“Phylogeography and Evolution of Butterflyfish

“PHYLOGEOGRAPHY AND EVOLUTION OF BUTTERFLYFISH IN THE SUBGENUS CORALLOCHAETODON: CHAETODON LUNULATUS, CHAETODON TRIFASCIATUS, CHAETODON AUSTRIACUS, CHAETODON MELAPTERUS”

A THESIS SUBMITTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF HAWAI`I IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

MASTER OF SCIENCE

ZOOLOGY (MARINE BIOLOGY)

MAY 2014

By Ellen Waldrop

Thesis Committee:

Brian Bowen, Chairperson

Robert Toonen

Robert Thomson

ACKNOWLEDGMENTS

Funding Sources National Science Foundation Grants OCE-0453167 and OCE-0929031 to B.W. Bowen Elizabeth Alison Kay Endowed Award Graduate Student Organization Travel Grant Papahanaumokuakea Marine National Monument NOAA National Marine Sanctuaries Program MOA No. 2005-008/66882 to R.J. Toonen National Geographic Society Grant 9024-11 to J.D. DiBattista KAUST Red Sea Research Center funding to M.L. Berumen University of Hawaii at Manoa Institutional Logistic Support Toonen-Bowen Laboratory University of Hawaii Diving Safety Program University of Hawaii sequencing lab Hawaii Institute of Marine Biology Hawaii Department of Land and Natural Resources King Abdullah University of Science and Technology University of Hawaii Annual Testers Symposium 9TH Indo-Pacific Fish Conference Conservation International Coral Reef Research Foundation Phoenix Island Protected Area Dept. of the Environment, Australian Government Government of Kiribati Administration of the British Indian Ocean Territories Fagatele Bay National Marine Sanctuary (NOAA) Dept. of Marine and Wildlife Resources, American Samoa Government of Fiji and the Chiefs and people of Wagamimi, Tavewa, and

Yasawas villages Government of French Polynesia U.S. Fish and Wildlife Service (Johnston Atoll)

For assistance with field work and collections, I thank Alexander Alfonso, Senifa Annadale, Kim Anderson, Paul H. Barber, Brian Bowen, W.K. Chan, J. Howard Choat, Richard Coleman, Pat and Laura Colin, Greg Concepcion, Joshua Copus, Matthew Craig, Toby S. Daly-Engel, Nancy Daschbach, Joseph DiBattista, Joshua A. Drew, John L. Earle, Jeff Eble, Kevin Flanagan, Michelle R. Gaither, Brian D. Greene, J.P Hobbs, Matthew Iacchei, Stephen A. Karl, Randall K. Kosaki, Carl G. Meyer, Darren Okimoto, Yannis P. Papastamatiou, David Pence, Mark Priest, Jon Puritz, Richard Pyle, John Randall, Joshua Reece, D. Ross Robertson, Luiz Rocha, Nick Russo, Jennifer Schultz, Charles Sheppard, Derek Skillings, Derek Smith, Zoltan Szabo, Sue Taei, Kim Tenggardjaja, Tukabu Teroroko, Robert Thorn, Allen Tom, Bill Walsh, Christie Wilcox, Ivor Williams, Jill Zamzow, and the crew of the R.V. Hi’ialakai.

For logistic support, advise, and valuable consultation, I thank Robert Toonen, Robert Thomson, John Randall, Stephen Karl, Jo-Ann Leong, David Pence, Luiz Rocha, Joseph DiBattista, Jason Jones, Narineh Nazarian, Les Watling, Michelle Gaither, Iria Fernandez-Silva, Zac Forsmen.

Finally, I would like to thank my husband Anthony and my family for all of their love and continued support.

iii

ABSTRACT

The subgenus Corallochaetodon consists of four partially overlapping specialized corallivores: Chaetodon lunulatus (Pacific Ocean), Chaetodon trifasciatus (Indian Ocean), Chaetodon melapterus (northern Indian Ocean), Chaetodon austriacus (Red Sea). Phylogenetic and population genetic analysis of mtDNA cytochrome b sequences reveal divergent patterns of genetic diversity and population structure. Phylogenetic reconstruction shows that C. lunulatus diverged from sister species C. trifasciatus approximately 3 million years ago, while C. melapterus and C. austriacus comprise a cluster of closely related haplotypes derived from C. trifasciatus over the last half million years. Microsatellite loci analysis revealed varying levels of population structure in C. lunulatus and C. trifasciatus. This evolutionary divergence may have been initiated along biogeographic boundaries such as the barrier between Indian and Pacific Oceans, and between the Indian Ocean and Red Sea. Species integrity in regions of overlap may be reinforced by ecological factors such as mate choice, habitat preferences, and the documented differences in coral prey between species.

TABLE OF CONTENTS

ACKNOWLEDGMENTS...... ii ABSTRACT...... iv LIST OF TABLES...... vii LIST OF FIGURES...... viii CHAPTER 1 GENERAL INTRODUCTION...... 9 CHAPTER 2 PHYLOGEOGRAPHIC PATTERNS IN THE BUTTERFLYFISH SUBGENUS CORALLOCHAETODON...... 15 INTRODUCTION...... 15 MATERIALS AND METHODS...... 17 Sample Collection...... 18 Mitochondrial DNA Sequencing...... 18 RESULTS...... 21 Genetic diversity within species...... 21 Genetic divergence between species...... 23 DISCUSSION...... 25 Population Structure...... 25 Range Size and Marine Biogeography...... 28 Implications of Specialization...... 32 CONCLUSION...... 34 CHAPTER 3 INTRA-SPECIES DIFFERENTIATION IN THE BUTTERFLYFISH SUBGENUS CORALLOCHAETODON (Chaetodon lunulatus and C. trifasciatus)...... 35 INTRODUCTION...... 35 MATERIALS AND METHODS...... 36 RESULTS...... 39 Chaetodon lunulatus...... 39 Chaetodon trifasciatus...... 42 DISCUSSION...... 43 Chaetodon lunulatus...... 43

Chaetodon trifasciatus...... 45 CONCLUSION...... 46 CHAPTER 4 FINAL SUMMARY...... 48 APPENDIX...... 49 LITERATURE CITED...... 50

LIST OF TABLES

Table 2.1. Sample size and molecular diversity indices for C. lunulatus, C. trifasciatus, C. melapterus, and C. austriacus based on mtDNA cytochrome b (cyt b) data...... 21

Table 2.2. Matrix of population pairwise ΦST values and associated P values based on mtDNA cyt b sequence data from C. lunulatus...... 22

Table 2.3. Matrix of population pairwise ΦST values and associated P values based on mtDNA cyt b sequence data from C. lunulatus Hawaiian sampling locations...... 22

Table 2.4. Matrix of population pairwise ΦST values and associated P values based on mtDNA cyt b sequence data: C. trifasciatus, C. melapterus, C. austriacus...... 23

Table 2.5. Estimation of divergence time between species...... 32

Table 3.1. Characteristics of microsatellites used in this study, developed for C. lunulatus by Lawton et al. 2010...... 37

Table 3.2. Matrix of population pairwise FST values and associated P values based on microsatellite genotypes for C. lunulatus...... 40

Table 3.3. Microsatellite statistics averaged over all loci for each location and species...41

Table 3.4. Matrix of population pairwise FST values and associated P values based on microsatellite genotypes for C. trifasciatus...... 44

Table A.1. Matrix of population pairwise ΦST values, FST values, and associated P values for C. lunulatus at Hawaiian sampling locations...... 49

vii

LIST OF FIGURES

Figure 1.1. The subgenus Corallochaetodon, including four sister species...... 10

Figure 1.2 Distribution map of the Corallochaetodon subgenus...... 13

Figure 2.1. Neighbor joining tree based on mtDNA cyt b sequences, highlighting the relationship between sister species in the subgenus Corallochaetodon...... 24

Figure 2.2. Statistical parsimony network for C. lunulatus, C. trifasciatus, C. melapterus, and C. austriacus based on mtDNA cyt b sequences...... 30

Figure 2.3. Known marine biogeographic barriers and their influence on the distribution range of Corallochaetodon...... 31

Figure 3.1. STRUCTURE bar plot C. lunulatus...... 41

Figure 3.2. Mean Ln probability of data (Ln P[D]) and Delta K for C. lunulatus as generated by the program STRUCTURE HARVESTER...... 42

Figure 3.3. Mean Ln probability of data (Ln P[D]) and Delta K for C. trifasciatus as generated by the program STRUCTURE HARVESTER...... 43

Figure 3.4. Prevailing currents in the north-western Indian Ocean...... 46

viii

CHAPTER 1

GENERAL INTRODUCTION

This research is part of an ongoing phylogeographic survey of coral reef fishes both at

Indo-Pacific and Hawaiian Archipelago scales. The broader scope is to resolve patterns of evolution, colonization, and diversification in the oceans and to support the conservation of coral reef fishes. Many marine species are declining due to habitat loss, overfishing, pollution, and other human-mediated challenges. Understanding how marine populations are connected and how different species evolve is crucial to understanding global patterns of biodiversity, the foundations of future biodiversity, and effective conservation strategies. Recent advancements in genetic techniques and analyses are leading to a better understanding of these processes.

Characterizing genetic connectivity in the marine environment has been an ongoing challenge to modern marine biology. In marine organisms, high dispersal potential may retard the accumulation of reproductive barriers and promote evolutionary stasis by connecting widely separated populations (Palumbi 1997; Bowen et al. 2013). The previously accepted view of largely open populations of coral reef organisms is shifting towards a realization that many species have limited dispersal and demographically discrete populations (Leis 2002; Swearer et al. 2002; Almany et al. 2007), highlighting the need for a better understanding of dispersal and population connectivity in marine environments (Rocha et al. 2007; Toonen et al. 2013).

Butterflyfishes (family Chaetodontidae) are diverse and abundant pan-tropical coastal fish that have stunning coloration, high levels of ecological and morphological

9 diversity, and are important and distinctive members of coral reef communities. Many are considered specialists in terms of diet, feeding exclusively on live coral and/or coral polyps. This dietary specialization makes them vulnerable to habitat shifts that alter resource availability (i.e. the declining state of coral reefs worldwide). Recent research has shown that butterflyfishes, because of their distinct diets and geographical range, are a useful model group for studying general questions about ecological specialization and patterns of gene flow (Craig et al. 2010; Lawton et al. 2011a). As noted by Fessler and

Westneat (2007), “The exploration of evolutionary, ecological, and biogeographic patterns remains limited by lack of a well-resolved molecular phylogeny involving species of chaetodontid.”

Figure 1.1. The subgenus Corallochaetodon, including four sister species. Clockwise from top left Chaetodon lunulatus, Chaetodon trifasciatus, Chaetodon melapterus, Chaetodon austriacus.

The subgenus Corallochaetodon includes four sister species: Chaetodon lunulatus

(Quoy and Gaimard 1824) in the Pacific and eastern Indian Ocean, Chaetodon trifasciatus (Park 1797) in the Indian Ocean, Chaetodon melapterus (Guichenot 1863) in the northern Indian Ocean, and Chaetodon austriacus (Rüppell 1836) in the Red Sea

(Figure 1.1). This subgenus is also a monophyletic group (Smith et al. 2003; Fessler and

Westneat 2007) and several studies have recently shed light on the evolutionary history of this group (Littlewood et al. 2004; Hsu et al. 2007; Bellwood et al. 2009). However, no study has used comprehensive geographic coverage of all four species in a hierarchical analysis of genetic partitions. Therefore, the history of the group remains unresolved.

All members of Corallochaetodon are obligate corallivores (Pratchett et al.

