.

297

Sodhi, N.S., Posa, M., Lee, T., Bickford, D., Koh, L. & Brook, B., 2010. The state and conservation of Southeast Asian biodiversity. Biodiversity and Conservation, Volume 19, pp. 317-328.

Soltis, D. E., Morris, A. B., Mclachlan, J. S., Manos, P. S. & Soltis, P. S., 2006. Comparative phylogeography of unglaciated eastern North America. Molecular Ecology, Volume 15, pp. 4261-4293.

Sosdian, S. & Rosenthal, Y., 2009. Deep-sea temperature and ice volume changes across the Pliocene-Pleistocene climate transitions. Science , Volume 325, pp. 306-310.

Sriwattanarothai, N., Steinke, D., Ruenwongsa, P., Hanner, R. & Panijpan, B., 2010. Molecular and morphological evidence supports the species status of the Mahachai fighter Betta sp. Mahachai and reveals new species of Betta from Thailand. Journal of Fish Biology , Volume 77, pp. 414-424.

Steindachner, F., 1901. Kükenthal's Ergebnisse einer zoologischen Forschungreise in den Molukken und Borneo. Fische. Abh. Senckenb. Naturforsch. Ges., Volume 25, pp. 409-464.

Steinke, D., Zemlak, T. S., Boutillier, J. A. & Hebert, P. D. N., 2009. DNA barcoding of Pacific Canada’s fishes. Marine Biology, Volume 156, pp. 2641-2647.

Stelbrink, B., 2014. A biogeographic view on Southeast Asia’s history. PhD Thesis. Humboldt University of Berlin.

Stelbrink, B., Albrecht, C., Hall, R. & von Rintelen, T., 2012. The biogeography of Sulawesi revisited: is there evidence for a vicariant origin of taxa on Wallace’s “anomalous island”? Evolution, Volume 66, pp. 2252-2271.

Stephens, M. & Scheet, P., 2005. Accounting for decay of linkage disequilibrium in haplotype inference and missing-data imputation. American Journal of Human Genetics, Volume 76, pp. 449-462.

Stephens, M., Smith, N. J. & Donnelly, P., 2001. A new statistical method for haplotype reconstruction from population data. American Journal of Human Genetics, Volume 68, pp. 978-989.

298

Stoeckle, M., 2003. Taxonomy, DNA, and the barcode of life. BioScience, Volume 53, pp. 2-3.

Su, Y. C., Chang, Y. H., Lee, S. C. & Tso, I. M., 2007. Phylogeography of the giant wood spider (Nephila pilipes, Araneae) from Asian-Australian regions. Journal of Biogeography, Volume 34, pp. 177-191.

Sulaiman, Z. H. & Mayden, R. L., 2012. Cypriniformes of Borneo (Actinopterygii, Otophysi): An extraordinary fauna for integrated studies on diversity, systematics, evolution, ecology, and conservation. Zootaxa , Volume 3586, pp. 359-376.

Tajima, F., 1983. Evolutionary relationship of DNA sequences to finite populations. Genetics , Volume 105, pp. 437-460.

Tajima, F., 1989. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics , Volume 123, pp. 585-595.

Takehana, Y., Naruse, K. & Sakaizumi, M., 2005. Molecular phylogeny of the medeka fishes genus Oryzias (Beloniformes, Adrianichthyidae) based on nuclear and mitochondrial DNA sequences. Molecular Phylogenetics and Evolution, Volume 36, pp. 417-428.

Tamura, K. & Nei, M., 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Molecular Biology and Evolution , Volume 10, pp. 512-526.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Filipski, A. & Kumar, S., 2013. MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Molecular Biology & Evolution, Volume 30, pp. 2725-2729.

Tan, H. H., 2006. The Borneo Suckers. Kota Kinabalu: Natural History Publications (Borneo).

Tan, H. H. & Lim, K. K. P., 2013. Three new species of freshwater halfbeaks (Teleostei: Zenarchopteridae: Hemirhamphodon) from Borneo. Raffles Bulletin of Zoology, Volume Supplement, pp. 7735-747.

299

Tan, H. H. & Ng, P. K. L., 2004. Two new species of freshwater fish (Teleostei: Balitoridae, Osphronemidae) from southern Sarawak. In: H. S. Yong, F. S. P. Ng & E. E. L. Yen, eds. Sarawak Bau, Limestone Biodiversity. Sarawak Museum Journal, Vol. LIX, p. No. 80 (New Series); Special Issue No. 6: 267–284.

Tan, H. H. & Ng, P. K. L., 2005. The fighting fishes (Teleostei:Osphronemidae: genus Betta) of Singapore, Malaysia and Brunei. Raffles Bulletin of Zoology, Supplement(13), pp. 43-99.

Tan, M. P., Jamsari, A. F. J. & Siti Azizah, M. N., 2012. Phylogeographic pattern of the striped snakehead, Channa striata in Sundaland: Ancient river connectivity, geographical and anthropogenic singnatures. PLoS ONE , 7(12), p. e52089.

Tan, M. P., Jamsari, A. F. J., Muchlisin, Z. A. & Siti Azizah, M. N., 2015. Mitochondrial genetic variation and population structure of the striped snakehead, Channa striata in Malaysia and Sumatra, Indonesia. Biochemical Systematics and Ecology , Volume 60, pp. 99-105.

Tchaicka, L., Eizirik, E., De Oliveira, T. G., Cândido, J. F. & Freitas, T. R.O., 2007. Phylogeography and population history of the crab-eating fox (Cerdocyon thous). Molecular Ecology, Volume 16, pp. 819-838.

Too, C. C., Adibah, A. B., Danial Hariz, Z. A. & Siti Azizah, M. N., 2016. Detection of mislabelled seafood products in Malaysia by DNA barcoding: Improving transparency in food market. Food Control, Volume 64, pp. 247-256.

Townsend, A. K., Rimmer, C. C., Latta, S. C. & Lovette, I. J., 2007. Ancient differentiation in the single-island avian radiation of endemic Hispaniolan chat- tanagers (Aves: calyptophilus). Molecular Ecology, Volume 16, pp. 3634-3642.

Tsoi, K. H., Wang, Z. Y. & Chu, K. H., 2005. Genetic divergence between Two Morphologically Similar Varieties of the Kuruma Shrimp Penaeus japonicas. Marine Biology, Volume 147, pp. 367-379.

Valdez-Moreno, M., Ivanova, N. V., Elías-Gutiérrez, M., Contreras-Balderas, S. & Hebert, P. D. N., 2009. Probing diversity in freshwater fishes from Mexico and Guatemala with DNA barcodes. Journal of Fish Biology, Volume 74, pp. 377- 402.

300

Valentini, A., Pompanon, F. & Taberlet, P., 2009. DNA barcoding for ecologists. Trends in Ecology and Evolution, Volume 24, pp. 110-117.

van der Kaars, W. A. & Dam, M. A. C., 1995. A 135,000-year record of vegetational and climatic change from the Bandung area, West Java, Indonesia. Palaeogeography, Palaeoclimatoloy, Palaeoecology, Volume 117, pp. 55-72.

van Houdt, J. K., Hellemans, B. & Volckaert, F. A. M., 2003. Phylogenetic relationships among Palearctic and Nearctic burbot (Lota lota): Pleistocene extinctions and recolonization. Molecular Phylogenetic and Evolution , Volume 29, pp. 599-612.

van Steenis, C. G. G. J., 1995. The delimitation of Malesia and its main plant geographical divisions. Flora Malesiana I, Volume 1, p. 120–125.

Verheyen, E. & Rubber, L., 2000. Conservation of the endemic cichilds of Lake Tanganyika, implication from population level studies based upon mtDNA sequences. Advances in Ecological Research, Volume 31, pp. 539-551.

Vidya, T. N. C., Sukumar, R. & Melnick, D. J., 2009. Range-wide mtDNA phylogeography yields insights into the origins of Asian elephants. Proceeding of the Royal Society of London B: Biological Sciences, Volume 276, pp. 893- 902.

