Phylogeography and molecular systematics of the rafting aeolid pinnata (Eschscholtz, 1831)

Jennifer S. Trickey

A thesis submitted for the degree of

Master of Science in Zoology

at the University of Otago,

New Zealand

August 2012

An undescribed of Fiona nudibranch (at center) on the mooring line of a rompong in SE Sulawesi, Indonesia. Also pictured are its egg masses and prey.

© Magnus Johnson [University of Hull]

i

ABSTRACT

The pelagic nudibranch (: ) occurs exclusively on macroalgal rafts and other floating substrata, and is found throughout tropical and temperate seas worldwide. Its cosmopolitan distribution has been attributed to its planktotrophic larval mode and propensity for passive rafting, and although it was one of the earliest aeolid nudibranchs to be described, this study produced the first molecular phylogeny for this ubiquitous . Mitochondrial and nuclear DNA sequence data was generated from specimens collected worldwide in order to elucidate the genetic structure and diversity within this obligate rafter. Phylogeographic analyses revealed three distinct lineages that were geographically partitioned in concordance with oceanic circulation patterns. Two clades were abundant and widespread, with one displaying a circum-equatorial distribution and the other exhibiting an anti-tropical distribution throughout temperate zones of the Pacific Ocean. A third lineage based on a single Indonesian specimen was also detected, and the genetic divergences and largely allopatric distributions observed among these three clades suggest that they may represent a cryptic species complex. Long-distance dispersal in this nudibranch appears to be current-mediated, and the North-South disjunction detected within New Zealand is concordant with known marine biogeographic breaks. In contrast, populations sampled in Chile and the South Island of New Zealand displayed close phylogenetic relationships, indicating that the West Wind Drift has facilitated trans-oceanic gene flow in Fiona pinnata. All sampled individuals formed a well-supported monophyletic group that was recovered in phylogenetic analyses of several independent molecular markers. Although much ambiguity has surrounded Fiona pinnata’s taxonomic status since its original discovery in 1775, the molecular evidence of the current study confirms that this group is substantially divergent from even its closest relatives in the and thus upholds the systematic standing of the historically monotypic family .

ii

ACKNOWLEDGMENTS

Many people contributed to the production of this thesis. First and foremost I must give thanks to the University of Otago for granting me an International Master’s Scholarship that made this project possible. I am also extremely grateful to my adviser, Prof. Jonathan Waters, who has been supportive every step of the way during this program; his feedback has been greatly appreciated throughout the entirety of this endeavor. Dr. Tania King has also been instrumental in my success; her guidance and patience in the lab were critical in getting this project up and running, and her troubleshooting skills were much appreciated. I am grateful to have passed through Otago’s Zoology Department and my fellow researchers in the rafting group have all contributed to my project in one way or another. I am deeply indebted to Chris Garden for always being willing to give a helping hand, whether with sample collection, data analysis, or GIS work. Laura Bussolini was also very helpful with sample collection, and I appreciate her braving seasickness in order to help me hunt for sea slugs. Raisa Nikula and Rebecca Cumming were both helpful in introducing me to the daunting world of data analysis and were always willing to lend an ear. Kim Currie, Phil Heseltine, and Bill Dickson were invaluable in their assistance in sampling, and I’m extremely appreciative that I was allowed to tag along on the RV Polaris cruises. They made every outing as smooth as could be expected, and it was always refreshing to get out in the field and see these research questions in action. All DNA sequencing was carried out at the University of Otago Genetic Analysis Services; their efficiency and cooperation made my work that much easier. Achieving global sampling of an elusive critter in under a year was a daunting task, but it was made a reality through the tremendous generosity of several people around the world. Martin Thiel was an outstanding collaborator who collected and contributed valuable specimens and went out of his way to help me in my quest for samples. Several people from all around the world went above and beyond by monitoring, collecting, organizing, and delivering samples: Richard Taylor of the University of Auckland, Yira Tibirica of Zavora Marine Lab, Yoichi Yusa of Nara Women’s University, and Marcus Eriksen of the 5 Gyres Institute. Many people from iii museums also generously provided loans of preserved specimens: Adam Baldinger of the Harvard Museum of Comparative Zoology, Kathe Jensen of the Zoological Museum of the University of Copenhagen (the original 1856 Bergh specimens no less!), Rafa Araujo of the Museo Nacional de Ciencias Naturales, Elizabeth Kools and Terry Gosliner of the California Academy of Sciences, Wilma Blom of the Auckland War Memorial Museum, Mandy Bemis and Terry Lott of the Florida Museum of Natural History, Christine Zorn of the Museum fur Naturkunde, Ricardo Araujo of the Museu de Historia Naturales do Funchal, Janet Waterhouse and Mandy Reid of the Australian Museum, and lastly Michael Schrödl and Enrico Schwabe of the Zoologische Staatssammlung Munchen. Cory Pittman was a great help, both in collecting samples and sharing his comprehensive knowledge of Hawaii’s rafting nudibranchs. Nerida Wilson greatly encouraged my early interest in nudibranchs and has been a great mentor and co- author over the past few years; she also contributed what proved to be some of the most interesting samples in this study. Bill Rudman kindly granted me the use of some of his pictures from the Forum, a truly fantastic resource. Graeme Loh from the Department of Conservation also provided some outstanding photographs taken during long days of fieldwork at sea. My family has of course been influential in my academic endeavors and I appreciate their constant support for my scientific pursuits, even when they take me thousands of miles from home. Lastly, I must thank my partner, Paul Nakauchi, for his support and patience during this project, and for enduring my many nudibranch-based ramblings over the past year. This work was supported by a Marsden fund courtesy of the Royal Society of New Zealand (contract 07-UOO-099 to J. M. Waters).

iv

TABLE OF CONTENTS

Abstract…………………………………...... i

Acknowledgments………………………………………………………...... ii

List of Tables……………………………….…………………………………...... vii

List of Figures………………………………….……………………………...... viii

List of Abbreviations….……………………………………………………...... ix

Chapter 1: General introduction……………………………………………………1

1.1 Cosmopolitan marine species and cryptic speciation…………………….…...... 1

1.2 The biogeographic importance of oceanic rafting…………………...... ………..3

1.3 The natural history of the aeolid nudibranch Fiona pinnata...……………...……..6

1.4 The taxonomic uncertainty of Fiona pinnata...... 9

1.5 Thesis objectives...... 11

Chapter 2: Genetic structure and diversity of the pelagic nudibranch Fiona pinnata within New Zealand & Chile and the influence of the West Wind Drift on population connectivity……………………………...... 13

2.1 Abstract………………………………………………………………………...... 13

2.2 Introduction………………………………...………………………………..…...14

2.3 Methods………….……………………………...………………………………..16

2.3.1 Sample collection…………...………………………………………….…….16

2.3.2 DNA extraction, PCR amplification, and sequencing……………...………...21

2.3.3 Analysis of mitochondrial DNA variation……………...….…....….………..24

2.3.4 Analysis of nuclear DNA variation……...…………………………………...25

2.4 Results………………………………………………………...... …………26 v

2.4.1 Mitochondrial COI variation………………………………………………....26

2.4.2 Nuclear ITS variation……………………...……………………………...... 31

2.5 Discussion………...…………………………………………………..…………33

2.5.1 Genetic structure of Dunedin populations……………………………..……34

2.5.2 North/South New Zealand dynamics…………………...……...……………35

2.5.3 Genetic structure of Coquimbo populations...………..……………………..36

2.5.4 Southern Hemisphere patterns……………………………………..………..36

Chapter 3: Global phylogeography, evolutionary history, and systematic status of the aeolid nudibranch Fiona pinnata…...... 38