2013a), but each species has distinct dietary characteristics. The Pacific C. lunulatus is considered a generalist and feeds on a broad range of hard corals from up to 52 different genera (Pratchett 2005). The Indian Ocean C. trifasciatus is more of a specialist preferring to feed on polyps, with some evidence of geographic variation in preferred genera (Reese 1989; Samways 2005; Graham 2007). C. melapterus (northern Indian

Ocean) feeds exclusively on coral polyps (Lieske and Myers 1996; Shokri et al. 2005), and has coral species-specific feeding selectivity (Pratchett et al. 2013a). C. austriacus

(Red Sea) is considered a generalist that feeds primarily on three genera of abundant hard corals (Acropora, Porites, and Pocillopora) (Alwany et al. 2003).

The Corallochaetodon species are morphologically similar and distinguished primarily by color pattern. All four species share a yellow to orange background overlaid by a series of 14-16 dark (blue-black) stripes that are nearly horizontal. C. lunulatus has narrow, posteriorly expanding, black markings along the base of the dorsal and anal fins,

11 a reddish anal fin, and a grey caudal peduncle with a vertical black bar at the end of the caudal fin. C. trifasciatus also has narrow, posteriorly expanding, black markings along the base of the dorsal and anal fins, but has a yellow anal fin and a yellow to orange caudal peduncle with a vertical black bar at the end of the caudal fin. Both C. melapterus and C. austriacus have anal and caudal fins that are completely black except for fringing white highlights (C. melapterus also has a black dorsal fin while C. austriacus has a white dorsal fin). This single derived feature suggest that these two species are more closely related to each other than either is to C. lunulatus or C. trifasciatus (species descriptions from Blum 1989).

The subgenus is found throughout the Indo-Pacific Ocean (except the eastern

Pacific) and the size of the geographical range of its members varies greatly (Figure 1.2).

C. lunulatus has by far the largest range and is distributed throughout the central and western Pacific Ocean, from western Australia north to Japan and east to Hawaii and the

Tuamotu Islands (Allen et al. 1998). It is replaced by C. trifasciatus in the Indian Ocean, with a range from East Africa to Bali, Indonesia, Cocos Keeling, and Christmas Islands

(Allen et al. 1998). These two sister species overlap at Christmas Island and Cocos

Keeling in the eastern Indian Ocean (Australia), where they form heterospecific pairs and hybridize (Hobbs et al. 2009; Montenari et al. 2011). C. melapterus occurs in the Persian

Gulf, Gulf of Aden, Gulf of Oman, and off the Yemen coast at the extreme southern tip of the Red Sea (Zekeria et al. 2005). C. austriacus has the smallest range and is endemic to the Red Sea, particularly the central and northern areas (Zekeria et al. 2005). However it has recently invaded the Mediterranean via the Suez Canal, raising interesting possibilities for future range expansion (Goren et al. 2011). While the distributions of C.

12 melapterus and C. austriacus are often listed as “northern Indian Ocean and Red Sea,” in fact the former barely gets inside the mouth of the Red Sea, while the latter is rarely seen south of the Farasan Islands in the lower Red Sea. These distributional limits have been recently been confirmed by personal observations (J.D. DiBattista, L.A. Rocha, and B.W.

Bowen). Hence the distribution of these two species is best characterized as parapatric, with effectively no overlap under contemporary conditions.

Figure 1.2 Distribution map of the Corallochaetodon subgenus (redrawn from Blum 1989). Chaetodon lunulatus is distributed throughout the Central and Western Pacific Ocean, from Eastern Australia north to Japan and east to Hawaiian and the Tuamotu Islands. C. trifasciatus is distributed throughout the Indian Ocean, from East Africa to Bali, Indonesia, Cocos Keeling, and Christmas Islands. C. melapterus occurs in the Persian Gulf, Gulf of Aden, and Gulf of Oman. C. austriacus occurs in the Northern and central Red Sea, and is considered endemic.

With this thesis I hope to understand the evolutionary history of this interesting group of fishes, specifically in terms of geography, oceanography, and ecology. The first part of the study analyzes mitochondrial DNA (mtDNA) cytochrome b (cyt b) sequences

13 to determine the relationship among sister species. Identifying population structure within species has important management implications, especially for the intensively-sampled

Hawaiian Archipelago and Papahanaumokuaka Marine National Monument. Therefore the second part of this study uses microsatellite analysis to investigate genetic partitions within the two widespread species (C. lunulatus and C. trifasciatus).

CHAPTER 2

PHYLOGEOGRAPHIC PATTERNS IN THE BUTTERFLYFISH SUBGENUS

CORALLOCHAETODON

INTRODUCTION

Even closely related species can have different levels of dispersal and population structure (e.g. Atlantic surgeonfishes and wrasses, Rocha et al. 2007). Within the subgenus Corallochaetodon, the varying diets, colorations, and distributions indicate the likelihood of distinct population structures and unique genetic features among these four species. Consequently I investigated how habitat requirements and geographic isolation both influence patterns of mtDNA diversity within and among these species.

This study is motivated by four primary questions. First, what is the phylogenetic history of this group? The widespread C. lunulatus (Quoy and Gaimard 1825) (Pacific

Ocean) and C. trifasciatus (Park 1797) (Indian Ocean) have been recognized alternately as a single species, subspecies, or sister species (Burgess 1978; Myers 1999; Randall

2007). The relationships of the two restricted range species, C. austriacus (Ruppell 1836)

(Red Sea), and C. melapterus (Guichenot 1863) (adjacent northern Indian Ocean), are unknown. Did they arise from the widespread C. trifasciatus, as geography would indicate? Did the restricted range species arise independently of one another?

Phylogenetic analyses should illuminate the history of speciation in this complex.

Second, what is the geographic distribution of genetic diversity between species?

Marine biogeographic barriers have been recognized for decades (Ekman 1953; Briggs

1974; Briggs and Bowen 2012), although in tropical marine habitats, only a few barriers

15 are apparent and well studied (Rocha et al. 2007). The widespread C. lunulatus and C. trifasciatus are divided by a well-known barrier between the Indian and Pacific Oceans

(Indo-Pacific Barrier [IPB] or Sunda Shelf Barrier). On the other hand, C. austriacus and

C. melapterus are both limited range species. The former may have been historically isolated in the Red Sea (DiBattista et al. 2013) but the latter has no clear barriers with C. trifasciatus to the south, and C. austriacus to the north. Here I am resolving the role of biogeographic provinces in shaping evolutionary history.

Third, does ecological specialization drive the patterns of diversity of butterflyfishes? Chaetodontids are among the most specialized of coral reef fishes, with many species having an obligate dependence on scleractinian corals, however it is currently unknown why some butterflyfishes are highly specialized and others are generalists (Pratchett et al. 2013b). Generalist species consume a wide range of coral species and have the ability to switch prey to cope with alterations in their environment.

On the other hand, more specialized species sacrifice the ability to withstand fluctuations in resource ability, presumably because dietary specialization will yield greater fitness.

Based on these observations I predict species with similar dietary specializations to have similar patterns of genetic structure. Identifying a link between diet preferences, range size, and dispersal ability in these species can aid in their successful management.

Fourth, what is the distribution of genetic diversity within species? Since the degree of genetic structure increases as range size decreases species with smaller distribution range should show a higher degree of genetic structuring than widespread species. Based on previous research (Eble et al. 2009), I hypothesize that the vast geographic range of C. lunulatus (Pacific Ocean) and C. trifasciatus (Indian Ocean)

16 should indicate strong population connectivity, and low genetic structure. Whereas the range-restricted C. melapterus should exhibit moderate population structure and range- restricted endemic C. austriacus should exhibit the most pronounced population structure. I hope to resolve population separations that are relevant to management. In particular, are there isolated management units within the Hawaiian Archipelago?

This study analyzed DNA sequence data from all members of the subgenus

Corallochaetodon to assess evolutionary history in the Indo-Pacific. The primary goals were to (1) resolve relationships among all four species, (2) establish how geographic barriers have influenced speciation in this subgenus, (3) determine how specialization might drive speciation in butterflyfishes, and (4) determine the degree of any population connectivity within and between species. My approach here in chapter 2 is to document levels of mtDNA diversity among individuals, populations, regions, and species. In chapter 3, some of the same issues are evaluated with microsatellite loci.

MATERIALS AND METHODS

This study uses mitochondrial DNA (mtDNA) cytochrome b (cyt b) sequences to resolve relationships from contemporary to evolutionary timescales. The mtDNA cyt b data comprises a single locus but offers the advantages of haploid inheritance, lack of recombination, potential for comparison to existing and future studies, and availability of universal primers for efficient production of sequence data.

Sample Collection

To characterize the range-wide phylogeographic structure of Corallochaetodon butterflyfishes, tissue samples (fin clips or gill filament) were collected with pole spears while scuba diving or snorkeling between 2005 and 2013. Specimens were obtained from

24 locations across the Indo-Pacific spanning the entire range of the subgenus (C. lunulatus N = 719, C. trifasciatus N = 160, C. melapterus N = 62, C. austriacus N = 30).

Within the Pacific range of C. lunulatus, I used a multi-scale approach consisting of range-wide sampling plus an intensive regional evaluation across the 2600 km length of the Hawaiian Islands, one of the largest and most isolated archipelagos in the world

(Randall 2007). Collected tissues were preserved in a saturated salt DMSO solution

(Seutin et al. 1991).

Mitochondrial DNA Sequencing

Genomic DNA was extracted using a “HotSHOT” protocol (Meeker et al. 2007), and aliquots were stored at -20 ˚C until sequenced. A 605 basepair (bp) segment of mtDNA cytochrome b (cyt b) gene was resolved using heavy-strand (Cyb 9) (FOR: 5’-

GTGACTTGAAAAACCACCGTTG-3’ Song et al. 1998) and light-strand primers (Cyb

7) (REV: 5’-AATAGGAAGTATCATTGCGGTTTGATG-3’ Taberlet et al. 1992).

Polymerase chain reaction (PCR) mixes contained 7.5 ul of 2x BioMix Red solution (BioMix Red; Bioline Ltd., London, UK), 0.3 ul (10 uM) of each primer, and 5–

50 ng template DNA in 15 ul total volume. PCRs had an initial denaturing step at 95°C for 3 min, then 35 cycles of amplification (30 s of denaturing at 94°C, 45 s of annealing at 52°C, and 45 s of extension at 72°C), followed by a final extension at 72°C for 10 min.

PCR products were visualized on a 1.5% agarose gel electrophoresis and purified by incubating with 1.1 units of ExoFAP (Fermentas) per 10 ul PCR product at 37°C for

60 min, followed by deactivation at 85°C for 15 min. DNA sequences were resolved with an ABI 3130XL Genetic Analyzer (Applied Biosystems, Foster City, CA). All specimens were sequenced in the forward direction and questionable haplotypes were resequenced in the reverse direction. The sequences were aligned, edited, and trimmed to a common length using Geneious Pro version 5.6.7 (Drummond et al. 2010); unique mtDNA cyt b haplotypes will be deposited in GenBank.

jModelTest version 2.1.3 (Posada 2008) was used with an Akaike information criterion (AIC) test to determine the best nucleotide substitution model in each species; for the four species, the HKY (Hasegawa, Kishino, and Yano 1985), HKY, HKY, and

TrN+G (Tamura and Nei 1993) models were selected for C. lunulatus, C. trifasciatus, C. austriacus, and C. melapterus, respectively. The Tamura/Nei is the only one of the models available in analytical software (Arlequin version 3.5, see below) and was therefore selected here for phylogeographic inferences.

Arlequin version 3.5.1.3 (Excoffier et al. 2005) was used to calculate haplotype

(h) and nucleotide diversity (π), Fu’s Fs test of neutrality (Fu 1997), as well as an analysis of molecular variance (AMOVA; Excoffier et al. 1992) to test for range-wide patterns of population structure. Tests were run within and between localities for each species separately.