Volckaert, F. A. M., Hänfling, B., Hellemans, B. & Carvalho, G. R., 2002. Timing of the population dynamics of bullhead Cottus gobio (Teleostei: Cottidae) during the Pleistocene. Journal of Evolutionary Biology, Volume 15, pp. 930-944.

von der Heyden, S., Lipinski, M. R. & Matthee, C. A., 2007. Mitochondrial DNA analyses of the Cape hakes reveal an expanding, panmictic population for Merluccius capensis and population structuring for mature fish in Merluccius paradoxus. Molecular Phylogenetic and Evolution, 42(2), pp. 517-527.

von Rintelen, T. & Glaubrecht, M., 2006. Rapid evolution of sessility in an endemic species flock of the freshwater bivalve Corbicula from ancient lakes on Sulawesi, Indonesia. Biology Letters, Volume 2, pp. 73-77.

301

Voris, H. K., 2000. Maps of Pleistocene sea levels in Southeast Asia: shorelines, river systems and time durations. Journal of Biogeography, Volume 27, pp. 1153- 1167.

Wallace, A. R., 1860. On the zoological geography of the Malay Archipelago. Journal of the Proceedings of the Linnean Society, Volume 4, pp. 172-184.

Wallace, A. R., 1863. On the physical geography of the Malay Archipelago. Journal of the Royal Geographical Society of London, Volume 33, pp. 217-234.

Wang, J. P., Hsu, K. C. & Chiang, T. Y., 2000. Mitochondrial DNA phylogeography of Acrossocheilus paradoxus (Cyprinidae) in Taiwan. Molecular Ecology, Volume 9, pp. 1483-1494.

Ward, R. D., 2000. Genetics in fisheries management. Hydrobiologia, Volume 420, pp. 191-201.

Ward, R. D., 2009. DNA barcode divergence among species and genera of birds and fishes. Molecular Ecology Resources, Volume 9, pp. 1077-1085.

Ward, R. D., Hanner, R. H. & Hebert, P. D. N., 2009. The campaign to DNA barcode all fishes, FISH-BOL. Journal of Fish Biology, Volume 74, pp. 329-356.

Ward, R. D., Woodwark, M. & Skibinski, D. O. F., 1994. A comparison of genetic diversity levels in marine, freshwater and anadromous fishes. Journal of Fish Biology, Volume 44, pp. 213-232.

Ward, R. D., Zemlak, T. S., Innes, B. H., Last, P. R. & Hebert, P.D.N., 2005. DNA barcoding Australia’s fish species. Philosophical Transactions of the Royal Society Biological Sciences , Volume 360, pp. 1847-1857.

Waters, J. M., Craw, D., Youngson, J. H. & Wallis, G. P., 2001. Genes meet geology: fish phylogeographic pattern reflects ancient, rather than modern, drainage connections. Evolution, Volume 55, pp. 1844-1851.

Watterson, G. A., 1975. On the number of segregating sites in genetic models without recombination. Theoretical Population Biology , Volume 7, pp. 256-276.

302

Whittaker, R. J., Araújo, M. B., Jepson, P., Ladle, R. J., Watson, J. E. M. & Willis, K., 2005. Conservation biogeography: assessment and prospect. Diversity and Distributions, Volume 11, pp. 3-23.

Whitten, T., Peter P., van Dijk., Curran, L., Meijaard, E., Wood, P., Supriatna, J. & Zellis, S., 2004. Sundaland. In: P. R. Gil, et al. eds. Hotspots Revisited. Mexico: CEMEX, pp. 164-172.

Wickman, A., 1981. Net in hand in Borneo. Aquar. Pondkeeper, 46(6), pp. 40-46.

Wiens, J. J. & Donoghue, M. J., 2004. Historical biogeography, ecology and species richness. Trends in Ecology and Evolution, Volume 19, pp. 639-644.

Wikramanayake, E., Dinerstein, E., Loucks, C., Olson, D., Morrison, J., Lamoreux, J., McKnight, M. & Hedao, P., 2002. Terrestrial ecoregions of the Indo-Pacific: a conservation assessment. Washington: Island Press.

Wilcove, D. S. & Koh, L. P., 2010. Addressing the threats to biodiversity from oil palm agriculture. Biodiversity and Conservation, 19(4), pp. 999-1007.

Will, K. W., Mishler, B. D. & Wheeler, Q. D., 2005. The perils of DNA barcoding and the need for integrative taxonomy. Systematic Biology, Volume 54, pp. 844-851.

Witt, J. D. S., Blinn, D. W. & Hebert, P. D. N., 2003. The recent evolutionary origin of the phenotypically novel amphipod Hyalella montezuma offers and ecological explanation for morphological stasis in a closely allied species complex. Molecular Ecology, Volume 12, pp. 405-413.

Wong, E. H. K. & Hanner, R. H., 2008. DNA barcoding detects market subtitution in north american seafood,. Food Research International, Volume 41, pp. 828-837.

Wong, K. M., 2011. A biogeographic history of Southeast Asian rainforests. In: R. Wickneswari & C. Cannon, eds. Managing the future of Southeast Asia’s valuable tropical rainforests. Dordrecht: Springer Netherlands, pp. 21-55.

Woodruff, D. S., 2010. Biogeography and conservation in Southeast Asia: How 2.7 million years of repeated environmental fluctuations affect today's patterns and

303

the future of the remaining refugial-phase biodiversity. Biodiversity and Conservation , 19(4), pp. 919-941.

Woodruff, D. S. & Turner, L. M., 2009. The Indochinese–Sundaic zoogeographic transition: a description of terrestrial mammal species distributions. Journal of Biogeography , Volume 36, pp. 803-821.

Wright, S., 1943. Isolation by distance. Genetics , Volume 28, pp. 114-138.

Wright, S. J., Muller-Landau, H. C. & Schipper, J., 2009. The future of tropical species on a warmer planet. Conservation Biology, Volume 23, pp. 1418-1426.

Xia, X. & Xie, Z., 2001. DAMBE: software package for data analysis in molecular biology and evolution. Journal of Heredity, Volume 92, pp. 371-373.

Yamamoto, S., Morita, K., Koizumi, I. & Maekawa, K., 2004. Genetic differentiation of white-spotted charr (Salvelinus leucomaenis) populations after habitat fragmentation: Spatial–temporal changes in gene frequencies. Conservation Genetics , Volume 5, pp. 529-538.

Yap, S. Y., 2002. On the distributional patterns of Southeast-East Asian freshwater fish and their history. Journal of Biogeography, Volume 29, p. 1187–1199.

Yeo, D. C. J., Shih, H. T., Meier, R. & Ng, P. K. L., 2007. Phylogeny and biogeography of the freshwater crab genus Johora (Crustacea, Brachyura, Potamidae) from the Malay Peninsula, and the origins of its insular fauna. Zoologica Scripta, Volume 36, pp. 255-269.

Zaki, S. A. H., Jordan, W. C., Reichard, M., Przybylski, M. & Smith, C., 2008. A morphological and genetic analysis of the European bitterling species complex. Biological Journal of the Linnean Society , Volume 95, pp. 337-347.

Zardoya, R. & Doadrio, I., 1999. Molecular evidence on the evolutionary and biogeographical patterns of European cyprinids. Journal of Molecular Evolution, Volume 49, pp. 227-237.

304

Zemlak, T. S., Habit, E. M., Walde, S. J., Battini, M. A., Adams, E. D. M. & Ruzzante, D. E., 2008. Across the southern Andes on fin: glacial refugia, drainage reversals and a secondary contact zone revealed by the phylogeographical signal of Galaxias platei in Patagonia. Molecular Ecology, Volume 17, pp. 5049-5061.

Zhang, D. X. & Hewitt, G. M., 2003. Nuclear DNA analyses in genetic studies of populations: practice, problems and prospects. Molecular Ecology, 12(3), pp. 563-584.

Zhang, J. B. & Hanner, R., 2011. DNA barcoding is a useful tool for the identification of marine fishes from Japan. Biochemical Systematics Ecology, Volume 39, pp. 31-42.