3.1 Abstract...... ………..………………………...………………………………..38

3.2 Introduction...... 39

3.3 Methods…………………...…………………...………………………...... ……41

3.3.1 Sample collection………………….……………...……………….....…...... 41

3.3.2 DNA extraction, PCR amplification, and sequencing…...…………..…...... 41

3.3.3 Analysis of global mitochondrial DNA variation……………………………44

3.3.4 Analysis of global nuclear DNA variation……….……………………...…...45

3.3.5 Phylogenetic analysis of the family Fionidae………………………………..45

3.4 Results……………………..………………………………………...………...…48

3.4.1 Global mitochondrial DNA variation………..……………………………….48

3.4.2 Global nuclear DNA variation……………………………………………….54

3.4.3 Phyleogenetic analysis of the family Fionidae………………………………58

3.5 Discussion………………………………………..………………………….…...62

3.5.1 Global phylogeography and evolutionary history …………………………..62

3.5.2 Systematic status of the family Fionidae…………………………………….69

Chapter 4: Summary and future research………………………………………...70 vi

4.1 Overall conclusions………………………….…………………………...…...70

4.2 Future work and recommendations …………………………………………..72

References………………………….....…..………………………………………….76

Appendix 1: A taxonomic overview of Fiona pinnata...... 93

Appendix 2: Mitochondrial cytochrome b sequence data...... 94

Appendix 3: COI haplotype frequencies in New Zealand………………………...…95

Appendix 4: COI haplotype frequencies in Coquimbo, Chile……………………….96

vii

LIST OF TABLES

Table 2.1 Collection details for New Zealand samples………………………………19

Table 2.2 Collection details for Chilean samples……………………...……………..20

Table 2.3 Primers used in PCR amplification………………………………………..21

Table 2.4 COI sequence divergence values for NZ & Chilean specimens…..………27

Table 2.5 COI AMOVA values for NZ & Chilean regional comparisons…………...29

Table 3.1 Collection details for global samples …………….……………………….47

Table 3.2 Primers used in PCR amplification……………………………..…………48

Table 3.3 Global COI sequence divergence values……………...... 51

Table 3.4 AMOVA values of global COI diversity among ocean basins…………....54

viii

LIST OF FIGURES

Figure 1.1 A floating bull kelp raft (Durvillaea antarctica) off the coast of NZ……...5

Figure 1.2 A kelp holdfast colonized by & Fiona nudibranchs…..…8

Figure 1.3 Australian specimens of Fiona pinnata & close-up of …...……….10

Figure 2.1 Map of New Zealand sampling sites……………………………………..18

Figure 2.2 Map of Coquimbo, Chile sampling sites…………………………………20

Figure 2.3 Maximum likelihood tree based on COI sequence data of New Zealand & Chilean specimens……………………………………………………………………28

Figure 2.4 Geographic distributions of COI clades within the South Pacific………..29

Figure 2.5 COI haplotype network of New Zealand & Chilean specimens……….....30

Figure 2.6 Maximum likelihood tree based on ITS1 sequence data of New Zealand & Chilean specimens.…...….………...…………………………………………………32

Figure 2.7 ITS1 haplotype network of New Zealand & Chilean specimens……...….32

Figure 2.8 ITS2 haplotype network of New Zealand & Chilean specimens…...…….33

Figure 3.1 Global map of sampling sites & geographic distribution of COI clades....49

Figure 3.2 Maximum likelihood tree based on global COI sequence data…...……...52

Figure 3.3 Global COI haplotype networks………...... 53

Figure 3.4 Maximum likelihood tree based on global ITS1 sequence data...………..56

Figure 3.5 Global ITS1 haplotype network……………………………………..……57

Figure 3.6 Global ITS2 haplotype network……………………………………...…...58

Figure 3.7 Deep-rooted maximum likelihood tree based on COI sequence data…….59

Figure 3.8 Deep-rooted maximum likelihood tree based on 16S sequence data…….61

Figure 3.9 Deep-rooted neighbor joining tree based on H3 sequence data………...... 61

Figure 3.10 Deep-rooted neighbour joining tree based on 18S sequence data………61

Figure 3.11 Simplified of Fiona pinnata…………………………………69 ix

LIST OF ABBREVIATIONS

16S mitochondrial 16S ribosomal RNA

18S nuclear 18S ribosomal RNA

AMOVA Analysis of molecular variance bp base pair

C Celsius

COI Cytochrome c oxidase subunit I

DNA deoxyribonucleic acid

Fst fixation index

H3 nuclear histone protein H3

ITS1 nuclear RNA internal transcribed spacer one

ITS2 nuclear RNA internal transcribed spacer two

K2P Kimura (1980) 2-parameter model of sequence evolution

µl microliter

µM micromolar mm millimeter min minute

NZ New Zealand

PCR Polymerase chain reaction

RNA ribonucleic acid rRNA ribosomal RNA s second spp. species

U unit 1

CHAPTER 1 General introduction

This chapter first introduces the concept of cosmopolitan marine species and describes what is currently known about the phylogenetics of some widespread nudibranchs. Second, I call attention to passive rafting as an important long-distance dispersal vector in the marine environment, particularly for that are widely distributed but seemingly lack dispersal ability. Third, I give an overview of nudibranch ecology and life history and introduce the aeolid nudibranch Fiona pinnata. Finally, I discuss the long-standing uncertainty surrounding the taxonomic status of Fiona pinnata. This introductory chapter ends by briefly outlining the structure and specific objectives of this thesis.