Demes with fewer than five individuals were excluded from all population-level analyses and pooled into their respective larger sampling locations to provide adequate

19 statistical power (Crandall et al. 2008). In addition to population pairwise comparisons for C. lunulatus, Hawaiian specimens were subdivided into the Main Hawaiian Islands

(high islands) and North Western Hawaiian Islands (low islands and atolls) to test for subdivision within the archipelago. For C. trifasciatus specimens from western Australia

(Cocos Keeling and adjacent Christmas Island), were pooled to increase statistical power since analysis using separate populations resulted in the same conclusions.

In addition to the population genetic analyses, neighbor-joining (NJ), maximum- likelihood (ML), and maximum-parsimony (MP) trees were constructed to examine phylogenetic relationships among cyt b haplotypes for all individuals (PAUP* implemented in Geneious Pro, Swofford 2003, and MEGA version 5.2.2, Tamura et al.

2011). Bootstrap support values for NJ, ML, and MP trees were calculated using default settings with 10,000 replicates. For simplicity, a subset of haplotypes was used to create the final tree. Two Chaetodon vagabundus samples (Genbank accession numbers:

JF458006, JF458008) were used to root the tree. To further explore evolutionary relationships NETWORK version 4.5.1.0 (Fluxus) was used to construct an unrooted network among all haplotypes using a median-joining algorithm and default settings

(Bandelt et al. 1999; DiBattista et al. 2012a).

Estimation of divergence between species was performed using a well- documented evolutionary rate, estimated to be 2% per million years (between lineages) for the mtDNA cyt b gene (Johns and Avise 1998; Bowen et al. 2001; Reece et al. 2011).

MEGA (version 5.2.2, Tamura et al. 2011) was used to calculate evolutionary distances based on the average number of nucleotide substitutions among lineages using the

Tamura-Nei model and 1,000 bootstrap replicates. Using this evolutionary rate, the approximate divergence times of Chaetodon species were estimated.

RESULTS

Genetic diversity within species

Haplotype diversity within each species (a measure of the frequency and number of haplotypes) was moderate to high (C. lunulatus h = 0.45 - 0.80; C. trifasciatus h = 0.67 -

0.80; C. melapterus h = 0.36 - 0.77; C. austriacus h = 0.20 - 0.87). Nucleotide diversity

(a measure of the mean divergence between sequences) was low (C. lunulatus π = 0.002 -

0.005; C. trifasciatus π = 0.001 - 0.09; C. melapterus π = 0.001 - 0.002; C. austriacus π =

0.000 - 0.005), indicating a cluster of closely-related haplotypes in each species.

Table 2.1. Sample size and molecular diversity indices for Chaetodon lunulatus, C. trifasciatus, C. melapterus, and C. austriacus based on mtDNA cytochrome b sequence data (significant Fu’s Fs values are in bold, P < 0.05).

Tests for Fu’s Fs were significant in five locations across three species: C. trifasciatus at Diego Garcia; C. melapterus at Maskali and Obock; C. austriacus at Jazirat

Baraqan and Yanbu (Table 2.1).

Table 2.2. Matrix of population pairwise ΦST values (above diagonal) and associated P values (below diagonal) based on 605 bp of mtDNA cyt b sequence data from Chaetodon lunulatus. Significant ΦST P values are indicated in bold (P < 0.05).

Significant population structure was observed in C. lunulatus (overall ΦST = 0.23;

P < 0.001). In comparisons among sample locations, 31 out of 78 pairwise comparisons were statistically significant (P < 0.05). Five locations accounted for all the significant comparisons (possibilities out of 12 comparisons for each location): Fiji with seven out of

12 significant comparisons; Johnston Atoll with three out of 12 significant comparisons;

Moorea (French Polynesia) with 12 out of 12 significant comparisons; Main Hawaiian

Islands, with five out of 12 significant comparisons; and the North West Hawaiian

Islands with 12 out of 12 significant comparisons (Table 2.2).

Table 2.3. Pairwise comparisons within Hawaiian sampling locations: mtDNA ΦST values are above the diagonal, and microsatellite FST values are below the diagonal. Significant P values values are highlighted in bold (P < 0.05).

Within the Hawaiian Archipelago, there were 13 out of 28 significant comparisons among sample locations (Table 2.3).

No significant structure overall or significant pairwise comparisons were detected among four locations in C. trifasciatus (ΦST < 0.01; P = 0.50), three locations in C. melapterus (ΦST = 0.03; P = 0.12), or three locations in C. austriacus (ΦST = 0.04; P =

0.21) (Table 2.4).

Table 2.4. Matrix of population pairwise ΦST values (above diagonal) and associated P values (below diagonal) based on 605 bp of mtDNA cyt b sequence data from Chaetodon trifasciatus, C. melapterus, C. austriacus.

Genetic divergence between species

Neighbor joining, maximum likelihood, and maximum parsimony reconstruction all produced nearly identical tree topologies with bootstrap support values for species level relationships between 80 and 100% (Figure 2.1). The primary feature of this phylogeny is a bifurcation at approximately d = 0.06 sequence divergence between Pacific C. lunulatus compared to the Indian Ocean C. trifasciatus, while C. melapterus and C. austriacus are

23 closely related to the Indian Ocean species (d = 0.015), and shared the most common haplotype in these two species. This relationship is apparent in the parsimony network, in which Pacific C. lunulatus and Indian Ocean C. trifasciatus are separated by 28 nucleotide substitutions, and the C. melapterus/austriacus cluster is separated from C. trifasciatus by three nucleotide substitutions (Figure 2.2).

Figure 2.1. Neighbor joining tree based on mtDNA cyt b sequences, highlighting the relationship between sister species in the subgenus Corallochaetodon (bootstrap values shown, 1000 replicates) (MEGA, Tamura et al. 2011).

DISCUSSION

Population Structure

The survey of Corallochaetodon mtDNA sequence data revealed contrasting levels of genetic structure across the subgenus. No structure was detected for C. trifasciatus, C. melapterus, and C. austriacus, while C. lunulatus showed significant population structure, even within the Hawaiian Archipelago. Pairwise population comparisons were significant between several locations (Table 2.2) the most pronounced in areas of moderate to extreme isolation (e.g. Fiji, Moorea, Johnston Atoll, and Hawaii).

Populations that are genetically distinct arise from several features including limited dispersal, geographic barriers, or ecological factors (e.g. Palumbi 1994; Rocha et al.

2007). In this study the genetically distinct and geographically distant populations are clearly consistent with known distribution barriers (Blum 1989; Hsu et al. 2007).

Additionally, while the mtDNA support an “open” system of dispersal for the majority of the C. lunulatus distribution range, which is connected by archipelagos that are separated by no more than 800 km (Schultz et al. 2008), this unexpected genetic subdivision suggests a degree of self-recruitment in some populations (e.g. Hourigan and Reese 1987;

Planes 1993; Jones et al. 1999; Swearer et al. 2002).

Pairwise comparisons with Fiji were statistically significant in seven out of 12 comparisons: American Samoa, Johnston Atoll, and Christmas Island (I) (P < 0.05), and the Marshall Islands, Moorea, MHI, and NWHI (P < 0.02) (Table 2.2). A study of cosmopolitan reef fishes showed pronounced genetic partitioning between populations in

Fiji and those from Papua New Guinea, Indonesia, and the Solomon Islands (Drew et al.

2008), indicating barriers to genetic exchange in this location. Strong patterns of regional isolation may indicate limited larval dispersal capabilities that contribute to the higher levels of genetic structure on the scale of the Fijian Archipelago (Drew and Barber 2012).

Pairwise comparisons with Moorea were all highly statistically significant (P <

0.001 in all cases) (Table 2.2). This result in consistent with a previous study of population structure in C. lunulatus in which the authors also found Moorea samples were significantly different from all other locations (five locations were used in this study: Lizard Island, North Great Barrier Reef; Heron Island, South Great Barrier Reef;

Kimbe Bay, Papua New Guinea; Noumea, New Caledonia; and Moorea, French

Polynesia) (Lawton et al. 2011b). Previous studies of reef fishes in Moorea also showed significant levels of population genetic structure in blacktail snapper (Lutjanus fulvus) and brown surgeonfish (Acanthurus nigrofuscus) (Gaither et al. 2010; Eble et al. 2011b).

Planes (1993) showed significant genetic structure in convict surgeonfish (Acanthurus triostegus) within French Polynesia alone with five out of 11 sampling locations genetically distinct, including Moorea. Based on the conclusions of these authors it is possible that oceanographic isolation, (specifically in relation to known current patterns in the region), may also explain the concordant break between Moorea and other central

Pacific sampling sites in C. lunulatus.

Pairwise comparisons with Johnston Atoll were statistically significant between three populations: Moorea and Hawaii (P < 0.001 in both cases), and Fiji (P < 0.05)

(Table 2.2). Johnston is a small open atoll extremely isolated from other shallow coral reefs. It lies 800km southwest of the nearest reefs of Hawaii and over 1,500km from other shallow reefs (Line Island and Marshall Islands). It exhibits low species diversity

26 and is primarily inhabited by species with broad geographic distributions. While it is in a position to serve as a stepping-stone in the Pacific (Kobayashi 2006), its remoteness, small size, lack of habitat diversity (Kosaki et al. 1991), and lack of favorable currents to transport larvae likely inhibit gene flow of C. lunulatus between Johnston Atoll and the

Hawaiian reefs (Maragos and Jokiel 1986; Skillings et al. 2011; Timmers et al. 2011).

The PLD for C. lunulatus is estimated at ~35 days (Nakamura et al. 2012), and likely falls short of the estimated “40–50-day PLD required to transit from Johnston Atoll to the

Hawaiian Archipelago” (Kobayashi 2006). Additionally, geological history may contribute to isolation in that habitat area and diversity could have been drastically reduced by temporarily lowered sea levels during Pleistocene glaciations (Kosaki et al.

1991). This effect has been observed in the hermit crab, Calcinus hazletti, where habitat availability during sea-level fluctuations was a significant force in driving the high level of differentiation across the Hawaiian Archipelago (Baums et al. 2014).

The Hawaiian Islands pairwise comparisons were statistically significant in 13 out of 28 population comparisons (Table 2.3). Population differentiation between the

Hawaiian Islands and other locations in the Pacific Ocean is not surprising since they are also one of the most isolated islands groups in the world with the highest level of endemism in the Pacific (DeMartini and Friedlander 2004; Randall 2007; Timmers et al.

2011). The recurrent trend of genetic distinctness in this region can be attributed to three factors: (1) the geographic isolation of the islands, and oceanographic features, especially current patterns; (2) the life history characteristics of the fishes, such as dispersal capabilities; and (3) the extent of adaptive differentiation to environmental conditions in

Hawaii (Hourigan and Reese 1987).

Within the Hawaiian archipelago, the results for C. lunulatus are consistent with a few previous mtDNA surveys that revealed genetic differentiation across the archipelago in reef fishes (Rivera et al. 2004; Ramon et al. 2008; Eble et al. 2011a). The pairwise comparison between MHI and NWHI was highly significant (ΦST = 0.49; P < 0.001).

However, in most cases this population structure is observed in restricted-range Hawaiian endemics (Eble et al. 2009). The opposite scenario is more prevalent, in which widespread Indo-Pacific species exhibited genetic homogeneity across the archipelago (Craig et al. 2007; Andrews et al. 2010; Gaither et al. 2010, 2011a; Eble et al. 2011b; Reece et al. 2011; DiBattista et al. 2011, 2012b; Ludt et al. 2012). The results for C. lunulatus illuminate the demographic linkages between the low islands of the NWHI and the high volcanic islands of the MHI. In this case NWHI populations are isolated from the MHI, a partition that is observed in other reef organisms (Toonen et al. 2011). This is pertinent to management concerns as the NWHI is protected as the Papahanaumokuakea Marine

National Monument. In Chapter 3 I will further explore the connectivity between these ecologically distinct zones of the Hawaiian Archipelago.