Zhao, L., Zhang, J., Liu, Z., Funk, S. M., Wei, F., Xu, M. & Li, M., 2008. Complex population genetic and demographic history of the Salangid, Neosalanx taihuensis, based on cytochrome b sequences. BMC Evolutionary Biology, Volume 8, p. 201.

305

APPENDICES

Appendix A: Geographical coordinates of sampling locations and voucher specimen repository. Sample label Species Country State/Province Latitude Longitude Institution Storing KWK H. pogonognathus Malaysia Perlis N: 6° 40' 39.2'' E: 100° 11' 5.1'' Universiti Sains Malaysia TE H. pogonognathus Malaysia Kedah N: 5° 56' 17.5'' E: 100° 26' 32.5'' Universiti Sains Malaysia TB H. pogonognathus Malaysia Penang N: 5° 26' 56.3'' E: 100° 12' 53.9'' Universiti Sains Malaysia BP H. pogonognathus Malaysia Penang N: 5° 9' 42.8'' E: 100° 32' 50.4'' Universiti Sains Malaysia PTG H. pogonognathus Malaysia Perak N: 5° 4' 46.4'' E: 100° 43' 20.1'' Universiti Sains Malaysia SK H. pogonognathus Malaysia Perak N: 4° 1' 5.7'' E: 101° 21' 57.5'' Universiti Sains Malaysia PJG H. pogonognathus Malaysia Selangor N: 3° 39' 12.4'' E: 101° 18' 1.9'' Universiti Sains Malaysia DM H. pogonognathus Malaysia Selangor N: 3° 11' 4.3'' E: 101° 35' 30'' Universiti Sains Malaysia SU H. pogonognathus Malaysia N. Sembilan N: 2° 54' 54'' E: 102° 15' 35.1'' Universiti Sains Malaysia SOM H. pogonognathus Malaysia Pahang N: 3° 56' 46.48' E: 102° 15' 28.8'' Universiti Sains Malaysia KS H. pogonognathus Malaysia Pahang N: 3° 28.850' E: 102° 57.335' Universiti Sains Malaysia PI H. pogonognathus Malaysia Pahang N: 3° 1' 37.3'' E: 102° 39' 22.8'' Universiti Sains Malaysia JP H. pogonognathus Malaysia Kelantan N: 5° 47' 47.4'' E: 102° 20' 15.8'' Universiti Sains Malaysia LB H. pogonognathus Malaysia Kelantan N: 5° 38' 25.5'' E: 102° 35' 35.3'' Universiti Sains Malaysia S H. pogonognathus Malaysia Terengganu N: 4° 58' 7'' E: 102° 57' 21.4'' Universiti Sains Malaysia RA H. pogonognathus Malaysia Terengganu N: 4° 52' 50'' E: 103° 23' 5'' Universiti Sains Malaysia PA H. pogonognathus Malaysia Terengganu N: 4° 17' 18.0'' E: 103° 21' 0.5'' Universiti Sains Malaysia PT H. pogonognathus Malaysia Johor N: 1° 47' 14.7'' E: 103° 56' 33.7'' Universiti Sains Malaysia ST H. pogonognathus Indonesia Riau N: 0° 25' 30'' E: 101° 21' 41'' Syiah Kuala University SG H. pogonognathus Indonesia Riau N: 0° 09' 48'' E: 101° 18' 53'' Syiah Kuala University STK H. pogonognathus Indonesia Riau N: 0° 08' 49'' E: 101° 15' 40'' Syiah Kuala University SKR H. pogonognathus Indonesia Indragiri Hulu S: 0° 13' 1'' E: 102° 14' 22'' Syiah Kuala University

306

Appendix A: continue… Sample label Species Country State/Province Latitude Longitude Institution Storing SBG H. pogonognathus Indonesia Indragiri Hulu S: 0° 22' 54'' E: 102° 23' 30'' Syiah Kuala University JBI H. pogonognathus Indonesia Jambi S: 1° 45' 54'' E: 103° 33' 21'' Syiah Kuala University JMP H. pogonognathus Indonesia Palembang S: 2° 15' 17'' E: 103° 48' 48'' Syiah Kuala University RBT H. pogonognathus Indonesia Banga Island S: 1° 55' 46'' E: 105° 16' 39'' Syiah Kuala University MRW H. pogonognathus Indonesia Banga Island S: 1° 57' 35'' E: 106° 2' 51'' Syiah Kuala University PLG H. pogonognathus Indonesia Banga Island S: 2° 6' 56'' E: 106° 0' 38'' Syiah Kuala University KA H. pogonognathus Indonesia Banga Island S: 2° 29' 29'' E: 106° 23' 30'' Syiah Kuala University SC H. byssus Malaysia Sarawak N: 1° 40' 7.7'' E: 109° 50' 18.5'' Universiti Sains Malaysia SST H. byssus Malaysia Sarawak N: 1° 28' 43.8'' E: 110° 0' 8.1'' Universiti Sains Malaysia SD H. byssus Malaysia Sarawak N: 1° 20' 45.5'' E: 110° 2' 37.1'' Universiti Sains Malaysia TR H. byssus Malaysia Sarawak N: 1° 4' 14.1'' E: 111° 17' 32.5'' Universiti Sains Malaysia SP H. byssus Malaysia Sarawak N: 1° 36' 26.0'' E: 111° 34' 29.7'' Universiti Sains Malaysia LS H. kuekenthali Malaysia Sarawak N: 1° 58' 50.6'' E: 111° 44' 52.8'' Universiti Sains Malaysia LG H. kuekenthali Malaysia Sarawak N: 2° 13' 21.6'' E: 111° 48' 17.4'' Universiti Sains Malaysia BK H. kuekenthali Malaysia Sarawak N: 2° 19' 37.5'' E: 111° 57' 25.9'' Universiti Sains Malaysia MK H. kuekenthali Malaysia Sarawak N: 2° 54' 29.7'' E: 112° 19' 33.5'' Universiti Sains Malaysia TT H. kuekenthali Malaysia Sarawak N: 2° 41' 7.9'' E: 112° 40' 30.3'' Universiti Sains Malaysia KMD H. kuekenthali Malaysia Sarawak N: 3° 19' 30.7'' E: 113° 28' 14.1'' Universiti Sains Malaysia LK H. kuekenthali Malaysia Sarawak N: 4° 14' 19'' E: 114° 3' 43'' Universiti Sains Malaysia LL H. kuekenthali Malaysia Sarawak N: 3° 46' 9.1'' E: 114° 25' 28.3'' Universiti Sains Malaysia KJI H. kuekenthali Malaysia Sarawak N: 3° 41' 32.8'' E: 114° 27' 20.8'' Universiti Sains Malaysia LBI H. kuekenthali Malaysia Sarawak N: 4° 10' 03.2' E: 114° 27' 53.1'' Universiti Sains Malaysia LMB H. kuekenthali Malaysia Sarawak N: 4° 43' 13.3'' E: 114° 59' 47.1'' Universiti Sains Malaysia

307

Appendix B: DNA extraction method. (Modified from Aljanabi & Martinez, 1997)

DNA EXTRACTION FROM FIN CLIPS

Extraction buffer 50ml

30mM Tris 10mM EDTA 1% SDS

Tris stock 1M dilution 33 use 1.5ml EDTA stock 0.5M dilution 50 use 1ml SDS stock 10% dilution 10 use 5ml Add 42.5ml H20

Procedure

• Cut fin clip from fish using flamed razor blade • Add 300µl extraction buffer, 10µl Proteinase K (stored at RT) • Incubate at 55℃ overnight (or a few hours) • Add 100µl 5M NaCl • Spin 5 min at 13000rpm on bench centri • Collect SN (200µl/tube) in strips/plate • Add 2 vol (400µl) of 100% ice cold ethanol • Invert tube to mix • Incubate 20min at -20℃ • Spin 30min at 13000rpm on bench centri • Pour off EtOH, tap on blue roll • Add 1ml of 70% EtOH (wash) • Spin 10min at 13000rpm on bench centri • Pour off EtOH, tap on blue roll • Air dry 30min at RT (inverted on blue roll) • Resuspend in 100µl dH2O • Store -20℃

308

Appendix C: Primers used to amplify DNA regions for Hemirhamphodon spp.