1.1 Cosmopolitan marine species and cryptic speciation Historically, biologists have assumed that speciation is less common in the oceans than in the terrestrial realm due to a lack of conventional, highly perceptible barriers to gene flow in the marine environment (e.g., Palumbi 1992; Lessios 1996; Benzie 1999; Bierne et al. 2003b). Marine invertebrates with long-lived planktonic larvae are often presumed to have high dispersal ability and levels of gene flow, and are thus predicted to display broad geographic ranges (Shuto 1974; Valentine and Jablonski 1983; Hedgecock 1986). Furthermore, cosmopolitan distributions have regularly been accepted for many marine organisms that seemingly lack any distinguishable morphological characteristics throughout their ranges (Knowlton 1993; Dawson and Jacobs 2001; Cox and Moore 2005). However, it is now understood that marine biodiversity is underestimated while the widespread distributions attributed to many taxa are overestimated, often as a result of undetected cryptic speciation. Although biologists have long relied on morpho-anatomical characters to make taxonomic decisions, it is clear that morphology alone is not an absolute reflection of phylogenetic signal, especially when a high degree of plasticity is associated with morphological characters used in species identification. With the rising prevalence of molecular techniques, scientists are now better equipped to elucidate phylogenetic relationships and thereby further our understanding of evolutionary processes in the 2 sea. Molecular phylogenies are rapidly entering the mainstream and are often used in conjunction with morphological examinations in order to verify taxonomic classifications (Knowlton 2000; Blanquer and Uriz 2007). As a result of the recent advance of molecular tools, many marine species previously deemed cosmopolitan or assumed to have high dispersal potential have instead been shown to comprise cryptic species complexes (e.g., Knowlton 1993, 2000; Palumbi et al. 2007; Avise 2000; Bierne et al. 2003; de Vargas et al. 2003; Miglietta et al. 2011). Additionally, while many of these sibling species previously appeared morphologically homogeneous, ensuing comprehensive examination often results in the discovery of subtle diagnostic characters (Knowlton 1993), although this is not always the case (Sponer 2002). Examples of cryptic speciation in the sea have been discovered across a diverse range of organisms including bryozoans (Nikulina et al. 2007; Schwaninger 2008), annelids (Bleidorn et al. 2006), nematodes (Zeppilli et al. 2011), fish (Colborn et al. 2001), mussels (Ladoukakis et al. 2002; Bierne et al. 2003a), copepods (Lee 2000; Goetze 2003), sponges (Klatau et al. 1999; Lazoski et al. 2001; Blanquer and Uriz 2007), echinoderms (Le Gac et al.; Williams 2000; Waters and Roy 2003; Zulliger and Lessios 2010), barnacles (Chan et al. 2009), and molluscs (Reid et al. 1996; Benzie and William 1997; Quattro et al. 2001). In contrast, the giant kelp Macrocystis pyrifera is an example of a truly cosmopolitan species, and its broad distribution is likely a result of high gene flow maintained via the long-distance dispersal of detached, reproductively viable sporophytes (Horta 2010). In light of many of these studies, truly cosmopolitan marine species are presumably quite rare, as the mounting genetic evidence suggests that the broad distributions attributed to many “ubiquitous” taxa are simply artifacts of inadequate taxonomy. This would suggest that marine invertebrates that are labeled cosmopolitan are in reality either masking a number of cryptic lineages, or are capable of some form of long-distance transport that surpasses their otherwise limited dispersal ability based strictly on autonomous mobility and larval period alone. Widespread distributions are commonly attributed to many species of nudibranchs (e.g., Edmunds 1977; Malaquias et al. 2009), although molecular data to verify these assumptions is generally unavailable due to a paucity of species-level phylogenies within opisthobranchs (a paraphyletic group nested within that 3 contains nudibranchs, see Schrödl et al. 2011). For example, approximately 12% of all opisthobranchs found in the Atlantic Ocean are considered amphi-Atlantic (Garcia-Talavera 1983; Valdes et al. 2006; Garcia and Bertsch 2009), but these distributions are based solely on morpho-anatomical similarities. Recent studies employing molecular tools (e.g., Malaquias and Reid 2008; Carmona et al. 2011) have revealed a number of cases of cryptic speciation within these Atlantic taxa, and phylogenetics-based research has also identified previously unrecognized sibling species of opisthobranchs on more localized scales as well (e.g., Morrow et al. 1992; Faucci et al. 2007; Krug et al. 2008; Wilson et al. 2009; Krug 2011; Yorifuji et al. 2012). There are two groups of pelagic aeolid nudibranchs, both of which display broad distributions throughout all oceans. The first group consists of two species within the family Glaucidae, Glaucilla marginata and atlanticus, which are drifters that prey upon floating cnidarians (Valdes and Campillo 2004). Although they are usually found floating upside down on the surface tension of the ocean, they have also been observed rafting on macroalgae (Fine 1970; Coston-Clements et al. 1991). Valdes and Campillo (2004) were unable to resolve the systematic status of the Glaucidae based on anatomical examination alone, and although molecular data is forthcoming (Celia Churchill, pers. comm.), the phylogeny of this group remained unknown at the time of this study. The second widespread pelagic nudibranch taxa is the Fionidae, a monotypic family represented only by Fiona pinnata, which generally occurs on floating objects in association with its Lepas spp. prey (Miller 1974; Willan 1979). Although Fiona pinnata has been the subject of several anatomical examinations (e.g., Alder and Hancock 1855; Casteel 1904; Suter 1913; Pruvot-Fol 1954; MacFarland 1966; Williams 1978), a molecular phylogeny of this widespread nudibranch did not exist before the present study.

1.2 The biogeographic importance of oceanic rafting Dispersal of macroalgae and benthic invertebrates in the marine environment is thought to occur primarily via propagules or planktonic larvae, and the role of larval dispersal in genetic connectivity has received much ongoing attention (Reed et al. 2002; Palumbi 2003; Grantham et al. 2003; Shanks et al. 2003; Gerber et al. 2005; Levin 2006). On the other hand, species that lack a planktonic larval stage, such as 4 brooders, direct developers, or clonal species, are typically considered to have diminished dispersal potential (Strathmann 1985; Scheltema 1986). However, molecular studies have uncovered high genetic similarity among disjunct populations of benthic marine invertebrates, which indicates that organisms lacking larval dispersal ability can nonetheless have puzzlingly vast geographic distributions (Sole- Cava et al. 1994; Grant and da Silva-Tatley 1997; Hurr et al. 1999; O’Foighil and Jozefowicz 1999; Oosthuizen et al. 2004). Sedentary life history traits such as brooding strongly contrast with the wide distributions attributed to many marine invertebrates, and thus passive rafting has often been invoked to explain this paradox (Johannesson 1988; Parker and Tunnicliffe 1994; O’Foighil et al. 1999). Although long-distance rafting is an old idea in marine biogeography (Darwin 1872), evidence supporting the role of this ecological process in shaping the dispersal and colonization patterns of certain macroalgae and associated epibiont communities has historically been inferred from circumstantial data alone (Moore et al. 1952; Highsmith 1985; Edgar 1987). In contrast, empirical data validating the frequency and efficiency of rafting as a long-distance dispersal mechanism has only emerged within the last decade (Waters and Roy 2004a; de Queiroz 2005; Donald et al. 2005; Gordillo 2006). Research on passive rafting typically focuses on buoyant macroalgal substrata, which are prevalent near benthic source populations in temperate zones (Rothausler 2012). Detachment and subsequent drifting of certain seaweeds is indeed a very common process in the higher latitudes: Smith (2002) estimated that seventy million rafts of the bull kelp Durvillaea antarctica are afloat in the Southern Ocean at any given time. The dominant rafting kelps in cold temperate waters are two species of brown algae: the honeyblade structure of the bull kelp Durvillaea antarctica (Fig. 1.1) allows it to float for extended periods of time and over great distances (Fraser et al. 2011), while the giant kelp Macrocystis pyrifera persists at the sea surface due to the presence of pneumatocysts (gas-filled tissues) and can also maintain reproductive viability post-detachment (Macaya et al. 2005; Hernandez-Carmona et al. 2006). These robust seaweeds can traverse vast stretches of open ocean and in some instances have been estimated to cover distances of over 1000 km (Harrold and Lisin 1989; Hobday 2000a; Fraser 2011). Floating seaweeds are frequently colonized by sessile epibionts, and thus these kelp-associated fauna can subsequently capitalize on the high dispersal ability of their macroalgal hosts. A diverse spectrum of marine 5

Figure 1.1 Rafting Durvillaea antarctica off the coast of Dunedin, New Zealand. [Photo © Graeme Loh]

invertebrates have indeed been recorded on floating substrata (for a review see Thiel and Gutow 2005b), and molecular studies have confirmed that population connectivity in buoyant seaweeds and benthic marine invertebrates can be greatly enhanced via rafting (Nikula et al. 2010; Fraser et al. 2011). Although the tropics represent a strong dispersal barrier to macroalgal transport, epifaunal organisms can still partake in passive rafting in lower latitudes due to the presence of other floating substrata such as driftwood (Nilsson-Cantell 1930; Edmands and Potts 1997), plastic debris (Carpenter and Smith 1972; Aliani and Molcard 2003), and volcanic pumice (Coombs and Landis 1966; Jokiel 1984). Passive rafting is occasionally invoked to explain the broad distributions and colonization patterns of some nudibranchs (e.g., Jensen 2005; Carmona et al. 2011; Martynov and Schrödl 2011), but it is often relegated to speculation because robust genetic evidence demonstrating high gene flow in a nudibranch as a result of long- distance dispersal has yet to surface in the literature. However, direct observations of rafting on the high seas exist for over forty known species of opisthobranchs (for a 6 review see Thiel and Gutow 2005b); these marine gastropods are most frequently encountered on floating macroalgae (e.g., Edgar 1987; Bushing 1994; Hobday 2000a) but also occur on other substrates such as plastic debris, driftwood, or pumice. Within nudibranchs in particular, rafting has been reported for a number of species including leonina (Bushing 1994; Hobday 2000a), Triopha maculata (Bushing, 1994; Hobday 2000a), crassicornis (Bushing 1994), Corambe pacifica (Bushing 1994; Martynov and Schrödl 2011), sp. (Aliani and Molcard 2003), sp. (Cory Pittman, pers. comm.), and valentini (Schrödl 2003). Rafting nudibranchs generally co-occur with their sessile epibiont prey (e.g., bryozoans, hydroids) and are often observed laying eggs on their raft substrate (Bushing 1994). Arguably the most frequently encountered rafting nudibranch is Fiona pinnata, which is considered to have a cosmopolitan distribution.