Range Size and Marine Biogeography

Counter to an initial hypothesis, there was not a general correlation between range size and dispersal ability, supporting the premise that range size does not always predict genetic structure (Lester et al. 2007). Data from the wide-ranging C. lunulatus indicates strong population structure. While C. trifasciatus (widespread throughout the Indian

Ocean, and also expected to have a low level of genetic structure due to its large range) showed no significant genetic structure.

Data from C. austriacus (Red Sea) and C. melapterus (Persian Gulf and Gulf of

Oman) failed to detect any significant population structure, which may indicate that each is a single species-wide population. This can be explained by the fact that they both have limited distributions in the Western Indian Ocean, with no apparent barriers within each range.

In the subgenus phylogeny a clear separation is observed between the Pacific (C. lunulatus) and Indian Ocean species (C. trifasciatus, C. melapterus, and C. austriacus)

(Figure 2.1). The statistical parsimony network indicates that geminate species, C. lunulatus and C. trifasciatus, are separated by 28 mutational steps based on mtDNA cytochrome b sequences (Figure 2.2), corresponding to a sequence divergence of d =

6.0%. Based on the conventional molecular clock of 2%/MY, the mtDNA distance indicates approximately three million years of divergence. This is close to the onset of modern glacial cycles at 2.6 – 2.8 Ma (Dwyer et al. 1995; Williams et al. 1997).

Therefore, early in the separation of Pacific C. lunulatus and Indian C. trifasciatus, they appear to have been isolated by a glacial-era barrier between these two oceans. The shallow Sunda shelf among Indonesian islands is exposed during low sea levels, forming a long land bridge and restricting exchange between the tropical Indian Ocean and the western Pacific (Randall 1998; Bay et al. 2004; Rocha et al. 2007). Therefore it seems likely that transient allopatry had a role in the formation of these species pairs, a process that is apparent (or strongly suspected) in other Indian/Pacific species pairs (Briggs and

Bowen 2013; Gaither and Rocha 2013). Notably there are hybrids between the two species near the boundary of the Indian and Pacific Oceans (Hobbs et al. 2009; Montanari et al. 2011). Several studies have examined natural hybridization between C. lunulatus

29 and C. trifasciatus at Christmas Island in the Eastern Indian Ocean, although these occurrences are considered rare (Hobbs et al. 2009) and limited to F1 (first generation) hybrids with no evidence of backcrossing (Montanari et al. 2011).

Figure 2.2. Statistical parsimony network for Chaetodon lunulatus, C. trifasciatus, C. melapterus, and C. austriacus based on mtDNA cytochrome b sequences. Area of circles is proportional to the abundance of the respective haplotype: small circles indicate rare or unique haplotypes and the largest circles indicate the most common haplotypes observed in sampled individuals. Black bars represent the number of mutations and colors indicate haplotype sampling location (see key).

Cases of allopatric sister species in which their ranges are strongly asymmetric in size and separated by wide areas of inhospitable habitats are examples of “classical founder speciation” (Paulay and Meyer 2002; Hsu et al. 2007). In this study C. trifasciatus was shown to be monophyletic and clearly distinct from its limited range neighboring species, C. austriacus and C. melapterus, supporting the concept of founder

30 speciation in these two sister species. Furthermore, the Red Sea region was isolated from the Indian Ocean several times during the Pliocene-Pleistocene when sea levels dropped creating a barrier at the narrow entrance (Randall 1998; DiBattista et al. 2013), isolating the endemic C. austriacus from other members of Corallochaetodon (Figure 2.3).

Figure 2.3. Known marine biogeographic barriers (Blum 1989; Hsu et al. 2007), and their influence on the distribution range of Corallochaetodon.

Unexpectedly, the subgenus phylogeny of C. austriacus and C. melapterus indicates that these species are not monophyletic (Figure 2.1). Additionally, the statistical parsimony network shows that major haplotypes are shared between these two species and not separated by any mutational steps (Figure 2.2). However, these two putative species are genetically distinct at a population level (ΦST = 0.06; P = 0.001) indicating early stages of speciation, or distinct color morphs separated by habitat discontinuities. This finding should be interpreted in light of the relatively recent origins of reef faunas currently inhabiting the Red Sea (DiBattista et al. 2013) and Persian Gulf

(Sheppard et al. 2010). Estimated time since divergence is approximately 50,000 years, so that incomplete lineage sorting in isolation may be the source of our observations

(Table 2.5). The parapatry of C. austriacus and C. melapterus (Randall 1994) was likely initiated by a vicariant event at the Bab-al-Mandab barrier, the mouth of the Red Sea which becomes a shelf during low sea level stands (Kemp 1998). This barrier flooded

31 about 20 Ka, and so contemporary species divergence may be maintained by environmental conditions. Over time I predict these two color morphs may proceed to reciprocal monophyly based on two main observations. First, they live in different regions, with unique conditions including salinity and temperature (Sheppard 1998).

Second, two environmental barriers separate faunas of south Arabia and Red Sea (Kemp

1998). One is located in the western Gulf of Aden, restricting eastward colonization of

Red Sea species. A second is located in the southern third of the Red Sea, preventing northward colonization by southern Arabian species. Roberts et al. (1992) described the ecological discontinuity in this region based on reefs that are sparsely distributed, poorly developed, and subject to high turbidity that appear to restrict the distribution of the northern and central endemics (e.g. C. austriacus). The barrier does not seem to affect the distribution of the more widely distributed species, which might have a prolonged pelagic larval life, facilitating long-range dispersal (Zekeria et al. 2005). Finally, the striking color differences between these two “species” may induce assortative mating if theses two species came into secondary contact (McMillan et al. 1999).

Table 2.5. Estimation of divergence between species using a rate of 2% per million years. Mean genetic distance between species is below the diagonal, and divergence time in million of years is above the diagonal.

Implications of Specialization

There are repeated suggestions in the scientific literature that generalist feeders are more dispersive than specialists, in part because they can persist in a broader array of habitats

(Rocha et al. 2002). A natural trade-off exists between a species level of specialization and the ability of that species to cope with changes in resource availability (Mundy 2004;

Berumen and Pratchett 2008). Generalist species consume a wide range of coral species and have the ability to switch prey in the presence of alterations in their environment.

More specialized species sacrifice the ability to withstand fluctuations in resource ability, presumably because dietary specialization will yield greater fitness. All four species of the subgenus Corallochaetodon are obligate corallivores but have a differing degree of specialization. Pacific C. lunulatus has very generalized feeding preferences (Lawton et al. 2012), the Red Sea endemic C. austriacus is a generalist, while the restricted range C. melapterus and the widespread Indian C. trifasciatus are regarded as specialists. In the

Chagos Archipelago (central Indian Ocean), C. trifasciatus is considered highly specialized, feeding predominantly on acroporid corals, and is the second most abundant coral-feeding species compared to generalist counterparts (Pratchett et al. 2013b). In cases like this where specialist fishes predominate, preferred prey are abundant, enhancing juvenile growth, survivorship, and lifetime reproductive output (Berumen and

Pratchett 2008; Pratchett et al. 2013b). This in turn could explain the higher population connectivity in C. trifasciatus compared to C. lunulatus; higher dispersal ability due to greater fitness. However, field observations of two butterflyfishes (Chaetodon trifascialis and C. plebeius) at Lizard Island (northern Great Barrier Reef, Australia) have shown specialized species did not outperform the generalist species and dietary specialization appears to be a questionable strategy for long-term persistence (Berumen and Pratchett

2008). The differences observed between Chagos and Lizard Island show that it is difficult to draw general conclusions between diet specialization and population

33 connectivity, and one must carefully take into account reef conditions before doing so.

Many locations are showing substantial declines in coral cover and reef condition, and in degraded locations, it is likely that specialist coral feeding species will be replaced by more generalist species (Pratchett et al. 2013b). The unique environmental conditions that

C. melapterus and C. austriacus inhabit can almost certainly explain habitat specialization for these two species.

Specialization, body size, and life-history traits help determine the likelihood that a species will undergo local losses and population declines following disturbances, whereas information on range size, occupancy, and rarity indicates whether declines may lead to global extinction (Graham et al. 2011). In this case, I propose ecological specializations do influence range size, dispersal ability, and population connectivity in this subgenus, however additional research is needed to further evaluate this hypothesis.

CONCLUSION

There was no general correlation between range size and dispersal ability, as initially expected. These data show that even closely related species can have different levels of population structure, which is a proxy for realized dispersal ability (see also Bird et al.

2007). The pattern of evolutionary differentiation here is coupled with generally high gene flow in three of the four species. Evolutionary divergence among species may have been initiated along biogeographic boundaries such as the barrier between Indian and

Pacific Oceans, and between the Indian Ocean and Red Sea. Contemporary overlap indicates that species integrity may be reinforced by ecological factors such mate choice, habitat preferences, and the documented differences in coral prey base between species.

CHAPTER 3

INTRA-SPECIES DIFFERENTIATION IN THE BUTTERFLYFISH SUBGENUS

CORALLOCHAETODON (Chaetodon lunulatus and C. trifasciatus)

INTRODUCTION

Based on the mtDNA results in Chapter 2, population structure was observed in C. lunulatus in the Pacific Ocean but not C. trifasciatus in the Indian Ocean. To evaluate the robustness of this conclusion, 10 microsatellite loci were surveyed across each of these two widespread species. In some cases, microsatellites can reveal fine-scale population structure that is not detected with mtDNA assays, however this is not always the case

(DiBattista et al. 2012b; Karl et al. 2012). Regardless of the sensitivity of individual classes of markers, conclusions about population structure are more robust when evaluated across both the mitochondrial and nuclear genomes (Bowen et al. 2014).

Microsatellites are appropriate in this situation because they tend to be highly polymorphic and combining the results from many loci provides a more precise and statistically powerful way of comparing populations and individuals. In addition, microsatellite analysis is particularly useful for assessing present day demography or population connectivity patterns (Selkoe and Toonen 2006, 2011).

Studies of reef fishes in the Hawaiian Islands show that widespread Indo-Pacific species usually exhibited genetic homogeneity across this 2600km archipelago, while endemics can have significant population subdivision over the same range (Ramon et al.

2008; Eble et al. 2009, 2011a; Toonen et al. 2011). Additionally, previous investigation of phylogeographic relationships, genetic connectivity, and population history for

Hawaii’s endemic butterflyfishes (family Chaetodontidae), revealed no evidence for population subdivisions across the Hawaiian Archipelago (Craig et al. 2010). A comparison of these endemic species to other more widespread chaetodontid species occupying the same region, specifically Chaetodon lunulatus, will enhance the scientific foundations for managing Hawaii’s coral reef ecosystems.

Is the extent of the distribution range of fishes a good predictor of dispersal ability? A recent meta-analysis indicates that the degree of genetic structure increases as range size decreases, with restricted-range species showing a higher degree of genetic structuring than widespread species (Eble and Bowen unpublished).

C. melapterus and C. austriacus specimens were not used in this part of the study because the microsatellite primers used were created specifically for C. lunulatus, applying them to different species, even ones closely related as in this case, can be complicated by allele dropout, homoplasy, and other problems (Selkoe and Toonen

2006).