Primer Sequence DNA region Reference F2: 5’- TCG ACT AAT CAT AAA GAT ATC mtDNA Ward, et al. (2005) GGC AC -3’ COI

R2: 5’- ACT TCA GGG TGA CCG AAG AAT CAG AA -3’

L14504: 5’- GCC AAW GCT GCW GAA TAM mtDNA Miya & Nashida, GCA AA -3’ Cyt b (2000)

Cyt-b3-3: 5’- GGC AAA TAG GAA Rta TCA Lovejoy, (2000) TTC -3’

Hp5-F: 5’- CCC CAT GAT TGA TTA GCT TG - nDNA de Bruyn, et al. 3’ (SCNP) (2010) Hp5 Hp5-R: 5’- CAC GAT TGT TGC AAC TCA CTC T -3’

Hp54-F:5’- ATT GGC CAA AAC CAT AGC AG nDNA de Bruyn, et al. -3’ (SCNP) (2010) HP54 Hp54-R: 5’- GCC ACA CCT TTT TCC CTT TT - 3’

309

Appendix D: PCR conditions used to amplify target DNA fragments.

DNA Product region Reaction Mix (25μL volume) Cycling conditions size COI 2.5μL 10X PCR Reaction Buffer 94oC - 2min ≈ 700 bp 1.0μL 50mM MgCl2 35 X 94 oC - 30s 1.0μL 10mM dNTPs 55 oC - 30s 0.25μM of each primer 72 oC - 1min 0.1μL of Taq DNA Polymerase 72 oC - 10min 1.0μL genomic DNA (50-100ng) 10 oC - hold 18.9μL ddH2O

Cyt b 2.5μL 10X PCR Reaction Buffer 94oC - 2min ≈ 900 bp 2.0μL 50mM MgCl2 35 X 94 oC – 1min 1.0μL 10mM dNTPs 50 oC – 1min 0.25μM of each primer 72 oC – 1.5min 0.1μL of Taq DNA Polymerase o 1.0μL genomic DNA (50-100ng) 72 C - 10min o 17.9μL ddH2O 10 C - hold

Hp5 and 2.5μL 10X PCR Reaction Buffer A standard touchdown ≈ 300 bp Hp54 2.0μL 50mM MgCl2 program used ≈ 300 bp 1.0μL 10mM dNTPs 95oC - 2min 0.25μM of each primer 10 X 95 oC – 35s 0.1μL of Taq DNA Polymerase 63 oC – 35s 1.0μL genomic DNA (50-100ng) (-0.5oC/ cycle) 17.9μL ddH2O 72 oC – 1min

10 X 95 oC – 35s 58 oC – 35s 72 oC – 1min

15 X 95 oC – 35s 52 oC – 35s (-0.5oC/ cycle) 72 oC – 1min

72 oC - 10min 10 oC – hold

310

Appendix E: Pairwise comparisons of the COI gene based on K2P distance among sampling locations of Hemirhamphodon species.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 1 Hp_TE 2 Hp_TB 0.002 3 Hp_BP 0.002 0.000 4 H_Pen 0.003 0.002 0.002 5 Hp_PI 0.005 0.003 0.003 0.005 6 Hp_PT 0.006 0.004 0.004 0.006 0.004 7 Hp_PTG 0.007 0.005 0.005 0.007 0.005 0.006 8 H_LR52 0.007 0.005 0.005 0.007 0.005 0.006 0.000 9 H_LR54 0.007 0.005 0.005 0.007 0.005 0.005 0.006 0.006 10 Hp_PJG 0.007 0.005 0.005 0.007 0.005 0.006 0.003 0.004 0.004 11 Hp_S 0.007 0.005 0.005 0.007 0.002 0.006 0.007 0.007 0.007 0.007 12 Hp_SOM 0.007 0.005 0.005 0.007 0.005 0.003 0.007 0.007 0.005 0.007 0.007 13 Hp_PA 0.007 0.005 0.005 0.007 0.002 0.006 0.007 0.007 0.007 0.007 0.003 0.007 14 Hp_RA 0.007 0.005 0.005 0.007 0.002 0.006 0.007 0.007 0.007 0.007 0.003 0.007 0.003 15 Hp_KJ 0.008 0.007 0.007 0.008 0.007 0.004 0.008 0.009 0.005 0.008 0.008 0.005 0.008 0.008 16 Hp_7140 0.008 0.007 0.007 0.008 0.007 0.004 0.008 0.009 0.005 0.008 0.008 0.005 0.008 0.008 0.000 17 H_SGK 0.008 0.007 0.007 0.008 0.007 0.004 0.008 0.009 0.005 0.008 0.008 0.005 0.008 0.008 0.000 0.000 18 Hp_SK 0.010 0.008 0.008 0.010 0.008 0.009 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.011 0.011 0.011 19 Hp_ST 0.010 0.009 0.009 0.010 0.008 0.010 0.009 0.010 0.010 0.009 0.009 0.010 0.009 0.009 0.012 0.012 0.012 0.004 20 Hp_SU 0.010 0.008 0.008 0.010 0.008 0.006 0.010 0.010 0.006 0.010 0.010 0.007 0.010 0.010 0.002 0.002 0.002 0.013 0.014 21 Hp_SKR 0.007 0.005 0.005 0.007 0.002 0.006 0.007 0.007 0.007 0.007 0.003 0.007 0.003 0.003 0.008 0.008 0.008 0.007 0.006 0.010 22 Hp_DM 0.011 0.010 0.010 0.011 0.010 0.011 0.011 0.012 0.011 0.011 0.011 0.011 0.011 0.011 0.013 0.013 0.013 0.002 0.005 0.015 0.008 23 H_Pek 0.015 0.014 0.014 0.014 0.012 0.015 0.014 0.014 0.015 0.013 0.014 0.015 0.014 0.014 0.017 0.017 0.017 0.010 0.012 0.019 0.011 0.012 24 Hp_SG 0.022 0.020 0.020 0.022 0.020 0.021 0.022 0.022 0.020 0.018 0.022 0.022 0.021 0.022 0.023 0.023 0.023 0.021 0.023 0.025 0.022 0.023 0.025 25 H_Sel 0.022 0.020 0.020 0.022 0.020 0.021 0.022 0.022 0.020 0.018 0.022 0.022 0.021 0.022 0.023 0.023 0.023 0.021 0.022 0.025 0.018 0.023 0.024 0.007 26 Hp_7126 0.022 0.020 0.020 0.022 0.020 0.021 0.022 0.022 0.020 0.018 0.022 0.022 0.021 0.022 0.023 0.023 0.023 0.021 0.022 0.025 0.018 0.023 0.024 0.007 0.000 27 Hp_STK 0.023 0.021 0.021 0.023 0.021 0.023 0.023 0.024 0.022 0.020 0.023 0.023 0.023 0.023 0.025 0.025 0.025 0.023 0.025 0.027 0.023 0.025 0.027 0.004 0.008 0.008 28 Hp_KS 0.024 0.022 0.022 0.024 0.022 0.023 0.021 0.021 0.020 0.017 0.024 0.024 0.024 0.024 0.022 0.022 0.022 0.027 0.027 0.024 0.024 0.029 0.028 0.009 0.009 0.009 0.010 29 Hp_KWK 0.025 0.023 0.023 0.025 0.023 0.025 0.022 0.021 0.023 0.018 0.025 0.025 0.025 0.025 0.027 0.027 0.027 0.029 0.028 0.029 0.025 0.030 0.030 0.018 0.018 0.018 0.020 0.017 30 Hp_7125 0.067 0.065 0.065 0.067 0.062 0.066 0.063 0.064 0.063 0.060 0.063 0.067 0.063 0.063 0.067 0.067 0.067 0.062 0.062 0.065 0.060 0.062 0.064 0.058 0.058 0.058 0.060 0.061 0.064 31 Hp_JMP 0.073 0.071 0.071 0.073 0.071 0.072 0.069 0.069 0.069 0.065 0.073 0.069 0.072 0.073 0.072 0.072 0.072 0.067 0.068 0.074 0.069 0.069 0.071 0.057 0.060 0.060 0.062 0.062 0.065 0.046 32 Hp_7127 0.076 0.074 0.074 0.076 0.071 0.074 0.073 0.073 0.072 0.069 0.073 0.073 0.072 0.073 0.076 0.076 0.076 0.071 0.071 0.078 0.069 0.071 0.074 0.071 0.071 0.071 0.073 0.073 0.076 0.033 0.044 33 Hp_JBI 0.080 0.078 0.078 0.080 0.074 0.078 0.076 0.076 0.076 0.073 0.076 0.076 0.076 0.076 0.080 0.080 0.080 0.074 0.072 0.082 0.073 0.074 0.078 0.074 0.074 0.074 0.076 0.077 0.076 0.037 0.046 0.017 34 Hp_MRW 0.080 0.078 0.078 0.080 0.078 0.079 0.076 0.077 0.076 0.073 0.080 0.076 0.080 0.080 0.080 0.080 0.080 0.074 0.073 0.078 0.076 0.076 0.078 0.062 0.062 0.062 0.063 0.064 0.075 0.067 0.047 0.062 0.064 35 Hp_SBG 0.082 0.080 0.080 0.082 0.076 0.080 0.078 0.077 0.078 0.074 0.078 0.078 0.078 0.078 0.082 0.082 0.082 0.076 0.076 0.084 0.074 0.076 0.080 0.076 0.076 0.076 0.078 0.079 0.078 0.035 0.047 0.015 0.008 0.069