1.3 The natural history of the aeolid nudibranch Fiona pinnata Nudibranchs, commonly referred to as sea slugs, are a diverse group of shell-less gastropod molluscs that inhabit all marine environments. They are simultaneous and their mating strategy consists of reciprocal, internal fertilization (Costello 1938; Beeman 1977). Following a mutual exchange of sperm, both deposit spirally-coiled egg masses onto the available substratum; these egg masses consist of fertilized embryos set in a gelatinous matrix (Wilson 2002) and their shape is often unique to species (Behrens 2005). Nudibranchs generally feed upon sedentary animals such as sponges, hydroids, cnidarians, or tunicates, and many exhibit remarkable prey specificity in correlation with various adaptations in radular morphology (Behrens 1980; Rudman and Bergquist 2007; Gosliner et al. 2008). The term ‘nudibranch’, meaning “naked gills”, refers to the exposed gills that dorid nudibranchs bear on their dorsal surfaces. In contrast, aeolid nudibranchs are a group broadly characterized by the possession of dorsal structures called cerata that are branching extensions of the digestive gland and function in respiration (Gosliner et al. 2008). Several cnidarian-feeding aeolids are able to sequester nematocysts from their colenterate prey and subsequently store them in at the ceratal tips for their own defense (Greenwood and Mariscal 1984). Fiona pinnata (Eschscholtz, 1831) is an aeolid nudibranch and obligate rafter (Thiel & Gutow 2005b) that has been found on a wide variety of floating substrata 7 including macroalgae (Casteel 1904; Adams 1960; Bushing 1994), driftwood (Inatsuchi et al. 2010), plastic debris (Aliani and Molcard 2003; Scarabino 2004), buoys (Williams 1978), cuttlebones (N. G. Wilson, pers. comm.; Rudman 2002), Janthina snail shells (Bergh 1882; Willan 1979), and even loggerhead sea turtles (Loza and Lopez-Jurado 2008). This aeolid is well adapted to a rafting lifestyle and accordingly displays some peculiar features such as a broad, mobile foot, short cerata arranged along each side of the body, and an anus that opens dorsally (Alder and Hancock 1855; Willan 1979). Taken together, these adaptations make for a low profile that is suitable for enduring conditions in the pleuston, and when found on macroalgal rafts this typically inhabits the less-exposed nooks and crannies within the holdfasts where it is less likely to fall off (pers. obs.). Furthermore, Fiona pinnata is not truly pelagic and cannot swim or float by itself (Rudman 2002), hence why it is found exclusively on floating objects. Accordingly, adults are only found in association with a food resource, because unlike other nudibranchs Fiona pinnata cannot drift or crawl around in search of prey. Once have settled on an object in the pleuston, their foraging opportunites are then restricted to the substrate on which they have metamorphosed. Like all nudibranchs, Fiona pinnata is carnivorous and has been reported to feed upon a variety of similarly pelagic animals, although it is most commonly found preying upon pelagic barnacles of the Lepas (Fig. 1.2). This group of goose barnacles exclusively colonizes floating objects (Skerman 1958; Hinojosa et al. 2006; Thiel and Gutow 2005b), and thus can be found on stationary objects (e.g., buoys, fishing floats), rafting substrata (macroalgae, driftwood, plastic debris), and as epibionts on other animals (whales, loggerhead turtles) (see Thiel and Gutow 2005b for a review). As reviewed in Willan (1979), Fiona pinnata has been reported feeding on (Bennett 1966; Holleman 1972; Williams 1978; McDonald and Nybakken 1978), (Bieri 1966), Lepas fascicularis (MacGintie and MacGintie 1949), Lepas hilli (MacFarland 1966), and Lepas testudinata (Foster 1978). Fiona pinnata was also observed eating spp. barnacles (McDonald and Nybakken 1978; Beeman and Williams 1980), and was capable of consuming the non-pelagic barnacles and , but could only attack them if they were damaged (Holleman 1972; Willan, 1979). As is observed with the Glaucidae nudibranchs (Miller 1974), Fiona pinnata 8

Figure 1.2 A Durvillaea antarctica holdfast collected off the coast of Dunedin, New Zealand and colonized by Lepas sp. goose barnacles and Fiona pinnata nudibranchs (circled in yellow). [Photo © Graeme Loh]

has also been documented living on and predating the pleustonic cnidarians velella, , and Physalia physalis (Pruvot-Fol, 1954; Bayer, 1963; Burn, 1966, 1967; MacFarland, 1966; McDonald and Nybakken 1978), but unlike the majority of aeolid nudibranchs Fiona pinnata does not possess cnidosacs and thus cannot hijack and store nematocysts from these cnidarian prey to use for its own defense. However, this seemingly opportunistic predator can concentrate pigments of its ingested prey, as this nudibranch is ordinarily cream to brown in colour while feeding upon barnacles, but becomes blue-purple after consuming the cnidarian Velella velella (Kropp 1931; Bayer 1963; Bennett 1966; Behrens 1991). Nudibranchs in general display ephemeral life histories, typically living less than a year (Gosliner et al. 2008), but Fiona pinnata nonetheless stands out as an example of a particularly accelerated life cycle. Although Schaefer (1996) gave 15-42 days as the average larval duration for planktonic opisthobranch larvae, Fiona pinnata’s free- swimming larvae take five days to hatch under laboratory conditions 9

(Holleman 1972), and following metamorphosis it displays the second highest growth rate known for nudibranchs (Willan 1979). It grows to an average length of 20 mm (Suter 1913), and reaches maturity very quickly. Bayer (1963) was one of the first authors to note this extremely rapid growth rate, as he documented the development of juveniles measuring 8 mm long into sexually mature adults about 20 mm in length after only four days in captivity. Fiona pinnata often occurs in dense aggregations and will continue feeding until its sedentary food source is completely depleted (Behrens 1992; Rudman 2002), upon which it begins spawning prolifically. When the eggs hatch, they disperse into the currents and this cycle begins anew. The challenge of coming across appropriate habitat (i.e., substrate with suitable prey) in the open ocean led Willan (1979) to postulate that Fiona pinnata veligers may be able to delay metamorphosis until triggered by an appropriate prey’s chemical cue. Some opisthobranchs are known to have the capacity to delay metamorphosis (Thompson 1958, 1962; Hadfield and Karlson 1969) so there is some precedent for this idea, but this hypothesis has never been explicitly tested in Fiona pinnata and thus its larval duration in situ remains unknown.