MATERIALS AND METHODS

A microsatellite primer database designed for Chaetodon species with large geographic ranges was developed by Lawton et al. (2010), including 16 microsatellite primers designed for C. lunulatus. In the present study both species were genotyped at 10 of these microsatellite loci (Table 3.1), based on the collections described in Chapter 2.

Polymerase chain reaction (PCR) mixes contained 5 ul of 2x BioMix Red solution

(BioMix Red; Bioline Ltd., London, UK), 0.15 ul (10 uM) of reverse primer and dye,

0.035 ul (10uM) of fluorescently labeled forward primer, and 5–50 ng template DNA in

10 ul total volume. PCRs included an initial denaturing step at 94°C for 5 min, then 35 cycles of amplification (30 s of denaturing at 94°C, 30 s of primer-specific annealing at temperatures provided in Table 3.1, and 90 s of extension at 72°C), followed by a final extension at 72°C for 10 min. PCR products were visualized through 1.5% agarose gel electrophoresis and compared to Hyperladder IV (Bioline, Taunton MA), a 100 bp standard size marker. All markers reliably amplified in both species and product sizes were consistent with expectations (Lawton et al. 2010; Montenari et al. 2011).

Repeat Ta Size Locus Label Motif Primer sequence (5'–3') (°C) Range

Lun03 PET (AG) TGTGTGTCACCACCTGGTCT 58 180-280 ACTCAGTTTTGAGCCGCTTC Lun05 6FAM (CAA) GCAACCCAGTCTCACATCAA 55 170-210 TCTGCTATTTCACAATTTTAGAGCA Lun07 NED (TG) AAGTGCCCTTTAGCAAAGCA 58 160-220 CTCCAGTCGCTTTCTGTGTG Lun08 NED (CA) GGCCTTTGTTTGTGGTCATT 55 180-240 CCTGAAGAGAGAGCTGCTCAA Lun09 6FAM (TG) CCTGTGTTTGTCATCCAACG 58 170–190 CTTTGGGACACACACTTCCA Lun14 VIC (TCA) TACGTTGGACAGTGGCTGTG 58 210-280 TGGCTCTGTGGCATGTATGT Lun20 6FAM (CTT) CAGTGTCGGAGAACAACGAA 58 220-280 TCACTGTGTCACCAATGCAC Lun29 6FAM (AC) CACCCACAGGCAGTGTATTG 55 230-290 GCCAGCCTGTCAAAACTTTA Lun34 PET (CA) CATGCTTGGGTGAGCATGTA 58 180–210 TGTGCGTTTGTGCAAGTGTA Lun36 VIC (GT) GCGTTTGACTTCACGTTTCA 58 200-270 TGCAAAACAACAACCTACGG Table 3.1. Characteristics of microsatellites used in this study, developed for Chaetodon lunulatus by Lawton et al. 2010. Primer sequences, repeat motifs, and annealing temperatures (Ta) are given as per the authors of the original study. Fluorescent dye labels and size ranges are from the present study.

PCR products labeled with different fluorescent dye colors were pooled for genotyping and resolved using an ABI 3130XL Genetic Analyzer (Applied Biosystems) along with a fluorescently labeled internal size standards (LIZ-500; Applied Biosystems) following the methods outlined by DiBattista et al. (2012b). Allele sizes were assigned using the Geneious Pro (BioMatters) version 5.6.7 microsatellite plug in.

Initially specimens from Hawaii were separated into individual sampling locations by island. However mtDNA cyt b sequence data revealed significant structure within the Hawaiian archipelago with a primary genetic break between the Main

Hawaiian Island (MHI) and North West Hawaiian Island (NWHI) concordant with a comparative analysis of multispecies connectivity in the archipelago (Toonen et al.

2011). As a result, for subsequent analysis sampling sites in Hawaii were partitioned into two groups; MHI and NWHI. However, a full comparison among Hawaiian sample sites is provided in the appendix (Table A.1).

For each locus the mean number of alleles (NA), observed (HO), and expected

(HE) heterozygosities, departure from Hardy-Weinberg proportions (HWE), and linkage disequilibrium (LD) were assessed with GENEPOP version 4.2 (Raymond and Rousset

1995). MICRO-CHECKER version 2.2.3 was used to identify genotyping errors including the possible presence of null alleles, allelic dropouts, and excessive stutter peaks (Van Oosterhout et al. 2004), and significance levels for multiple comparisons were adjusted using the sequential Bonferonni correction. GENODIVE version 2.0b23

(Meirmans and Tienderen 2004) was used to estimate overall population structure for each species separately and to estimate pairwise comparisons (FST calculations).

STRUCTURE version 2.3.4 was used to assign individuals to distinct genetic clusters (populations) without presumption from predefined geographic locations

(Pritchard at al. 2000). STRUCTURE uses a Bayesian approach to assign individual multilocus genotypes to one or more clusters by minimizing deviations from Hardy-

Weinberg and linkage equilibrium. The most likely number of clusters in the dataset was identified based on the probability of K=1 to K= 12 or K=1 to K=4 for C. lunulatus and

C. trifasciatus respectively. Analyses were repeated five times and results averaged. Each replicate run consisted of 1,000,000 MCMC repetitions, a burn-in of 10,000 iterations, and assumed correlated allele frequencies and admixed populations (as per DiBattista et al. 2012b). STRUCTURE HARVESTER version 0.6.93 was used to determine most likely value of K. This program processes STRUCTURE results and uses the “Evanno” method (Evanno et al. 2005) to provide a fast way to assess and visualize likelihood values across multiple values of K and hundreds of iterations for easier detection of the number of genetic groups that best fit the data (Earl and vonHoldt 2012).

RESULTS

Chaetodon lunulatus

Significant population structure was detected for C. lunulatus (FST = 0.05, P = 0.001).

Microsatellite allele frequencies (FST) were significantly different in 49 out of 91 comparisons for C. lunulatus (Table 3.2; see also Table A.1 in the appendix). Four different locations yielded consistently significant FST values: Moorea, Johnston Atoll,

MHI, and NWHI. See Table 3.3 for Microsatellite statistics averaged over all loci for each location and species.

Table 3.2. Matrix of population pairwise FST values and associated P values based on microsatellite genotypes for Chaetodon lunulatus. FST values are above the diagonal and P values are below the diagonal. Significant P values are highlighted in bold (P < 0.05).

The mean number of alleles per locus was 17.9 (range: 7 to 29 alleles), allelic richness was 3.56 (range: 1.54 to 5.44), and observed heterozygosity ranged from 0.54

(Lun 14) to 0.95 (Lun 3). Few loci deviated from Hardy-Weinberg equilibrium for within site comparisons (7 out of 140, P < 0.05). Linkage disequilibrium was detected in 2 out of

45 within site comparisons after correcting for multiple tests: Moorea Lun 14 and Lun 34,

Johnston Atoll Lun 8 and Lun 20 (P < 0.011 across all populations), but none of these deviations are consistent within loci or sites.

MICROCHECKER analysis revealed no evidence for scoring error due to stuttering, and no evidence for large allele dropout. Evidence of null alleles was detected in only 12 out of 140 within site comparisons, although one locus was disproportionately represented, Lun 20. Therefore analyses were run including and excluding this locus; finding were no different between datasets and so results from all 10 loci are presented.

Overall, there was no consistent evidence for departure from HWE, linkage disequilibrium, or null alleles across all sampled locations, supporting the decision to retain and examine the entire data set.

N NA HO HE C. lunulatus Christmas Island (I) 6 6.3 0.883 0.824 Indonesia 27 11.1 0.802 0.795 American Samoa 18 8.5 0.744 0.764 Fiji 37 12.2 0.784 0.79 Kanton Island 17 9.8 0.831 0.79 Marshall Islands 54 13.6 0.785 0.795 Moorea 32 9.5 0.776 0.729 Okinawa 14 8.7 0.752 0.808 Pohnpei 30 10.7 0.793 0.787 Kiribati 40 12.7 0.779 0.783 Palau 33 12.3 0.788 0.795 Johnston Atoll 42 8.6 0.731 0.695 MHI 50 9 0.782 0.747 NWHI 203 13.3 0.688 0.671 C. trifasciatus Diego Garcia 29 11.6 0.686 0.775 Seychelles 34 14.3 0.764 0.800 Australia 58 15.0 0.757 0.811 Indonesia 17 9.9 0.765 0.800 Table 3.3. Microsatellite statistics averaged over all loci for each location and species: Number of samples (N), number of alleles (NA), observed heterozygosities (HO) and expected heterozygosities (HE).

STRUCTURE identified mean probabilities as being highest at K = 3 (Figure

3.1), and K = 3 was verified using STRUCTURE HARVESTER (Figure 3.2).

Figure 3.1. STRUCTURE bar plot for Chaetodon lunulatus, showing the highest mean probability of K = 3. Locations: 1 Christmas Island (I), 2 Indonesia, 3 Palau, 4 Okinawa, 5 Pohnpei, 6 Marshall Islands, 7 Fiji, 8 American Samoa, 9 Moorea, 10 Kanton Island, 11 Kiribati, 12 Johnston Atoll, 13 MHI, 14 NWHI.

Figure 3.2 STRUCTURE HARVESTER analysis used to determine that K = 3 is the most likely value for Chaetodon lunulatus.

Chaetodon trifasciatus

Low but significant population structure was detected for C. trifasciatus (FST = 0.003, P =

0.03). Microsatellite allele frequencies were significantly different in 3 out of 6 comparisons (Table 3.4).

The mean number of alleles per locus was 17.9 (range: 7 to 29 alleles), allelic richness was 5.99 (range: 3.04 to 9.92), and observed heterozygosity ranged from 0.41

(Lun 36) to 0.91 (Lun 3). Linkage Disequilibrium was detected in only one pair of loci

(Australia, Lun 3 and Lun 9) out of 45 within site comparisons after correcting for multiple tests (adjusted P < 0.001). The test for Hardy-Weinberg disequilibrium was significant in 7 out of 40 within site comparisons (adjusted P < 0.005).

Table 3.4. Matrix of population pairwise FST values and associated P values based on microsatellite genotypes for Chaetodon trifasciatus. FST values are above the diagonal and P values are below the diagonal. Significant P values are highlighted in bold (P < 0.05).

Microchecker analysis again revealed no evidence for scoring error due to stuttering, or large allele dropout. Evidence of null alleles was detected in only 9 out of

40 within site comparisons; again 2 loci were disproportionately represented, Lun 7 and

Lun 20. Therefore analyses were run including and excluding this locus; again findings were no different between datasets and so results from all 10 loci are presented.

To test for the possible impact of null alleles in Lun 7 and Lun 20 on these results,

I also ran FreeNA which estimates unbiased FST (Weir 1996) following the excluding null alleles (ENA) method described in Chapuis and Estoup (2007). This test resulted in

FST = 0.0038 not using ENA, and FST = 0.0044 using ENA.

STRUCTURE identified mean probabilities as being highest at K = 2, which is consistent with the results from STRUCTURE HARVESTER (Figure 3.3).

Figure 3.3 STRUCTURE HARVESTER analysis used to determine that the most likely value was K = 2 for Chaetodon trifasciatus.

DISCUSSION

Chaetodon lunulatus

Microsatellite analyses for C. lunulatus were consistant with the mtDNA results in indicating divergent populations at MHI and Johnston Atoll, NWHI, and Moorea. All

43 lines of evidence indicate that the large range of C. lunulatus is subdivided into isolated populations at the extremes of the range. The homogenous region through the Central-

West Pacific range, the Indo-Polynesian Biogeographic Province, is speckled with atolls and islands, so that there is no distance greater than 800 km between suitable shallow reef habitat (Schultz et al. 2008; Briggs and Bowen 2012). However, the peripheral isolation is not maintained by distance alone, as prevailing currents also work against colonization of Hawaii and French Polynesia (Hourigan and Reese 1987; Gaither et al. 2010).