311

Appendix E: continue…

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 36 Hp_PLG 0.082 0.080 0.080 0.082 0.080 0.081 0.078 0.079 0.078 0.074 0.082 0.078 0.082 0.082 0.082 0.082 0.082 0.076 0.075 0.080 0.078 0.078 0.080 0.063 0.063 0.063 0.065 0.066 0.077 0.068 0.047 0.066 0.066 0.007 37 Hp_KA 0.082 0.080 0.080 0.082 0.080 0.081 0.078 0.079 0.078 0.074 0.082 0.078 0.082 0.082 0.082 0.082 0.082 0.076 0.075 0.080 0.078 0.078 0.080 0.063 0.063 0.063 0.065 0.066 0.077 0.068 0.047 0.067 0.067 0.007 38 H_Jam 0.082 0.080 0.080 0.081 0.076 0.079 0.078 0.077 0.078 0.074 0.078 0.078 0.078 0.078 0.081 0.081 0.081 0.076 0.074 0.083 0.074 0.076 0.079 0.076 0.076 0.076 0.078 0.078 0.078 0.038 0.047 0.018 0.003 0.065 39 Hp_RBT 0.095 0.093 0.093 0.095 0.093 0.093 0.091 0.092 0.091 0.087 0.095 0.091 0.093 0.095 0.095 0.095 0.095 0.089 0.089 0.097 0.091 0.091 0.094 0.082 0.082 0.082 0.084 0.084 0.091 0.069 0.058 0.069 0.074 0.082 40 H_LR53 0.110 0.108 0.108 0.110 0.108 0.105 0.106 0.106 0.103 0.102 0.110 0.104 0.110 0.110 0.102 0.102 0.102 0.106 0.105 0.104 0.106 0.106 0.111 0.109 0.108 0.108 0.112 0.107 0.102 0.101 0.111 0.113 0.113 0.124 41 Hp_JP 0.110 0.108 0.108 0.110 0.108 0.105 0.106 0.106 0.103 0.102 0.110 0.104 0.110 0.110 0.102 0.102 0.102 0.106 0.105 0.104 0.106 0.106 0.111 0.109 0.108 0.108 0.112 0.107 0.102 0.101 0.111 0.113 0.113 0.124 42 Hp_LB 0.114 0.112 0.112 0.114 0.112 0.109 0.110 0.111 0.107 0.106 0.114 0.106 0.114 0.114 0.106 0.106 0.106 0.110 0.109 0.108 0.110 0.110 0.115 0.109 0.108 0.108 0.112 0.107 0.102 0.105 0.109 0.111 0.115 0.118 43 H_LR697 0.126 0.128 0.128 0.130 0.128 0.129 0.126 0.125 0.125 0.122 0.130 0.128 0.130 0.128 0.128 0.128 0.128 0.122 0.121 0.126 0.126 0.122 0.129 0.127 0.120 0.120 0.126 0.123 0.116 0.123 0.126 0.131 0.114 0.122 44 H_Sar 0.121 0.123 0.123 0.125 0.123 0.124 0.121 0.121 0.120 0.117 0.125 0.123 0.125 0.123 0.123 0.123 0.123 0.119 0.121 0.121 0.121 0.119 0.126 0.126 0.119 0.119 0.126 0.123 0.115 0.124 0.120 0.128 0.120 0.122 45 Hb_SC 0.122 0.124 0.124 0.126 0.124 0.125 0.122 0.121 0.121 0.118 0.126 0.124 0.126 0.124 0.124 0.124 0.124 0.120 0.121 0.122 0.122 0.120 0.126 0.127 0.120 0.120 0.126 0.123 0.116 0.125 0.120 0.129 0.120 0.120 46 Hb_SST 0.122 0.124 0.124 0.126 0.124 0.125 0.122 0.121 0.121 0.118 0.126 0.124 0.126 0.124 0.124 0.124 0.124 0.120 0.121 0.122 0.122 0.120 0.126 0.127 0.120 0.120 0.126 0.123 0.116 0.125 0.120 0.129 0.120 0.120 47 Hk_7136 0.131 0.133 0.133 0.135 0.133 0.135 0.131 0.131 0.131 0.129 0.131 0.133 0.135 0.133 0.133 0.133 0.133 0.129 0.131 0.131 0.131 0.131 0.132 0.134 0.127 0.127 0.134 0.132 0.119 0.132 0.136 0.138 0.134 0.140 48 Hb_SD 0.133 0.135 0.135 0.136 0.135 0.137 0.133 0.133 0.133 0.131 0.133 0.135 0.137 0.135 0.135 0.135 0.135 0.132 0.133 0.133 0.133 0.133 0.133 0.136 0.129 0.129 0.136 0.134 0.125 0.138 0.142 0.144 0.140 0.138 49 Hb_TR 0.134 0.136 0.136 0.138 0.136 0.133 0.134 0.134 0.131 0.130 0.134 0.132 0.138 0.136 0.132 0.132 0.132 0.132 0.134 0.134 0.134 0.132 0.138 0.139 0.132 0.132 0.139 0.137 0.128 0.121 0.128 0.122 0.118 0.133 50 Hb_SP 0.133 0.135 0.135 0.137 0.135 0.136 0.133 0.132 0.132 0.129 0.133 0.135 0.137 0.135 0.135 0.135 0.135 0.131 0.132 0.133 0.133 0.131 0.136 0.134 0.127 0.127 0.133 0.132 0.119 0.123 0.127 0.127 0.123 0.127 51 Hk_LS 0.112 0.114 0.114 0.116 0.114 0.115 0.112 0.112 0.112 0.108 0.112 0.114 0.116 0.116 0.116 0.116 0.116 0.110 0.108 0.114 0.112 0.109 0.116 0.115 0.110 0.110 0.119 0.115 0.110 0.106 0.104 0.108 0.108 0.102 52 Hk_LG 0.112 0.114 0.114 0.116 0.114 0.115 0.112 0.111 0.113 0.112 0.112 0.114 0.116 0.116 0.116 0.116 0.116 0.110 0.109 0.114 0.112 0.110 0.117 0.119 0.114 0.114 0.124 0.119 0.114 0.117 0.114 0.118 0.114 0.112 53 Hk_BK 0.122 0.124 0.124 0.126 0.124 0.124 0.122 0.121 0.122 0.118 0.122 0.124 0.126 0.126 0.126 0.126 0.126 0.120 0.118 0.124 0.122 0.120 0.125 0.125 0.120 0.120 0.130 0.125 0.120 0.116 0.118 0.122 0.122 0.116 54 H_LR699 0.117 0.119 0.119 0.121 0.119 0.119 0.117 0.117 0.118 0.117 0.117 0.119 0.121 0.121 0.121 0.121 0.121 0.115 0.116 0.119 0.117 0.115 0.121 0.124 0.119 0.119 0.129 0.122 0.125 0.120 0.122 0.126 0.130 0.126 55 Hk_MK 0.116 0.118 0.118 0.120 0.118 0.118 0.116 0.116 0.118 0.116 0.116 0.118 0.120 0.120 0.120 0.120 0.120 0.114 0.115 0.118 0.116 0.114 0.120 0.123 0.118 0.118 0.128 0.121 0.124 0.120 0.122 0.126 0.130 0.126 56 Hk_TT 0.122 0.124 0.124 0.126 0.124 0.125 0.122 0.121 0.122 0.118 0.122 0.124 0.126 0.126 0.126 0.126 0.126 0.120 0.119 0.124 0.122 0.120 0.124 0.121 0.116 0.116 0.126 0.121 0.118 0.109 0.106 0.114 0.110 0.104 57 Hk_KMD 0.134 0.136 0.136 0.138 0.140 0.142 0.138 0.138 0.138 0.134 0.138 0.140 0.142 0.142 0.142 0.142 0.142 0.136 0.135 0.140 0.138 0.136 0.140 0.145 0.136 0.136 0.147 0.140 0.134 0.128 0.134 0.134 0.134 0.134 58 Hk_LK 0.114 0.116 0.116 0.118 0.120 0.121 0.118 0.117 0.118 0.114 0.122 0.120 0.122 0.122 0.122 0.122 0.122 0.120 0.117 0.120 0.118 0.120 0.123 0.121 0.116 0.116 0.126 0.121 0.116 0.120 0.116 0.122 0.118 0.122 59 Hk_LL 0.114 0.116 0.116 0.118 0.116 0.117 0.114 0.114 0.114 0.110 0.118 0.116 0.118 0.118 0.118 0.118 0.118 0.116 0.114 0.116 0.114 0.116 0.119 0.118 0.112 0.112 0.122 0.117 0.112 0.114 0.114 0.122 0.112 0.114 60 Hk_KJI 0.114 0.116 0.116 0.118 0.116 0.117 0.114 0.114 0.114 0.110 0.118 0.116 0.118 0.118 0.118 0.118 0.118 0.116 0.114 0.116 0.114 0.116 0.119 0.118 0.112 0.112 0.122 0.117 0.112 0.114 0.114 0.122 0.112 0.114 61 Hk_7138 0.108 0.110 0.110 0.112 0.114 0.115 0.112 0.112 0.112 0.108 0.116 0.114 0.116 0.116 0.116 0.116 0.116 0.114 0.112 0.114 0.112 0.114 0.117 0.116 0.110 0.110 0.120 0.115 0.110 0.112 0.112 0.120 0.110 0.112 62 Hk_LBI 0.108 0.110 0.110 0.112 0.114 0.115 0.112 0.112 0.112 0.108 0.116 0.114 0.116 0.116 0.116 0.116 0.116 0.114 0.112 0.114 0.112 0.114 0.117 0.116 0.110 0.110 0.120 0.115 0.110 0.112 0.112 0.120 0.110 0.112 63 Hk_LMB 0.114 0.116 0.116 0.118 0.116 0.117 0.114 0.114 0.114 0.110 0.114 0.116 0.118 0.118 0.118 0.118 0.118 0.116 0.114 0.116 0.114 0.116 0.121 0.120 0.116 0.116 0.122 0.117 0.112 0.115 0.108 0.116 0.112 0.116 64 H_phaiosoma 0.140 0.142 0.142 0.144 0.142 0.143 0.140 0.140 0.140 0.136 0.144 0.144 0.142 0.142 0.144 0.144 0.144 0.142 0.141 0.142 0.144 0.142 0.147 0.137 0.142 0.142 0.134 0.133 0.134 0.128 0.149 0.141 0.147 0.146 65 H_chrysopunctatus0.173 0.171 0.171 0.172 0.171 0.170 0.168 0.168 0.167 0.164 0.168 0.173 0.170 0.173 0.169 0.169 0.169 0.164 0.166 0.171 0.168 0.166 0.168 0.166 0.162 0.162 0.168 0.167 0.162 0.165 0.174 0.175 0.177 0.178 66 H_tengah 0.147 0.149 0.149 0.151 0.153 0.154 0.147 0.146 0.150 0.147 0.155 0.151 0.155 0.155 0.153 0.153 0.153 0.145 0.148 0.155 0.151 0.147 0.152 0.140 0.143 0.143 0.145 0.144 0.145 0.155 0.145 0.153 0.145 0.155 67 Outgroup 0.182 0.179 0.179 0.182 0.184 0.185 0.182 0.182 0.181 0.177 0.182 0.182 0.184 0.184 0.184 0.184 0.184 0.177 0.176 0.181 0.182 0.179 0.182 0.173 0.177 0.177 0.175 0.178 0.180 0.203 0.194 0.212 0.210 0.190