1.4 The taxonomic uncertainty of Fiona pinnata The pelagic nudibranch Fiona pinnata (Eschscholtz, 1831) is currently considered a single species displaying circum-tropical and circum-temperate distribution in all oceans. Although it was one of the first aeolid nudibranchs to be identified and recognized (Willan 1979), the taxonomic classification of Fiona pinnata has been contentious ever since its original discovery. The first recorded finding of this organism was in 1775 by the Swedish naturalist Peter Forsskal. However, the name he assigned this nudibranch, marinus, was already in use by Gunnerus (Pruvot- Fol 1954) and thus did not give rise to the currently accepted nomenclature. For well over a hundred years following Forsskal’s early finding, this ubiquitous nudibranch was repeatedly “discovered” and described by biologists working in diverse geographic regions the world over (see Appendix 1 for an overview of its taxonomic history). It has been assigned over ten different scientific names since its original discovery, but the specimen is based on samples found by F. Eschscholtz in Sitka, Alaska, which he called Eolidia pinnata at the time (later synonymized into Fiona pinnata). Eventually all of these designations were resolved into a single 10

Figure 1.3 Fiona pinnata specimens collected from driftwood in Eden, Australia. On left: the subdorsal position of the anus is visible midway down the animal, and the membranous frilly edge along the cerata is shown in the inset. On right: several individuals on a Lepas sp. goose barnacle. [Photos © Bill Rudman]

cosmopolitan species, as a review of these various taxonomic accounts revealed that the morphological descriptions and reported natural histories were in agreement. It thus appears likely that far-flung populations were originally regarded as separate species because each subsequent author failed to surmise that such a sedentary animal could occur in all major oceans over a broad latitudinal range. Although Fiona pinnata was eventually deemed a single species, most authors have placed it within a monotypic family, the Fionidae (Gray, 1857), a name meaning “feathered”. The Fionidae in turn belongs to the superfamily Fionoidea (Gray, 1857), a group containing four speciose other families: the (Bergh, 1889), the (Odhner, 1934), the Pseudovermidae (Thiele, 1931), and the (Iredale and O’Donoghue, 1923). The name Fionoidea reflects the fact that the Fionidae was the earliest member of this superfamily to be described. Many descriptions of Fiona pinnata (or its various synonyms) have remarked upon the peculiarities exhibited by this aeolid nudibranch and subsequently recommended it be given distinct genus or family status due to the degree of morphological divergence 11 from all other described aeolids (e.g., Alder & Hancock 1851; Suter 1913). These defining morphological characters include smooth that greatly resemble the oral tentacles, irregularly arranged cerata that lack cnidosacs but have unique frilly membranes running along their edges (Fig. 1.3), an anus that opens dorso-laterally, reproductive apertures that open independently, and the jaws and also show several unusual features (Suter 1913; Pruvot-Fol 1954; MacFarland 1966; Willan 1979; Behrens 1991). Clearly, much confusion has surrounded the taxonomic status of Fiona pinnata since its original discovery and this uncertainty has continued into the present day. Gosliner et al. (2008) recently revised the Tergipedidae to include Fiona pinnata, thereby eliminating the systematic status of the family Fionidae; this decision conflicts with the previously discussed morpho-anatomical evidence that suggests Fiona to be sufficiently divergent from other members of the Fionoidea as to merit the creation of a monotypic family (e.g., Alder and Hancock 1851b; Pruvot-Fol 1954; Suter 1913; MacFarland 1966). Furthermore, an increasing number of molecular phylogenies have demonstrated that a number of “cosmopolitan” marine invertebrates are in fact complexes of cryptic species. Morphological data alone cannot resolve the taxonomic status of Fiona pinnata, and thus a phylogenetic analysis on a global scale is necessary to clarify the systematics of this ubiquitous organism.

1.5 Thesis objectives The pelagic nudibranch Fiona pinnata has been recorded in the tropical and temperate zones of all oceans. Such a vast geographic distribution is potentially supported by its planktotrophic developmental mode and propensity for passive rafting, and thus its natural history may be consistent with its current classification as a single cosmopolitan species. However, in light of a host of recent studies unveiling molecular evidence for cryptic speciation within widespread marine invertebrates, it is possible that Fiona pinnata actually masks a species complex comprised of cryptic regional lineages. Although Fiona pinnata was one of the first aeolid nudibranchs to be identified and described, this study generated the first molecular phylogeny for this ubiquitous animal. The broad aim of this project was to relate how this research pertains to larger questions involving cryptic speciation and the role of long-distance rafting in marine biogeography. By exploring the genetic structure of Fiona pinnata 12 populations worldwide, we can shed light on its evolutionary history and thereby further our understanding of the role of passive rafting in maintaining large-scale genetic connectivity. I also sought to construct deep-rooted molecular phylogenies using mitochondrial and nuclear DNA sequences in order to explore the implications for this nudibranch’s systematic standing.

The specific objectives of this thesis are as follows:

• Chapter 2: To examine the genetic structure and diversity found within and among populations of Fiona pinnata in New Zealand and Chile, and to investigate the possibility that the West Wind Drift has served to mediate connectivity between these trans-oceanic populations. • Chapter 3: To explore the global phylogeography of Fiona pinnata, discuss possible explanations for its evolutionary history, and to also re-assess the systematic status of the family Fionidae using phylogenetic techniques.

These two objectives are addressed separately and thus each chapter is written in the style of an independent paper. Within each of these two chapters, mitochondrial and nuclear DNA sequence data are reported separately. Chapter 4 integrates the overall results of this research and discusses implications and future research directions in the study of oceanic rafting and marine invertebrate phylogeography. All references are compiled in a single section following this final chapter.

13

CHAPTER 2 Genetic structure and diversity of the rafting nudibranch Fiona pinnata within New Zealand and Chile, and the influence of the West Wind Drift on population connectivity

2.1 ABSTRACT

The aeolid nudibranch Fiona pinnata is found worldwide over a broad latitudinal range in both hemispheres. This marine invertebrate has a planktonic larval mode and becomes an obligate rafter when its free-swimming veligers settle on detached buoyant seaweeds and other floating objects. Gene flow in marine invertebrates is generally associated with larval dispersal potential, but passive rafting has also been recognized as an important mechanism of long-distance dispersal for sedentary organisms. Accordingly, marine invertebrates with high dispersal ability may not show genetic breaks consistent with biogeographic barriers. Given that there are several well-known marine biogeographic zones around New Zealand and Chile, I sought to determine whether genetic variation in Fiona pinnata was partitioned along any of these oceanographic boundaries. Specimens collected from these trans-oceanic coastlines were also compared in order to explore the possibility that the West Wind Drift serves to maintain genetic connectivity among rafting Fiona pinnata populations in the Southern Hemisphere. DNA sequence data (COI, ITS) was thus generated and analyzed for a total of 64 Fiona pinnata individuals collected from 33 rafts. No evidence of regional partitioning was found within Dunedin, New Zealand or Coquimbo, Chile, which indicates that genetic diversity in the local Fiona pinnata populations of these coastal areas was evenly distributed throughout the macroalgal rafts on which they occurred. Furthermore, relatively low levels of genetic differentiation were detected between the isolated coastlines of Chile and New Zealand, suggesting that the West Wind Drift has facilitated trans-Pacific gene flow in this rafting nudibranch. In contrast, a prominent phylogeographic break in Fiona pinnata populations was detected between the North and South Islands of New Zealand, which was concordant with known marine biogeographic boundaries and differing oceanic circulation regimes. The detection of moderately high 14 mitochondrial sequence divergences (maximum 6.9%) between these North and South clades may suggest that a species complex exists within “Fiona pinnata”. Although these two major clades were partitioned along temperate/sub-tropical boundaries, latitude alone cannot account for their geographic distribution patterns, as Coquimbo lies at lower latitudes than the North Island of New Zealand and yet Chilean specimens grouped firmly within the temperate clade. Instead, oceanic currents appeared to exert the largest influence on phylogeographic diversity and structuring, an idea that is supported by Fiona pinnata having planktonic larvae and also displaying a propensity for passive rafting.