Furthermore, this large range encompasses spatial, temporal, and environmental heterogeneity in coral-reef habitats.

An interesting outcome is the separation between the high islands of the MHI and the low islands and atolls of the NWHI, indicating some kind of barrier between these regions. This may be due to a lack of appropriate habitat between these regions, specifically abundant coral cover and areas protected from high wave energy. The specialist C. lunulatus prefers areas with greater than 60-70% coral cover. Notably, at the

MHI region adjacent to this break (Kauai) our own efforts and previous transect data indicate a near-complete absence of C. lunulatus. This provides an obvious explanation for the geographic partition on the finest scale observed in this study, although additional fine-scale breaks may have eluded detection with our sampling design. In understanding the population genetic partitions in this species, we return to an earlier biogeographic appraisal of the Hawaiian Islands, which identified three factors defining the distribution of reef fishes: “ (1) the geographic isolation of the islands, and oceanographic features, especially current patterns; (2) the life history characteristics of the fishes, especially their

44 dispersal capabilities; and (3) the extent of adaptive differentiation to environmental conditions after they reached Hawaii” (Hourigan and Reese 1987).

Chaetodon trifasciatus

This species shows low but significant overall structure in microsatellites (trifasciatus

(FST = 0.003, P = 0.03), but not with mtDNA (ΦST = -0.01; P = 0.50). There is a persistent but fallacious belief in the field of population genetics that the hypervariable microsatellites will show population structure that may be missed in mtDNA assays (Karl et al. 2012). However, pragmatic comparisons of these two approaches show that both can reveal population structure not detected by the other assay, depending on a variety of conditions (DiBattista et al. 2012b). Therefore it is sufficient to conclude that these findings demonstrate the value of using multiple loci for population assessment.

The microsatellite data for C. trifasciatus indicate that the population at Diego

Garcia (Chagos Archipelago), in the middle of the Indian Ocean, is statistically differentiated from samples sites to the east and west. Part of this outcome may be explained by sparse sampling, especially at Indonesia. However, Chagos is also an unusual location in several respects. First, based on faunal composition the Chagos aligned with the Western Indian Ocean biogeographic province, but also shares some faunal elements with the Indo-Polynesian Province to the east (Winterbottom and

Anderson 1997; Craig 2008; Briggs and Bowen 2012). Second, the Chagos is subject to seasonal monsoon-driven currents that switch direction between easterly and westerly, possibly limiting larval dispersal to this location (Figure 3.4, Sheppard et al. 2012). C. trifasciatus spawn from August-January and March-May (Vijay Anand and Pillai 2002)

45 so it is possible that larvae are exposed to unfavorable conditions by these changing currents, which could also effect nutrients, temperatures and larval dispersal ability and ultimately have an isolating effect. Third, the population genetic separation of Chagos has been observed in other reef fauna (Gaither et al. 2010; Eble et al. 2011b; Vogler et al.

2012), indicating that the isolation of this mid-ocean archipelago is a general feature of the reef fauna, akin to that observed for Hawaii (DiBattista et al. 2011; Eble et al. 2011a;

Gaither et al. 2011b).

Figure 3.4. Prevailing currents in the north-western Indian Ocean. (left) Winter: November – March. (right) Summer: April – October (figure from Kemp 1998).

CONCLUSION

Genetic data indicate that the Indian Ocean C. trifasciatus shows less structure and is perhaps more dispersive than the Pacific sister species C. lunulatus, regardless of the assay. This conclusion is somewhat counter-intuitive for at least three reasons. First, C. lunulatus has a larger range, which can be associated with higher dispersal ability (Eble and Bowen, unpublished). Second, the Pacific domain of C. lunulatus has many more

46 island groups to provide stepping stones across the species range. Third, C. trifasciatus is a dietary specialist relative to C. lunulatus. All three factors would predict greater dispersal in C. lunulatus, although a counterargument could be made to point three, that specialists need greater dispersal to find appropriate habitat. While the explanation for this pattern is not clear, there is one pertinent observation in the pattern of population subdivisions: with the possible exception of Fiji, all the isolated populations identified here are either at the margins of the range (French Polynesia), at an isolated mid-oceanic archipelago (Chagos), or both (Hawaii). Recent studies indicate that such peripheral marine habitats can be engines of biodiversity (Duda and Lee 2009; Budd and Pandolfi

2010; Bird et al. 2011; Bowen et al. 2013; Baums et al. 2014). If population genetic separations are the starting point for speciation, then the evolutionary hotspots for corallivore butterflyfishes may be at these isolated and peripheral habitats.

CHAPTER 4

FINAL SUMMARY

The value of comparative phylogeography is clearly shown here, and is supported by the power of multiple species comparisons, where multiple species serve as independent tests of regional environmental and geological processes (Toonen et al. 2011; Drew and

Barber 2012). While all four species are closely related and have similar life histories they exhibit contrasting population genetic signatures emphasizing “the hazards of making sweeping predictions about population connectivity” (Bird et al. 2007). This study shows that patterns of differentiation can vary within a single species, and closely related species, reinforcing the need to identify and protect unique lineages in marine species. It also allowed us to determine populations of obligate corallivores that can and should be defined as distinct management units. The definition of these management units will indicate how these corallivore specialists will react to declining resource availability. Obligate corallivores are considered good “indicator organisms” because changes in their distribution and abundance should reflect conditions on the reef (Reese

1981). Butterflyfish can be particularly effective indicator organisms since they are relatively large and conspicuous. Overall, determining the degree to which populations interact is critical for understanding how species are formed, how communities are maintained and how conservation efforts should be prioritized on Indo-Pacific coral reefs

(Eble et al. 2009).

APPENDIX

Table A.1. Pairwise comparisons within Hawaiian sampling locations: mtDNA ΦST values are above the diagonal, and microsatellite FST values are below the diagonal. Significant P-values values are highlighted in bold (P<0.05).

LITERATURE CITED

Allen, G.R., R.C. Steene, and M. Allen. 1998. A guide to angelfishes & butterflyfishes. Odyssey Publishing/Tropical Reef Research, Perth.

Almany, G.R., M.L. Berumen, S.R. Thorrold, S. Planes, and G.P. Jones. 2007. Local replenishment of coral reef fish populations in a marine reserve. Science 316: 742–744.

Alwany, M., E. Thaler, and M. Stachowitsch. 2003. Food selection in two corallivorous butterflyfishes, Chaetodon austriacus and C. trifascialis, in the Northern Red Sea. Marine Ecology 24: 165–177.

Andrews, K.R., L. Karczmarski, W.W.L. Au, S. Rickards, C.A. Vanderlip, B.W. Bowen, and R.J. Toonen. 2010. Rolling stones and stable homes; Social structure, habitat diversity, and population genetics of the Hawaiian spinner dolphin (Stenella longirostris). Molecular Ecology 19: 732-748.

Bandelt, H.J., P. Forster, A. Röhl. 1999. Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution 16: 37–48.

Baums, I.B., L.S. Godwin, E.C. Franklin, D.B. Carlon, and R.J. Toonen. 2014. Discordant population expansions in four species of coral-associated Pacific hermit crabs (Anomura: Diogenidae) linked to habitat availability resulting from sea-level change. Journal of Biogeography 41: 339–352.

Bay, L.K., J.H. Choat, L. van Herwerden, and D.R. Robertson. 2004. High genetic diversities and complex genetic structure in an Indo-Pacific tropical reef fish (Chlorurus sordidus): evidence of an unstable evolutionary past? Marine Biology 144: 757–767.

Bellwood, D.R., S. Klanten, P.F. Cowman, M.S. Pratchett, N. Konow, and L. van Herwerden. 2009. Evolutionary history of the butterflyfishes (f: Chaetodontidae) and the rise of coral feeding fishes. Journal of Evolutionary Biology 23: 335– 349. 50

Berumen, M.L. and M.S. Pratchett. 2008. Trade-Offs Associated with Dietary Specialization in Corallivorous Butterflyfishes (Chaetodontidae: Chaetodon). Behavioral Ecology and Sociobiology 62: 989-994.

Bird, C.E., B.S. Holland, B.W. Bowen, and R.J. Toonen. 2007. Contrasting phylogeography in three endemic Hawaiian limpets (Cellana spp.) with similar life histories. Molecular Ecology 16: 3173–3186.

Bird, C.E., S.A. Karl, P.E. Smouse, and R.J. Toonen. 2011. Detecting and measuring genetic differentiation. In: Crustacean Issues: Phylogeography and Population Genetics in Crustacea (eds Koenemann S, Held C, Schubart C), pp. 31–55. CRC Press, Boca Raton, Florida.

Blum, S.D. 1989. Biogeography of the Chaetodontidae: an analysis of allopatry among closely related species. Environmental Biology of Fishes 25: 9-31.

Bowen, B.W., A.L. Bass, L.A. Rocha, W.S. Grant, and D.R. Robertson. 2001. Phylogeography of the trumpetfishes (Aulostomus): Ring species complex on a global scale. Evolution 55: 1029–1039.

Bowen, B.W., L.A. Rocha, R.J. Toonen, S.A. Karl, and The ToBo Laboratory. 2013. The origins of tropical marine biodiversity. Trends in Ecology & Evolution 28: 359-366.

Bowen, B.W., K. Shanker, N. Yasuda, M.C.D. Malay, S. von der Heyden, G. Paulay, L.A. Rocha, K.A. Selkoe, P.H. Barber, S.T. Williams, H.A. Lessios, E.D. Crandall, G. Bernardi, C.P. Meyer, K.E. Carpenter, and R.J. Toonen. 2014. Phylogeography unplugged: Comparative surveys in the genomic era. Bulletin of Marine Science 90:13- 46. DOI: 10.5343.

Briggs, J.C. 1974. Marine zoogeography. McGraw-Hill, New York.

Briggs, J.C., and B.W. Bowen. 2012. A realignment of marine biogeographic provinces with particular reference to fish distributions. Journal of Biogeography 39: 12–30.

Briggs, J.C., and B.W. Bowen. 2013. Evolutionary patterns: Marine shelf habitat. Journal of Biogeography 40: 1023–1035. doi:10.1111/jbi.12082.

Budd, A.F., and J.M. Pandolfi. 2010. Evolutionary novelty is concentrated at the edge of coral species distributions. Science 328: 1558-1561.

Burgess, W.E. 1978. Butterflyfishes of the world: a monograph of the family Chaetodontidae. T. F. H. Publications, Neptune, N.J.

Chapuis, M.P. and A. Estoup. 2007. Microsatellite Null Alleles and Estimation of Population Differentiation. Molecular Biology and Evolution 24: 621–631.

Craig, M.T., J.A. Eble, B.W. Bowen, and D.R. Robertson. 2007. High genetic connectivity across the Indian and Pacific Oceans in the reef fish Myripristis berndti (Holocentridae). Marine Ecology Progress Series 334: 245–254.

Craig, M.T. 2008. The goldrim surgeonfish (Acanthurus nigricans; Acanthuridae) from Diego Garcia, Chagos Archipelago: first record for the central Indian Ocean. Zootaxa 1850: 65–68.

Craig, M.T., J.A. Eble, and B.W. Bowen. 2010. Origins, ages, and population histories: Comparative phylogeography of endemic Hawaiian butterflyfishes (genus Chaetodon). Journal of Biogeography 37: 2125 – 2136.

Crandall, E.D., M.A. Frey, R.K. Grosberg, and P.H. Barber. 2008. Contrasting demographic history and phylogeographical patterns in two Indo-Pacific gastropods. Molecular Ecology 17: 611-626.