312

Appendix E: continue…

35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 36 Hp_PLG 0.071 37 Hp_KA 0.073 0.003 38 H_Jam 0.010 0.067 0.069 39 Hp_RBT 0.074 0.080 0.080 0.074 40 H_LR53 0.112 0.122 0.125 0.114 0.147 41 Hp_JP 0.112 0.122 0.125 0.114 0.145 0.002 42 Hp_LB 0.115 0.117 0.119 0.116 0.148 0.010 0.008 43 H_LR697 0.120 0.127 0.127 0.118 0.138 0.118 0.118 0.124 44 H_Sar 0.121 0.126 0.126 0.123 0.134 0.124 0.124 0.130 0.030 45 Hb_SC 0.122 0.125 0.125 0.124 0.134 0.124 0.124 0.130 0.028 0.003 46 Hb_SST 0.122 0.125 0.125 0.124 0.134 0.124 0.124 0.130 0.028 0.003 0.000 47 Hk_7136 0.132 0.142 0.142 0.137 0.150 0.123 0.123 0.129 0.058 0.045 0.046 0.046 48 Hb_SD 0.138 0.140 0.140 0.143 0.157 0.132 0.132 0.138 0.067 0.056 0.054 0.054 0.015 49 Hb_TR 0.116 0.137 0.137 0.122 0.138 0.106 0.106 0.112 0.058 0.044 0.042 0.042 0.040 0.046 50 Hb_SP 0.121 0.127 0.132 0.127 0.150 0.119 0.119 0.121 0.054 0.043 0.041 0.041 0.039 0.045 0.029 51 Hk_LS 0.106 0.106 0.106 0.110 0.126 0.122 0.122 0.124 0.078 0.085 0.084 0.084 0.099 0.101 0.090 0.090 52 Hk_LG 0.112 0.116 0.116 0.116 0.132 0.124 0.124 0.126 0.076 0.083 0.081 0.081 0.091 0.093 0.083 0.084 0.015 53 Hk_BK 0.120 0.120 0.120 0.124 0.136 0.118 0.118 0.120 0.083 0.087 0.085 0.085 0.096 0.098 0.087 0.088 0.020 0.025 54 H_LR699 0.124 0.126 0.130 0.132 0.135 0.126 0.126 0.128 0.099 0.099 0.097 0.097 0.111 0.107 0.101 0.096 0.036 0.038 0.032 55 Hk_MK 0.124 0.126 0.131 0.132 0.134 0.126 0.126 0.128 0.099 0.098 0.097 0.097 0.110 0.106 0.101 0.096 0.035 0.037 0.032 0.001 56 Hk_TT 0.112 0.109 0.109 0.112 0.132 0.120 0.120 0.122 0.078 0.085 0.083 0.083 0.095 0.095 0.084 0.086 0.029 0.030 0.025 0.044 0.044 57 Hk_KMD 0.136 0.139 0.139 0.136 0.153 0.124 0.124 0.126 0.093 0.084 0.085 0.085 0.093 0.100 0.083 0.092 0.071 0.073 0.074 0.084 0.084 0.080 58 Hk_LK 0.120 0.126 0.126 0.120 0.138 0.120 0.120 0.117 0.085 0.075 0.076 0.076 0.087 0.096 0.074 0.078 0.057 0.056 0.058 0.066 0.065 0.051 0.053 59 Hk_LL 0.114 0.120 0.120 0.114 0.134 0.118 0.118 0.120 0.083 0.073 0.074 0.074 0.089 0.096 0.072 0.080 0.058 0.058 0.062 0.071 0.071 0.060 0.055 0.018 60 Hk_KJI 0.114 0.120 0.120 0.114 0.134 0.118 0.118 0.120 0.083 0.073 0.074 0.074 0.089 0.096 0.072 0.080 0.058 0.058 0.062 0.071 0.071 0.060 0.055 0.018 0.000 61 Hk_7138 0.112 0.118 0.118 0.112 0.132 0.116 0.116 0.118 0.082 0.071 0.072 0.072 0.087 0.098 0.070 0.079 0.058 0.058 0.062 0.071 0.071 0.060 0.053 0.015 0.005 0.005 62 Hk_LBI 0.112 0.118 0.118 0.112 0.132 0.116 0.116 0.118 0.082 0.071 0.072 0.072 0.087 0.098 0.070 0.079 0.058 0.058 0.062 0.071 0.071 0.060 0.053 0.015 0.005 0.005 0.000 63 Hk_LMB 0.114 0.120 0.120 0.114 0.132 0.124 0.124 0.130 0.087 0.085 0.085 0.085 0.097 0.108 0.084 0.092 0.062 0.071 0.069 0.087 0.086 0.067 0.064 0.040 0.042 0.042 0.044 0.044 64 H_phaiosoma 0.149 0.149 0.151 0.146 0.151 0.162 0.162 0.166 0.135 0.137 0.141 0.141 0.147 0.153 0.156 0.152 0.136 0.143 0.147 0.150 0.149 0.151 0.144 0.150 0.151 0.151 0.149 0.149 0.147 65 H_chrysopunctatus0.175 0.177 0.177 0.180 0.184 0.177 0.177 0.182 0.193 0.193 0.193 0.193 0.170 0.175 0.173 0.174 0.166 0.170 0.172 0.175 0.175 0.181 0.188 0.186 0.178 0.178 0.175 0.175 0.191 0.201 66 H_tengah 0.143 0.155 0.155 0.146 0.149 0.159 0.159 0.161 0.144 0.147 0.148 0.148 0.150 0.156 0.144 0.160 0.144 0.147 0.157 0.147 0.147 0.143 0.164 0.136 0.137 0.137 0.131 0.131 0.143 0.177 0.161 67 Outgroup 0.214 0.192 0.192 0.209 0.183 0.201 0.201 0.199 0.199 0.217 0.218 0.218 0.215 0.210 0.215 0.221 0.202 0.207 0.214 0.211 0.210 0.203 0.204 0.208 0.203 0.203 0.201 0.201 0.201 0.201 0.208 0.150