2.2 INTRODUCTION

The aeolid nudibranch Fiona pinnata is classified as a pelagic organism but because it cannot swim or even float by itself post-metamorphosis, it thus spends its entire life cycle on floating objects (Rudman 2002). Accordingly, when free- swimming veligers colonize objects such as floating macroalgae or other flotsam with long-distance dispersal potential, they become obligate rafters (Thiel and Gutow, 2005b). This carnivorous invertebrate is also only found in association with a food resource due to its restricted foraging ability, and is generally found feeding upon fellow pelagic organisms such as Lepas sp. goose barnacles or pleustonic cnidarians (e.g., Velella velella, Porpita porpita). Fiona pinnata is considered to have a cosmopolitan distribution throughout tropical and temperate seas, and numerous New Zealand locality records exist for this obligate rafter (Willan 1979). New Zealand is an ideal location for phylogeographic research as it is geographically isolated, displays well-characterized oceanographic features, lies across a broad latitudinal range, and possesses a comprehensive geologic record (Buckley et al. 2001; Waters and Roy 2004b). Furthermore, previous work has found prominent breaks in the population structures of marine biota between the North and South Islands of New Zealand, owing to the competing influences of sub-tropical water regimes in the North and sub-Antarctic oceanic processes in the South (Dawson 1965; Heath 1985; Francis 1996). Indeed, two sister species of Fiona pinnata’s primary prey, goose barnacles of the genus Lepas, show a marked North/South division: the waters of New Zealand’s North Island are dominated by Lepas anatifera, 15 whereas Lepas australis is typically only found south of 39°S and is thus distributed throughout the South Island (Skerman 1958; Foster 1978; Hinojosa et al. 2006). Given that Fiona pinnata frequently associates with Lepas barnacles and shares their pelagic life history, a potential hypothesis could then follow that populations of this nudibranch along the New Zealand coastline will display a similar biological transition over this oceanographic boundary. On a more regional scale, the Southland Front off the east coast of New Zealand represents the boundary between sub-tropical and sub-Antarctic water masses (Chiswell 1996, 2009; Hopkins et al. 2010). The interaction between these two major circumpolar oceanic systems thus makes for a distinctive shelf-break system off the coast of Dunedin in the South Island. The Southern Ocean is known to exhibit an abundance of rafting macroalgae (Smith 2002) and floating seaweeds are particularly likely to occur near benthic source populations (Ingolfsson 1995; Hobday 2000b), especially in complex coastal systems with frontal zones such as the Dunedin region (Kingsford 1992; Fraser et al. 2011; Rothausler 2012). While oceanic currents certainly help to determine the direction and distance traveled by macroalgal rafts, the spatial dispersal patterns of floating seaweeds in coastal environments are also shaped by the prevailing winds (Tapia et al. 2004; Thiel and Gutow 2005a), and thus the interplay of these oceanographic and abiotic factors provides a unique opportunity to study patterns of fine-scale gene flow in the local rafting communities. The southeast Pacific coastline of South America is also noted to have distinct marine biogeographic zones (Knox 1960, Lancellotti and Vásquez 1999; Thiel 2002). The boundary between the warm-temperate Peru-Chilean province and the central Chilean transition zone (which contains elements of the southern cold-temperate Magellanic province) lies at 30°S and is also subject to coastal upwelling events (Brattstrom and Johanssen 1983; Camus 2001; Hinojosa et al. 2006; Vidal et al. 2008). Phylogeographic breaks at this latitude have been noted for certain species of marine invertebrates and brown algae (Meneses and Santelices 2000; Tellier et al. 2009), but other studies have demonstrated that species with high dispersal potential, such as the giant kelp Macrocystis pyrifera (Horta 2010), do not display genetic breaks across this biogeographic boundary (Gallardo and Carrasco 1996; Gomez- Uchida et al. 2003; Cardenas et al. 2009). Given that Fiona pinnata has free- swimming planktonic larvae and is frequently observed rafting on buoyant macralgae 16 at mid- to high-latitudes (Willan 1979; Hobday 2000a), a prediction could be made that levels of gene flow within this nudibranch are quite high and thus genetic partitioning across local biogeographic transitions will not be observed. In addition to serving as a regional comparison to the study of genetic structure in Dunedin, the examination of Fiona pinnata populations in Coquimbo will also allow an exploration of the possibility that these two southern temperate localities both harbor individuals of a common source population distributed by the West Wind Drift. Previous work (Waters and Roy 2004; Donald et al. 2005; Waters 2008; Nikula et al. 2010) has demonstrated that the West Wind Drift (aka Antarctic Circumpolar Current) can strongly influence the genetic connectivity of Southern Hemisphere marine invertebrates that rely on passive rafting for long-distance dispersal. I therefore sought to characterize the levels of genetic differentiation within and between populations of Fiona pinnata in New Zealand and Chile in order to determine whether this steady circumpolar current maintains genetic connectivity in this nudibranch. Obligate rafters such as Fiona pinnata are good models with which to test long- distance dispersal hypotheses, and by exploring patterns of genetic structure on varying geographic scales we can investigate the concordance of possible phylogeographic breaks with known biogeographic boundaries. DNA sequence data was thus generated from several independent molecular markers in order to address questions of phylogenetics and phylogeography on both local and trans-oceanic scales. The protein-coding cytochrome c oxidase I (COI) was chosen as the primary mitochondrial gene used in molecular analysis because it has repeatedly been shown to be suitable for resolving species-level diversity within nudibranchs (Thollesson 2000; Wollscheid-Lengeling et al. 2001; Grande et al. 2004; Turner & Wilson 2008). The nuclear non-coding internal transcribed spacer region (18S-ITS1-5.8S-ITS2-28S) has also proven useful for examining phylogenetic relationships at lower taxonomic levels within various molluscs, including nudibranchs (Raahauge and Kristenson 2000; Oliverio et al. 2002; Eriksson et al. 2006; Aguilera-Munoz et al. 2009).

2.3 METHODS

2.3.1 Sample collection 17

A total of forty-four specimens of Fiona pinnata were collected for genetic analysis from twenty-three different rafts around the North and South Islands of New Zealand (Table 2.1, Fig. 2.1). Thirty-four of these samples were taken at sea from the

R/V Polaris II during 60 km transects off the coast of Dunedin, Otago in the South Island of New Zealand. These cruises occurred approximately every two months, beginning in January 2011 and ending in May 2012. Fiona pinnata specimens were collected from nineteen floating rafts, and the geographic coordinates of each collection site were recorded using the boat’s GPS device. Sample sites 4 and 13 occurred beyond the Southland Front, while all other rafts were collected within the inshore side of this oceanographic boundary. In addition to these offshore samples, material was also collected in Dunedin following strong southerlies that deposited fresh beach-cast wrack on St. Clair Beach and Sandfly Bay. All Dunedin samples of Fiona pinnata were found on macroalgal rafts (either Durvillaea antarctica or Macrocystis pyrifera) in association with their Lepas sp. goose barnacle prey. Samples of Fiona pinnata were also collected from the North Island of New Zealand, including specimens found on a floating chilly bin lid off Goat Island, as well as material provided by the Auckland War Memorial Museum (voucher AK118324) that was taken from beach-cast wrack in Opoutere. All New Zealand samples were immediately preserved in absolute ethanol upon collection and subsequently stored in a refrigerator at ~6°C. Twenty specimens of Fiona pinnata were also obtained for genetic analysis from ten different floating rafts off the coast of Coquimbo, Chile (Table 2.2, Fig. 2.2). These cruises took place over six days in January 2012, and the GPS waypoints of each raft sampling site were recorded. Sample sites 1, 2, and 6 occurred at latitudes greater than 30°S while all other sites occurred at latitudes lower that 30°S. All specimens of Fiona pinnata were taken from floating macroalgal rafts (either Macrocystis pyrifera or Durvillaea antarctica) and were generally found in association with either Lepas pectinata or Lepas australis goose barnacles. Nudibranchs were immediately preserved in absolute ethanol upon collection.