DeMartini, E.E. and A.M. Friedlander. 2004. Spatial patterns of endemism in shallow- water reef fish populations of the Northwestern Hawaiian Islands. Marine Ecology Progress Series 271: 281-296.

DiBattista, J.D., C. Wilcox, M.T. Craig, L.A. Rocha, and B.W. Bowen. 2011. Phylogeography of the Pacific Blueline Surgeonfish Acanthurus nigroris reveals a cryptic species in the Hawaiian Archipelago. Journal of Marine Biology, Article ID 839134.

DiBattista J.D., E.M. Waldrop, B.W. Bowen, J.K. Schultz, M.R. Gaither, R.L. Pyle, and L.A. Rocha. 2012a. Twisted sister species of pygmy angelfishes: discordance between taxonomy, coloration, and phylogenetics. Coral Reefs 31: 839–851.

DiBattista, J.D., L.A. Rocha, M.T. Craig, K.A. Feldheim, and B.W. Bowen. 2012b. Phylogeography of two closely related Indo-Pacific butterflyfishes reveals divergent evolutionary histories and discordant results from mtDNA and microsatellites. Journal of Heredity. 103: 617–629.

DiBattista, J.D., M.L. Berumen, M.R. Gaither, L.A. Rocha, J.A. Eble, J.H. Choat, M.T. Craig, D.J. Skillings, and B.W. Bowen. 2013. After continents divide: comparative phylogeography of reef fishes from the Red Sea and Indian Ocean. Journal of Biogeography 40: 1170-1181.

Drew J., G.R. Allen, and L. Kaufman. 2008. Endemism and regional color and genetic differences in five putatively cosmopolitan reef fishes. Conservation Biology 22: 965– 975.

Drew, J.A., and P.H. Barber. 2012. Comparative phylogeography in Fijian coral reef fishes: A multi-taxa approach towards marine reserve design. PLoS ONE 7(10): e47710. doi:10.1371/journal.pone.0047710.

Drummond, A.J., B. Ashton, S. Buxton, M. Cheung, A. Cooper, C. Duran, M. Field, J. Heled, M. Kearse, S. Markowitz, R. Moir, S. Stones-Havas, S. Sturrock, T. Thierer, and A. Wilson. 2010. Geneious version 5.4, Available from http://www.geneious.com.

Duda, T.F., and T. Lee. 2009. Isolation and population divergence of a widespread Indo- West Pacific marine gastropod at Easter Island. Marine Biology 156: 1193-1202.

Dwyer, G.S., T.M. Cronin, P.A. Baker, M.E. Raymo, J.S. Buzas, and T. Correge. 1995. North Atlantic deepwater temperature change during late Pliocene and late Quaternary climatic cycles. Science 270: 1347–1351.

Earl, D.A., and B.M. vonHoldt. 2012. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources 4: 359-361.

Eble, J.A., R.J. Toonen, and B.W. Bowen. 2009. Endemism and dispersal: comparative phylogeography of three surgeonfishes across the Hawaiian Archipelago. Marine Biology 156: 689-698.

Eble, J.A., R.J. Toonen, L.L. Sorensen, L. Basch, Y. Papastamatiou, and B.W. Bowen. 2011a. Escaping paradise: Larval export from Hawaii in an Indo-Pacific reef fish, the Yellow Tang (Zebrasoma flavescens). Marine Ecology Progress Series 428: 245–258.

Eble, J.A., L.A. Rocha, M.T. Craig, and B.W. Bowen. 2011b. Not all larvae stay close to home: Long-distance dispersal in Indo-Pacific reef fishes, with a focus on the Brown Surgeonfish (Acanthurus nigrofuscus). Journal of Marine Biology, Article ID 518516.

Ekman, S. 1953. Zoogeography of the sea. Sidgwick and Jackson, London.

Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14: 2611–2620.

Excoffier, L., G. Laval, and S. Schneider. 1992. Arlequin (version 3.0): An integrated software package for population genetics data analysis. Evolutionary bioinformatics 1: 47-50.

Excoffier, L., G. Laval, and S. Schneider. 2005. Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evolutionary bioinformatics Online 1: 47–50.

Fessler, J.L. and M.W. Westneat. 2007. Molecular phylogenetics of the butterflyfishes (Chaetodontidae): Taxonomy and biogeography of a global coral reef fish family. Molecular Phylogenetics and Evolution 45: 50–68.

Fu, Y.X. 1997. Statistical tests of neutrality of mutations against population growth, hitchhiking, and background selection. Genetics 147: 915-925.

Gaither, M.R., R.J. Toonen, D.R. Robertson, S. Planes, and B.W. Bowen. 2010. Genetic evaluation of marine biogeographical barriers: perspectives from two widespread Indo- Pacific snappers (Lutjanus kasmira) and (Lutjanus fulvus). Journal of Biogeography 37: 133–147.

Gaither, M.R., B.W. Bowen, T.R. Bordenave, L.A. Rocha, S.J. Newman, J.A. Gomez, L. van Herwerden, and M.T. Craig. 2011a. Phylogeography of the reef fish Cephalopholis argus (Epinephelidae) indicates Pleistocene isolation across the Indo-Pacific barrier with contemporary overlap in the coral triangle. BMC Evolutionary Biology 11: 189.

Gaither, M.R., S.A. Jones, C. Kelley, S.J. Newman, L. Sorenson, and B.W. Bowen. 2011b. High connectivity in the deepwater snapper Pristipomoides filamentosus

(Lutjanidae) across the Indo-Pacific with isolation of the Hawaiian Archipelago. PLoS One 6(12): e28913. doi:10.1371/journal.pone.0028913.

Gaither, M.R. and L.A. Rocha. 2013. Origins of species richness in the Indo-Malay- Philippine biodiversity hotspot: evidence for the center of overlap hypothesis. Journal of Biogeography 40: 1638–1648.

Goren, M., R. Gvili, and B.S. Galil. 2011. The reef-associated butterfly fish Chaetodon austriacus Ruppell, 1836 in the Mediterranean: The implications of behavioral plasticity for bioinvasion hazard assessment. Aquatic Invasions 6: S143 – 145.

Graham, N.A.J. 2007. Ecological versatility and the decline of coral feeding fishes following climate driven coral mortality. Marine Biology 153: 119-127.

Graham, N.A.J., P. Chabanet, R.D. Evans, S. Jennings, Y. Letourneur, M.A. MacNeil, T.R McClanahan, M.C. Öhman, N.V.C. Polunin, and S.K. Wilson. 2011. Extinction vulnerability of coral reef fishes. Ecology Letters 14: 341-348.

Hasegawa, M., H. Kishino, and T. Yano. 1985. Dating of the human-ape splitting by a molecular clock of Mitochondrial DNA. Journal of Molecular Evolution 22: 160-174.

Hobbs, J.P.A., A.J. Frisch, G.R. Allen, and L. van Herwerden. 2009. Marine hybrid hotspot at Indo-Pacific biogeographic border. Biology Letters 5: 258–61.

Hourigan, T.F., and E.S. Reese. 1987. Mid-ocean isolation and the evolution of Hawaiian reef fishes. Trends in Ecology and Evolution 2: 187-191.

Hsu, K.C., J.P. Chen, and K.T. Shao. 2007. Molecular phylogeny of Chaetodon (Teleostei: Chaetodontidae) in the Indo-West Pacific: evolution in geminate species pairs and species groups. The Raffles Bulletin of Zoology Supplement 14: 77-86.

Johns, G.C., and J.C. Avise. 1998. A comparative summary of genetic distances in the vertebrates from the mitochondrial cytochrome b gene. Molecular Biology and Evolution 15: 1481–90.

Jones, G.P., M.J. Milicich, M.J. Emslie, and C. Lunow. 1999. Self-recruitment in a coral reef fish population. Nature 402: 802–804.

Karl, S.A., R.J. Toonen, W.S. Grant, and B.W. Bowen. 2012. Common misconceptions in molecular ecology: Echoes of the modern synthesis. Molecular Ecology 21: 4171–89.

Kemp, J.M. 1998. Zoogeography of the coral reef fishes of the Socotra Archipelago. Journal of Biogeography 25: 919–933.

Kobayashi, D.R. 2006. Colonization of the Hawaiian Archipelago via Johnston Atoll: a characterization of oceanographic transport corridors for pelagic larvae using computer simulation. Coral Reefs 25: 407-417.

Kosaki, R.K., R.L. Pyle, J.E. Randall, and D.K. Irons. 1991. New records of fishes from Johnston Atoll, with notes on biogeography. Pacific Science 45: 186-203.

Lawton, R.J., L.K. Bay, and M.S. Pratchett. 2010. Isolation and characterization of 29 microsatellite loci for studies of population connectivity in the butterflyfishes Chaetodon trifascialis and Chaetodon lunulatus. Conservation Genetics Resources 2: 209 – 213.

Lawton, R.J., M.S. Pratchett, and L.K. Bay. 2011a. Cross-species amplification of 44 microsatellite loci developed for Chaetodon trifascialis, C. lunulatus and C. vagabundus in 22 related butterflyfish species. Molecular Ecology Resources 11: 323–327.

Lawton, R.J., V. Messmer, M.S. Pratchett, and L.K. Bay. 2011b. High gene flow across large geographic scales reduces extinction risk for a highly specialised coral feeding butterflyfish. Molecular Ecology 20: 3584-3598.

Lawton, R.J., A.J. Cole, M.L. Berumen, and M.S. Pratchett. 2012. Geographic variation in resource use by specialist versus generalist butterflyfishes. Ecography 35: 566–576.

Leis, J.M. 2002. Pacific coral-reef fishes: The implications of behaviour and ecology of larvae for biodiversity and conservation, and a reassessment of the open population paradigm. Environmental Biology of Fishes 65: 199-208.

Lester, S.E, B.I. Ruttenberg, S.D. Gaines, and B.P. Kinlan. 2007. The relationship between dispersal ability and geographic range size. Ecology Letters 10: 745–58.

Lieske, E., and R. Myers. 1996. Coral Reef Fishes. Princeton, NJ : Princeton University Press.

Littlewood, D.T.J., S.M. McDonald, A.C. Gill, and T.H. Cribb. 2004. Molecular phylogenetics of Chaetodon and the Chaetodontidae (Teleostei: Perciformes) with reference to morphology. Zootaxa 779: 1–20.

Ludt, W.B., M. Bernal, B.W. Bowen, and L.A. Rocha. 2012. Living in the past: Phylogeography and population histories of Indo-Pacific wrasses (Genus Halichoeres) in shallow lagoons versus outer reef slopes. PLoS One 7: e38042 doi: 10.1371/journal.pone.0038042.

Maragos, J.E., and P.L Jokiel. 1986. Reef corals of Johnston Atoll: One of the worlds most isolated reefs. Coral Reefs 4: 141-150.

McMillan, W.O., L.A. Weight, and S.R. Palumbi. 1999. Color pattern evolution, assortative mating, and genetic differentiation in brightly colored butterflyfishes (Chaetodontidae). Evolution 53: 247-260.

Meeker, N.D., S.A. Hutchinson, L. Ho, and N.S. Trede. 2007. Method for isolation of PCR-ready genomic DNA from zebrafish tissues. BioTechniques 43: 610-614.

Meirmans, P.G., and P.H. Van Tienderen. 2004. GENOTYPE and GENODIVE: Two programs for the analysis of genetic diversity of asexual organisms. Molecular Ecology Notes 4: 92-794.

Montenari, S.R., L. van Herwerden, M.S. Pratchett, J.P.A Hobbs, and A. Fugedi. 2011. Reef fish hybridization: lessons learnt from butterflyfishes (genus Chaetodon). Ecology and Evolution. 2: 310-328.

Mundy, P.L. 2004. Habitat loss, resource specialization, and extinction on coral reefs. Global Change Biology 10: 1642–1647.