313

Appendix F: Haplotype frequencies of cyt b across 25 H. pogonognathus populations in Sundaland.

Haplotype KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW n 15 6 15 15 14 13 15 15 7 15 13 15 15 14 15 13 15 9 7 13 15 15 4 15 15 Hpb01 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb02 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb03 0 0 0 0.933 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb04 0 0 0 0.067 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb05 0 0 0 0 0.929 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb06 0 0 0 0 0.071 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb07 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0.615 0 0 0 0 0 Hpb08 0 0 0 0 0 0 0.933 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb09 0 0 0 0 0 0 0.067 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb10 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb11 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb12 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb13 0 0 0 0 0 0 0 0 0 0 0.692 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb14 0 0 0 0 0 0 0 0 0 0 0.154 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb15 0 0 0 0 0 0 0 0 0 0 0.154 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb16 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 Hpb17 0 0 0 0 0 0 0 0 0 0 0 0 0.6 0 0 0 0 0 0 0 0 0 0 0 0 Hpb18 0 0 0 0 0 0 0 0 0 0 0 0 0.067 0 0 0 0 0 0 0 0 0 0 0 0 Hpb19 0 0 0 0 0 0 0 0 0 0 0 0 0.333 0 0 0 0 0 0 0 0 0 0 0 0 Hpb20 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Hpb21 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 Hpb22 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.846 0 0 0 0 0 0 0 0 0 Hpb23 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.154 0 0 0 0 0 0 0 0 0 Hpb24 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 Hpb25 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.8 0 0 0 0 0 0 0 0 Hpb26 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 Hpb27 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.714 0 0 0 0 0 0 Hpb28 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.286 0 0 0 0 0 0 Hpb29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.385 0.4 0 0 0 0 Hpb30 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.067 0 0 0 0 Hpb31 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.267 0 0 0 0 Hpb32 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.133 0 0 0 0 Hpb33 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.133 0 0 0 0 Hpb34 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 Hpb35 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.25 0 0 Hpb36 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.75 0 0 Hpb37 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 Hpb38 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.933 Hpb39 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.067

314

Appendix G: Haplotype frequencies of Hp5 across 25 H. pogonognathus populations in Sundaland.

Haplotype KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW n 26 6 10 20 10 10 22 20 18 10 24 26 6 7 9 20 10 18 22 32 20 22 36 24 30 Hp501 0.615 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp502 0.231 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp503 0.154 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp504 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp505 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.219 0 0 0 0 0 Hp506 0 0 0 0.95 1 1 0.136 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp507 0 0 0 0.05 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp508 0 0 0 0 0 0 0.091 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp509 0 0 0 0 0 0 0.364 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp510 0 0 0 0 0 0 0.091 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp511 0 0 0 0 0 0 0.227 0 0 0 0 0 0 0 0 0 0 0 0 0.156 0 0 0 0 0 Hp512 0 0 0 0 0 0 0.091 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp513 0 0 0 0 0 0 0 0.7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp514 0 0 0 0 0 0 0 0.3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp515 0 0 0 0 0 0 0 0 0.889 0 0.167 0.308 0 0 0 0 0 0 0 0.188 0 0.091 0 0 0 Hp516 0 0 0 0 0 0 0 0 0.111 0 0 0 0 0 0 0 0 0 0 0 0.1 0 0 0 0 Hp517 0 0 0 0 0 0 0 0 0 1 0.583 0.692 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp518 0 0 0 0 0 0 0 0 0 0 0.25 0 0 0 1 0.4 1 0 0.409 0 0 0 0 0 0.2 Hp519 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 Hp520 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.3 0 0 0 0 0 0 0 0 0 Hp521 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.25 0 0 0 0 0 0 0 0 0 Hp522 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.05 0 0 0 0 0 0 0 0 0 Hp523 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.278 0 0 0 0 0 0 0 Hp524 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 0 0 0 0 0 Hp525 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.222 0 0 0 0 0 0 0 Hp526 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.111 0 0 0 0 0 0 0 Hp527 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.111 0 0 0 0 0 0 0 Hp528 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.167 0 0 0 0 0 0 0 Hp529 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 0 0 0 0 0 Hp530 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.091 0 0 0 0 0 0 Hp531 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.091 0 0 0 0 0 0 Hp532 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.091 0 0 0.091 0 0 0.2 Hp533 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.182 0.188 0.3 0 0 0 0 Hp534 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.046 0 0 0 0 0 0 Hp535 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.091 0 0 0 0 0 0

315

Appendix G: continue…

Haplotype KWK TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW n 26 6 10 20 10 10 22 20 18 10 24 26 6 7 9 20 10 18 22 32 20 22 36 24 30 Hp536 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.063 0 0 0 0 0 Hp537 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.063 0 0 0 0 0 Hp538 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.063 0 0 0 0 0 Hp539 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.063 0 0 0 0 0 Hp540 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.3 0 0 0 0 Hp541 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0 0 Hp542 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.1 0 0 0 0 Hp543 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.182 0 0 0 Hp544 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.091 0 0 0 Hp545 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.091 0 0 0 Hp546 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.227 0 0 0 Hp547 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.091 0 0 0 Hp548 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.136 0 0 0 Hp549 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.222 0 0 Hp550 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 Hp551 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 Hp552 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.111 0 0 Hp553 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp554 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 Hp555 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.111 0 0 Hp556 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 Hp557 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 Hp558 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 Hp559 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 Hp560 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 Hp561 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 Hp562 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.056 0 0 Hp563 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.708 0 Hp564 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.25 0 Hp565 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.042 0 Hp566 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.133 Hp567 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.133 Hp568 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.067 Hp569 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.133 Hp570 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.067 Hp571 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.067 316

Appendix H: Haplotype frequencies of Hp54 across 25 H. pogonognathus populations in Sundaland.