18

Figure 2.1 Collection localities of 13 sites sampled for Fiona pinnata in New Zealand. Sample sites correspond to Table 2.1. The major oceanic surface currents off the northern and eastern coasts of New Zealand are shown in highly simplified form (after Chiswell 2009). The Dunedin area is represented by 11 distinct sample locations (inset), wherein the Southland Front divides sites 4 and 13 from all other rafts sampled in Dunedin.

19

Table 2.1 Collection details for Fiona pinnata specimens sampled in New Zealand. Sample sites are numbered chronologically and correspond to Figure 2.1. Latitude and longitude waypoints for offshore Dunedin samples are included in parentheses in collection locality data. Sample size (n) indicates the number of individuals sequenced for COI. AK=Auckland War Memorial Museum, AM=Australian Museum.

Sample Collection locality Voucher Collection Raft substrate n site (lat, long) date 1 Ohui Beach, AK118314 1/2/2008 Beach-cast wrack 2 Opoutere 2 St. Clair Beach, AM 26/2/2009 Beach-cast kelp 2 Dunedin, Otago C.476827 (D. antarctica) 3 Dunedin, Otago AM 10/1/2011 4 floating kelp rafts 9 (45° 49' 26" S, C.476807 - (D. antarctica) 171° 05' 02" E) 476810 4 Dunedin, Otago AM 1/3/2011 2 floating kelp rafts 4 (45° 50' 02" S, C.476811 - (D. antarctica) 171° 29' 58" E) 476812 5 Sandfly Bay, AM 20/10/2011 Beach-cast kelp raft 3 Dunedin, Otago C.476826 (D. antarctica) 6 Dunedin, Otago AM 8/11/2011 Floating kelp raft 2 (45° 48' 25" S, C.476813 (D. antarctica) 171° 08' 17" E) 7 Dunedin, Otago AM 8/11/2011 Floating kelp raft 2 (45° 48' 51" S, C.476814 (D. antarctica) 171° 11' 42" E) 8 Dunedin, Otago AM 8/11/2011 2 floating kelp rafts 4 (45° 48' 49" S, C.476815 - (D. antarctica) 171° 12' 31" E) 476816 9 Dunedin, Otago AM 8/11/2011 Floating kelp raft 2 (45° 48' 48" S, C.476817 (D. antarctica) 171° 14' 01" E) 10 Dunedin, Otago AM 8/12/2011 Floating kelp raft 3 (46° 11' 31" S, C.476819 (D. antarctica) 170° 14' 34" E) 11 Dunedin, Otago AM 10/1/2012 Floating kelp raft 2 (45° 47' 23" S, C.476818 (D. antarctica) 170° 58' 33" E) 12 Cape Rodney, AM 20/2/2012 Floating chilly bin 3 Goat Island, Leigh C.476793 lid 13 Dunedin, Otago, AM 24/5/2012 6 floating kelp rafts 6 (45° 47' 23" S, C.476820 - (M. pyrifera & 170° 58' 33" E) 476825 D. antarctica)

20

Figure 2.2 Collection localities for Fiona pinnata samples taken from 10 rafts off the coast of Coquimbo, Chile (inset). Sample sites are numbered chronologically and correspond to Table 2.2 (sites 1, 2, and 6 occurred at latitudes greater than 30°S). The major oceanic surface currents along the Chilean coast are shown in highly simplified form (after Vasquez and Vega 2007): HC=Humboldt Current, ACC=Antarctic Circumpolar Current, CHC=Cape Horn Current.

Table 2.2 Collection details for Fiona pinnata specimens sampled off the coast of Coquimbo, Chile. Each sample site represents a single floating raft, and two individuals from each raft were sequenced for COI. AM=Australian Museum.

Sample Collection site Voucher Collection Raft substrate site (latitude, longitude) date 1 30° 17' 15" S, AM C.476797 4/1/2012 M. pyrifera 71° 49' 35" W 2 30° 6' 18" S, AM C.476798 5/1/2012 M. pyrifera 71° 55' 17" W 3 29° 36' 33" S, AM C.476799 6/1/2012 M. pyrifera 71° 33' 46" W 4 29° 47' 00" S, AM C.476800 6/1/2012 M. pyrifera 71° 47' 20" W 5 29° 47' 38" S, AM C.476801 6/1/2012 M. pyrifera 71° 46' 15" W 6 30° 10' 03" S, AM C.476802 13/1/2012 D. antarctica 71° 29' 41" W 7 29° 49' 23" S, AM C.476803 16/1/2012 M. pyrifera 71° 42' 07" W 8 29° 49' 52" S, AM C.476804 16/1/2012 M. pyrifera 71° 44' 19" W 9 29° 51' 04" S, AM C.476805 16/1/2012 M. pyrifera 71° 46' 31" W 10 29° 22' 48" S, AM C.476806 18/1/2012 D. antarctica 72° 01' 60" W 21

Table 2.3 Primers used in PCR amplification of mitochondrial and nuclear genes. Region Primer name Direction Primer sequence (5’-3’) Reference COI LCO1490 forward GGTCAACAAATCATAA AGATATTGG Folmer et al. HCO2198 reverse TAAACTTCAGGGTGAC (1994) CAAAAAATCA H7005 reverse CCGGATCCACNACRTA Hafner et al. RTANGTRTCRTG (1994) CoxAF forward CWAATCAYAAAGATAT TGGAAC Colgan et al. CoxAR reverse AATATAWACTTCWGGG (2003) TGACC ITS1 18S forward TAACAAGGTTTCCGTAT Armbruster GTGAA et al. (2002) 5.8S reverse GCGTTCTTCATCGATGC ITS2 LSU-1 forward CTAGCTGCGAGAATTA ATGTGA Wade et al. LSU-3 reverse ACTTTCCCTCACGGTAC (2006) TTG Fiona_ITS2_F forward GTCTGAGGGTCGGACG ATAC This study Fiona_ITS2_R reverse ATCCCGGTTGGTTTCTT TTC

2.3.2 DNA extraction, PCR amplification, and sequencing DNA was extracted using a Chelex procedure (Walsh et al. 1991). A small piece of foot tissue (approximately 2 mm3) was dissected from each nudibranch and placed in 300-500 µl of 5% Chelex solution (BIO-RAD, Hercules, CA, USA). After 3 µl of proteinase-K (20 mg/ml) was added, the solution was incubated in a dry heat block at 60°C overnight, with the tissue constantly submerged in the Chelex beads. The extraction was subsequently vortexed, raised to 90°C for ten minutes, vortexed again, and then centrifuged at 14,000 rpm for ten minutes. The cooled supernatant was then ready for use in polymerase chain reaction (PCR) amplifications, and was otherwise kept refrigerated at 6°C when not in use. All PCR amplifications were performed in 20-µl volumes and contained 1 U of MyTaq DNA polymerase (Bioline), 1 µl of each primer (10 µM), 4 µl of 5X MyTaq 22