Myers, R.F. 1999. Micronesian Reef Fishes. Third revised and expanded edition. Coral Graphics, Guam. 330 pp.

Nakamura, S.Y., T. Shibuno, and K. Yamaoka. Relationship between pelagic larval duration and abundance of tropical fishes on temperate coasts of Japan. Journal of Fish Biology 80: 346-357.

Palumbi, S.R. 1994. Genetic divergence, reproductive isolation, and marine speciation. Annual Review of Ecology and Systematics 25: 547-572.

Palumbi, S.R. 1997. Molecular biogeography of the Pacific. Coral Reefs 16, Suppl.: S47—S52.

Paulay, G., and C. Meyer 2002. Diversification in the tropical Pacific: Comparisons between marine and terrestrial systems and the importance of founder speciation. Integrative and Comparative Biology 49: 922-934.

Planes, S. 1993. Genetic differentiation in relation to restricted larval dispersal of the convict surgeonfish Acanthurus triostegus in French Polynesia. Marine Ecology Progress Series 98: 237-246.

Posada, D. 2008. jModelTest: Phylogenetic model averaging. Molecular Biology and Evolution 25: 253–1256.

Pratchett, M.S. 2005. Dietary overlap among coral-feeding butterflyfishes (Chaetodontidae) at Lizard Island, northern Great Barrier Reef. Marine Biology 148: 373–382.

Pratchett, M.S., A.S. Hoey, D.A. Feary, A.G. Bauman, J.A. Burt, and B.M. Riegl. 2013a. Functional composition of Chaetodon butterflyfishes at a peripheral and extreme coral reef location, the Persian Gulf. Marine Pollution Bulletin 72: 333–341.

Pratchett, M.S., N.A.J. Graham, and A.J. Cole. 2013b. Specialist corallivores dominate butterflyfish assemblages in coral-dominated reef habitats. Journal of Fish Biology 82: 1177–1191.

Pritchard, J.K., M. Stephens, and P. Donnelly. 2000. Inference of population structure using multilocus genotype data. Genetics 155: 945–959.

Ramon, M.L., P.A. Nelson, E. De Martini, W.J. Walsh, and G. Bernardi. 2008. Phylogeography, historical demography, and the role of post-settlement ecology in two Hawaiian damselfish species. Marine Biology 153: 1207-1217.

Randall, J.E. 1994. Twenty-two new records of fishes from the Red Sea. Fauna Saudi Arabia 14: 259–275.

Randall, J.E. 1998. Zoography of shore fishes of the Indo-Pacific region. Zoological Studies 37: 227-268.

Randall, J.E. 2007. Reef and Shore Fishes of the South Pacific. University of Hawaii Press, Honolulu.

Raymond, M., and F. Rousset. 1995. GENEPOP (version 1.2): Population genetics software for exact tests and ecumenicism. Journal of Heredity 86: 248-249.

Reece, J.S., B.W. Bowen, and A.F. Larson. 2011. Long larval duration in moray eels (Muraenidae) ensures ocean-wide connectivity despite differences in adult niche breadth. Marine Ecology Progress Series 437: 269–277.

Reese, E.S. 1973. Duration of residence by coral reef fishes on "home" reefs. Copeia 1973: 145-149.

Reese, E.S. 1981. Predation on corals by fishes of the family Chaetodontidae: Implications for conservation and management of coral reef ecosystems. Bulletin of Marine Science 31: 594-604.

Rivera, M.A.J., C.D. Kelley, and G.K. Roderick. 2004. Subtle population genetic structure in the Hawaiian grouper, Epinephelus quernus (Serranidae) as revealed by mitochondrial DNA analyses. Biological Journal of the Linnean Society 81: 449–468.

Roberts, C.M., A.R.D. Shepherd, and R.F.G. Ormond. 1992. Large scale variation in assemblage structure of Red Sea butterflyfishes and angelfishes. Journal of Biogeography 19: 239-250.

Rocha, L.A., A.L. Bass, R. Robertson, and B.W. Bowen. 2002. Adult habitat preferences, larval dispersal, and the comparative phylogeography of three Atlantic surgeonfishes (Teleostei: Acanthuridae). Molecular Ecology 11: 243–252.

Rocha, L.A., M.T. Craig, and B.W. Bowen. 2007. Phylogeography and the conservation genetics of coral reef fishes. Coral Reefs 26: 501-512.

Samways, M.J. 2005. Breakdown of butterflyfish (Chaetodontidae) territories associated with the onset of a mass coral bleaching event. Aquatic Conservation: Marine and Freshwater Ecosystems 15: S101–S107.

Schultz, J.K., Feldheim, K.A., Gruber, S.H., Ashley, M.V., McGovern, T.M., and Bowen, B.W. (2008) Global phylogeography and seascape genetics of the lemon sharks (genus Negaprion). Molecular Ecology 17: 5336–5348.

Selkoe, K.A., and R.J. Toonen 2006. Microsatellites for ecologists: a practical guide to using and evaluating microsatellite markers. Ecology Letters 9: 615-629.

Selkoe, K.A., and R.J. Toonen 2011. Marine connectivity: a new look at pelagic larval duration and genetic metrics of dispersal. Marine Ecology Progress Series 436: 291-305.

Seutin, G., B.N. White, and P.T. Boag. 1991. Preservation of avian blood and tissue samples for DNA analyses. Canadian Journal of Zoology 69: 82-90.

Sheppard, C.R.C. 1998. Biodiversity patterns in Indian Ocean corals, and effects of taxonomic error in data. Biodiversity and Conservation 7: 847-868.

Sheppard, C., M. Al-Husiani, F. Al-Jamali, F. Al-Yamani, R. Baldwin, J. Bishop, F. Benzoni, E. Dutrieux, N.K. Dulvy, S.R.V. Durvasula, D.A. Jones, R. Loughland, D. Medio, M. Nithyanandan, G.M. Pilling, I. Polikarpov, A.R.G. Price, S. Purkis, B. Riegl, Saburova, M., K.S. Namin, O. Taylor, S. Wilson, and K. Zainal. 2010. The Gulf: A young sea in decline. Marine Pollution Bulletin 60: 13–38.

Sheppard, C., M. Ateweberhan, B.W. Bowen, P. Carr, C.A. Chen, C. Clubbe, M. T. Craig, R. Ebinghaus, J. Eble, N. FitzSimmons, M.R. Gaither, C.-H. Gan, M. Gollock, N. Guzman, N.A.J. Graham, A. Harris, R. Jones, S. Keshavmurthy, H. Koldewey, C.G. Lundin, J.A. Mortimer, D. Obura, M. Pfeiffer, A.R.G. Price, S. Purkis, P. Raines, J.W. Readman, B. Riegl, A. Rogers, M. Schleyer, M.R.D. Seaward, A.L.S. Sheppard, J.

Tamelander, J.R. Turner, S. Visram, C. Vogler, S. Vogt, H. Wolschke, J. M.-C. Yang, S.- Y. Yang, and C. Yesson. 2012. Reefs and islands of the Chagos Archipelago, Indian Ocean: Why it is the world’s largest no-take marine protected area. Aquatic Conservation: Marine and Freshwater Ecosystems 22: 232 – 261.

Shokri, M.R., S.M.R. Fatemi, and M.P. Crosby. 2005. The status of butterflyfishes (Chaetodontidae) in the northern Persian Gulf, I.R. Iran. Aquatic Conservation: Marine and Freshwater Ecosystems 15: S91–S99.

Skillings, D.J., C.E. Bird, and R.J. Toonen. 2011. Gateways to Hawai‘i: Genetic population structure of the tropical sea cucumber Holothuria atra. Journal of Marine Biology, Article ID 783030.

Smith, W.L., J.F. Webb, and S.D. Blum. 2003. The evolution of the laterophysic connection with a revised phylogeny and taxonomy of butterflyfishes (Teleostei: Chaetodontidae). Cladistics 19: 287–306.

Song, C.B., T. J. Near, and L.M. Page. 1998. Phylogenetic relations among percid fishes as inferred from mitochondrial cytochrome b DNA sequence data. Molecular Phylogenetics and Evolution 10: 343–353.

Swearer, S.E., J.S. Shima, M.E. Hellberg, S.R. Thorrold, G.P. Jones, D.R. Robertson, S.G. Morgan, K.A. Selkoe, G.M. Ruiz, and R.R. Warner. 2002. Evidence of self- recruitment in demersal marine populations. Bulletin of Marine Science 70: 251-271.

Swofford, D.L. 2003. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4. Sinauer Associates, Sunderland, Massachusetts.

Taberlet, P., A. Meyer, and J. Bouvert. 1992. Unusually large mitochondrial variation in populations of the blue tit, Parus caeruleus. Molecular Ecology 1: 27–36.

Tamura, K., and M. Nei. 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Molecular Biology and Evolution 10: 512–526.

Tamura, K., D. Peterson, N. Peterson, G. Stecher, M. Nei, and S. Kumar. 2011. MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biology and Evolution 28: 2731- 2739.

Timmers, M.A., K.R. Andrews, C.E. Bird, M.J. deMaintenton, R.E. Brainard, and R.J. Toonen. 2011. Widespread dispersal of the crown-of-thorns sea star, Acanthaster planci, across the Hawaiian Archipelago and Johnston Atoll. Journal of Marine Biology, Article ID 934269.

Toonen, R.J., K.R. Andrews, I.B. Baums, C.E. Bird, G.T. Concepcion, T.S. Daly-Engel, J.A. Eble, A. Faucci, M.R. Gaither, M. Iacchei, J.B. Puritz, J.K. Schultz, D.J. Skillings, M.A. Timmers, and B.W. Bowen. 2011. Defining boundaries for ecosystem-based management: A multispecies case study of marine connectivity across the Hawaiian Archipelago. Journal of Marine Biology. Article ID 460173.

Toonen, R.J., T.A. Wilhelm, S.M. Maxwell, D. Wagner, B.W. Bowen, C.R.C. Sheppard, S.M. Taei, T. Teroroko, R. Moffitt, C.F. Gaymer, L. Morgan, N. Lewis, A.L.S. Sheppard, J. Parks, A.M. Friedlander, and The Big Ocean Think Tank. 2013. One size does not fit all: The emerging frontier in large-scale marine conservation. Marine Pollution Bulletin 77: 7–10.

Van Oosterhout, C., W.F. Hutchinson, D.P.M. Willis, and P. Shipley. 2004. MICRO- CHECKER: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4: 535–538.

Vijay Anand, P.E. and N.G.K. Pillai. 2002. Reproductive biology of some common coral reef fishes of the Indian EEZ. Journal of the Marine Biological Association of India 44: 122 – 135.

Vogler, C., J.A.H. Benzie, K. Tenggardjaja, Ambariyanto, P.H. Barber, and G. Wörheide. 2012. Phylogeography of the crown-of-thorns starfish in the Indian Ocean. Coral Reefs 32: 515–525.

Williams, D.F., J. Peck, E.B. Karabanov, A.A. Prokopenko, V. Kravchinsky, J. King, and M.I. Kuzmin. 1997. Lake Baikal record of continental climate response to orbital insolation during the past five million years. Science 278: 1114–1117.

Winterbottom, R. and R.C. Anderson. 1997. A revised checklist of the epipelagic and shore fishes of the Chagos Archipelago, Central Indian Ocean. Ichthyological Bulletin, J.L.B. Smith Institute of Ichthyology 66: 1–2.

Zekeria, Z.A., Y. Afeworki, and J.J. Videler. 2005. The distribution patterns of Red Sea Chaetodontid assemblages. Aquatic Conservation: Marine and Freshwater Ecosystems 15: S71–S76.