Haplotype TE TB BP PTG SK PJG DM SU SOM KS PI JP LB S RA PA PT KJ ST SKR JBI JMP RBT MRW n 8 7 28 12 26 26 16 8 10 20 18 10 9 16 14 20 5 18 12 20 18 20 20 14 Hp5401 0.75 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5402 0.25 0 0 0.167 0 0 0.5 0 0 0 0 0 0 0.125 0.286 0.15 0 0.167 0 0 0 0 0 0 Hp5403 0 1 0.643 0 0 0.154 0 0 0 0 0 0 0 0 0 0.1 0 0 0 0 0 0 0 0 Hp5404 0 0 0.286 0 0 0 0 0 0 0 0.222 0 0 0.875 0 0.2 0 0 0 0.2 0 0 0 0 Hp5405 0 0 0.071 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5406 0 0 0 0.25 0 0 0 0 0 0 0 0 0 0 0 0.1 0 0 0 0 0 0 0 0 Hp5407 0 0 0 0.083 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5408 0 0 0 0.5 0 0 0.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5409 0 0 0 0 0.462 0.154 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5410 0 0 0 0 0.538 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5411 0 0 0 0 0 0.385 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5412 0 0 0 0 0 0.154 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5413 0 0 0 0 0 0.077 0 0 0 0 0 0 0 0 0.071 0 0 0 0 0 0 0 0 0 Hp5414 0 0 0 0 0 0.077 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5415 0 0 0 0 0 0 0 1 1 0.45 0.444 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5416 0 0 0 0 0 0 0 0 0 0.15 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5417 0 0 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5418 0 0 0 0 0 0 0 0 0 0.1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5419 0 0 0 0 0 0 0 0 0 0.1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5420 0 0 0 0 0 0 0 0 0 0 0.167 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5421 0 0 0 0 0 0 0 0 0 0 0.111 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5422 0 0 0 0 0 0 0 0 0 0 0.056 0 0 0 0 0 0 0 0 0 0 0 0 0 Hp5423 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 Hp5424 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.5 0 0 0 0 0 0 0 0 0 Hp5425 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.143 0 0 0 0 0 0 0 0 0 Hp5426 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.25 1 0.722 0 0 0 0 0 0 Hp5427 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.1 0 0 0 0 0 0 0 0 Hp5428 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.1 0 0.111 0 0 0 0 0 0 Hp5429 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.167 0.3 0 0 0 0 Hp5430 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.667 0 0 0 0 0 Hp5431 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.167 0 0 0 0 0 Hp5432 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0 0 Hp5433 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.1 0 0 0 0 Hp5434 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.1 0 0 0 0 Hp5435 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.1 0 0 0 0 Hp5436 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.222 0 0 0 Hp5437 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.444 0 0 0 Hp5438 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.111 0 0.2 0 Hp5439 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.111 0 0 0 Hp5440 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.111 0 0 0 Hp5441 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.45 0 0

317

Appendix I: Haplotype frequencies of cyt b, Hp5 and Hp54 across five H. byssus populations.

Haplotype SC SST SD TR SP cyt b n 15 15 15 14 15 Hbb01 1 Hbb02 1 Hbb03 0.867 Hbb04 0.067 Hbb05 0.067 Hbb06 0.214 Hbb07 0.714 Hbb08 0.071 Hbb09 1

Hp5 n 10 22 24 10 10 Hb501 1 0.909 Hb502 0.091 Hb503 0.5 Hb504 0.167 Hb505 0.333 Hb506 1 Hb507 1

Hp54 n 16 20 10 20 26 Hb5401 0.438 Hb5402 0.375 Hb5403 0.063 Hb5404 0.063 Hb5405 0.063 Hb5406 0.9 Hb5407 0.1 Hb5408 0.5 0.077 Hb5409 0.5 Hb5410 0.85 Hb5411 0.1 Hb5412 0.05 Hb5413 0.231 Hb5414 0.692

318

Appendix J: Haplotype frequencies of cyt b, Hp5 and Hp54 across eleven H. kuekenthali populations.

Haplotype LS LG BK MK TT KMD LK LL KJI LBI LMB cyt b n 15 6 15 15 15 15 15 15 15 15 14 Hkb01 0.2 Hkb02 0.8 Hkb03 1 Hkb04 1 Hkb05 1 Hkb06 0.2 Hkb07 0.333 Hkb08 0.133 Hkb09 0.333 Hkb10 1 Hkb11 0.733 Hkb12 0.2 Hkb13 0.067 Hkb14 0.867 Hkb15 0.133 Hkb16 0.067 Hkb17 0.8 Hkb18 0.133 Hkb19 0.133 Hkb20 0.733 Hkb21 0.133 Hkb22 0.286 Hkb23 0.714

Hp5 n 20 12 20 14 18 10 18 20 10 18 8 Hk501 0.3 0.917 0.45 Hk502 0.7 Hk503 0.083 0.55 Hk504 1 Hk505 0.944 Hk506 0.056 Hk507 1 Hk508 1 0.95 1 0.222 Hk509 0.05 Hk510 0.056 Hk511 0.056 Hk512 0.667 Hk513 1

Hp54 n 16 2 18 28 6 20 18 10 10 18 18 Hk5401 0.75 1 0.944 Hk5402 0.25 Hk5403 0.056 Hk5404 0.286 1 Hk5405 0.25 Hk5406 0.071 Hk5407 0.25 Hk5408 0.071 Hk5409 0.071 Hk5410 0.85 0.111 0.778 Hk5411 0.15 0.611 1 1 0.889 Hk5412 0.278 Hk5413 0.111 Hk5414 0.111 Hk5415 0.111

319

LIST OF PUBLICATIONS / CONFERENCE PRESENTATIONS

1. Lim, H.C., Zainal Abidin M, Pulungan CP, de Bruyn M, Mohd Nor SA (2016). DNA Barcoding Reveals High Cryptic Diversity of the Freshwater Halfbeak Genus Hemirhamphodon from Sundaland. PLoS ONE 11(9): e0163596. doi:10.1371/journal.pone.0163596

2. Lim, H.C., Siti Azizah, M.N. (2016). Utilization of DNA Barcode Analysis in Species Discovery of the Genus Hemirhamphodon. 2016 ASI Annual Meeting, 18-21 May 2016, Taipei, Taiwan.

3. Lim, H.C., Siti Azizah, M.N. (2013). Population Study of the Freshwater Halfbeak Hemirhampodon pogonognathus (Bleeker, 1866) in Peninsular Malaysia Using Cytochrome b Gene and the Preliminary Evidence of New Taxon. SAGE 2013 2nd Southeast Asian Gateway Evolution Meeting. 11-15th March 2013, Berlin, Germany.

4. Lim Hong Chiun. (2013). Utilization of Mitochondrial DNA in the Population Study of the Freshwater Halfbeak Hemirhmphodon pogonognathus (Bleeker, 1866) in Peninsular Malaysia and the Preliminary Evidence of New Taxon. Biodiversity Asian Freshwater Fishes Conference, 10-13rd Jan 2013, Brunei.

5. Lim, H. C., Siti Azizah, M.N. (2012). Population Study of the Freshwater Halfbeak Hemirhampodon pogonognathus (Bleeker, 1866) in Peninsular Malaysia: Preliminary Evidence of New Taxon. The 2nd Annual International Conference In-conjunction with The 5th IMT-GT UNINET Bioscience Conference, 22-24th Nov 2012, Banda Ache, Indonesia.

320