Red Buffer (Bioline), and 1 µl of extracted DNA. Thermal cycling for all reactions was carried out using Eppendorf Mastercycler ep gradient proS machines. A portion of the mitochondrial cytochrome c oxidase I (COI) was amplified for sixty-four individuals using various primers (Table 2.3). Most specimens successfully amplified using the primers LCO1490 (Folmer et al. 1994) and H7005 (Hafner et al. 1994), yielding 988 base pairs (bp) of sequenced data. Six individuals failed to amplify with these primers, but successfully amplified using either LCO1490 paired with HCO2198 (Folmer et al. 1994) or with the primer pair CoxAF and CoxAR (Colgan et al. 2003), resulting in either 700-bp or 567-bp fragments of COI, respectively. Thermal cycling for COI consisted of an initial denaturation step at 94°C for 3 mins, followed by 40 cycles of 94°C for 45 s (denaturation), 45-48°C for 45 s (annealing), and 72°C for 1.25 mins (extension), and a final extension phase at 72°C for 10 mins. Initially, a fragment of mitochondrial 16S rRNA (16S) was also sequenced for analysis, but this gene was extremely conserved across all sampled individuals (2 variable sites out of 421 sequenced bases; max 0.5% divergence among thirty sequences). Accordingly, 16S was discarded as a mitochondrial marker and a portion of cytochrome b was briefly examined in a subset of samples, as this gene has been useful in resolving the phylogenetic relationships among closely related groups of molluscs (Merritt et al. 1998). The cytochrome b findings were congruent with the COI topology and exhibited similar levels of genetic divergence (data not shown but see Appendix 2), and thus corroborated the mitochondrial diversity observed in COI. As a result, COI was henceforward used as the sole mitochondrial gene under investigation. In order to compare mitochondrial and nuclear patterns, sequence data was also generated from the nuclear internal transcribed spacer (ITS) region in a subset of Fiona pinnata samples. Not all specimens successfully amplified for this nuclear region, and thus only fifteen ITS1 sequences were generated for analysis. A fragment of the ITS1 region was amplified using the universal primers 18S and 5.8S (Armbruster et al. 2002), yielding an approximate 560-bp fragment containing the 3’ end of 18S, complete ITS1, and the 5’ end of 5.8S. The thermal cycling parameters for ITS1 were an initial denaturation step at 95°C for 2 mins, followed by 40 cycles of 23

93°C (30 s), 52-55°C (30 s), and 72°C (1 min), and a final extension phase at 72°C for 5 min. A fragment of the nuclear ITS2 region was amplified for a subset of fifteen individuals. The primers LSU-1 and LSU-3 (Wade et al. 2006) were used to amplify an approximate 714-bp fragment containing the 3’ end of 5.8S, complete ITS2, and the 5’ end of 28S. However, not all specimens amplified using these universal primers so novel Fiona-specific primers (Fiona_ITS2_F and Fiona_ITS2_R) were designed in order to target a 327-bp fragment nested within the ITS2-28S region that contained all variation observed in the longer portions of ITS2 obtained earlier. Thermal cycling conditions for ITS2 were an initial denaturation step at 96°C for 2 mins, followed by 45 cycles of 93°C (30 s), 48-59°C (30 s), and 72°C (1 min), with a final extension phase at 72°C for 7 mins. To confirm that amplifications were successful, all resulting PCR products were visualized by gel electrophoresis. Two µl aliquots of PCR products were run alongside size standards at 100 V on a 1% agarose gel (2 g of agarose in 200 mL 1X TAE) containing 1% SYBR Safe dye (Invitrogen). Gel purification was occasionally required to obtain clean ITS1 and ITS2 sequences, as these regions often result in multiple amplification products. All PCR products were purified using an Ultra-Sep Gel Extraction Kit (Omega) according to the manufacturer’s instructions. Purified DNA was subsequently quantified using a Nanodrop ND-1000 spectrophotometer. DNA sequencing was carried out at the University of Otago Genetic Analysis Services facility using an ABI 3730xl DNA Analyser. Sequencing reactions used the original PCR primers and all DNA fragments were sequenced in both forward and reverse directions. Reconciliation of forward and reverse reads was carried out in Sequencher v.5.0 (Gene Codes Corporation, Ann Arbor, MI), wherein sequences were assembled and aligned, and manual adjustments were made by eye. Sequences were further optimized in MEGA v.5.05 (Tamura et al. 2011), and for further confirmation of alignment, COI sequences were translated into amino acids to check for the presence of premature stop codons, which would have indicated sequencing errors or nuclear pseudogenes. COI sequences were easily aligned due to a lack of insertions and deletions (indels). ITS regions did contain indels, as is common in rRNA gene 24 sequences (Wade et al. 2006), and thus were first aligned in MEGA using the ClustalW alignment program run at default parameters (Higgins et al. 1994) and then re-checked by eye. In order to verify that contamination had not occurred and that the correct gene had been amplified, GenBank “blast” searches (Benson et al. 2005) were performed on the earliest sequences obtained for Fiona pinnata; in the case of COI, these searches revealed moderate genetic similarity between Fiona pinnata and sequences available for other members of the superfamily Fionoidea (data not shown). All DNA sequences generated from Fiona pinnata in this study will be deposited in the NCBI GenBank database.

2.3.3 Analysis of mitochondrial DNA variation COI sequence divergences among Fiona pinnata samples were calculated using the Kimura (1980) two-parameter (K2P) model of sequence evolution. Sequences were inspected using TCS v.1.21 analysis (Clement et al. 2000) in order to identify haplotypes that were shared among individuals, and an unrooted statistical parsimony network based on the unique COI haplotypes was then generated. Three sequences were excluded from TCS network analysis due to missing data (i.e., < 988-bp fragments), which would have skewed statistical parsimony analysis because all variation outside of the incomplete sequence data available for these samples would have been ignored. These three incomplete New Zealand sequences included one specimen from the South Island and two specimens from the North Island. The phylogenetic relationships among the COI haplotypes were estimated using maximum likelihood and Bayesian methods (Geyer 1991). The best-fit model of sequence evolution for the COI dataset (HKY+I; Hasegawa et al. 1985) was determined using the Akaike information criterion (Akaike 1974) implemented in jModelTest v.0.1.1 (Guindon and Gascuel 2003; Posada and Buckley 2004; Posada 2008). As Fiona pinnata is thought to be closely related to the Tergipedidae (e.g., Gosliner et al. 2008), two members of this family (Cuthona ocellata, AY345043; , AY345032) were included as outgroups in phylogenetic analyses. Initial assessment using neighbour joining was also performed (data not shown). A maximum likelihood (ML) tree was constructed using MEGA, and branch support was evaluated by bootstrapping (1000 replicates). Bayesian analysis (settings: nst=2, rates=inv) was performed in Mr. Bayes v.3.1.2 (Huelsenbeck and Ronquist 2001; 25

Ronquist and Huelsenbeck 2003) in order to estimate Bayesian posterior probability (PP) values. The Markov chain Monte Carlo search ran for twenty million generations, and trees were sampled every 100 generations. The first 10,000 trees (i.e., 1,000,000 generations) were discarded as burn-in. Incomplete COI sequences were included in all model-based analyses, as the incorporation of samples with missing data is not only preferable to excluding them, but presumably improves overall phylogenetic resolution (Philippe et al. 2004; Wiens 2006). In order to investigate the distribution of genetic variation at various geographic scales, hierarchical analyses of molecular variation (AMOVA) were performed using ARLEQUIN v.3.1 (Excoffier et al. 2005). I tested whether genetic diversity in Fiona pinnata populations was partitioned in concordance with biogeogeographic breaks represented by: (1) the Southland Front off the east coast of New Zealand; (2) a North/South split in New Zealand; (3) trans-Pacific coastlines; and (4) the 30°S split between Peruvian and Magellan marine provinces in Chile. For each AMOVA, a raft was defined as a population.

2.3.4 Analysis of nuclear DNA variation Pairwise divergence values among Fiona pinnata ITS1 sequences were calculated using the K2P model of sequence evolution. TCS 1.21 was used to identify all unique ITS1 sequences, and a haplotype network was constructed after the removal of redundant sequences. Treatment of alignment gaps (indels) as characters in network analysis differed little from analyses in which alignment gaps were ignored, and thus the ITS haplotype networks presented herein are based on analyses in which gaps were excluded. Phylogenetic analysis of ITS1 consisted of maximum likelihood methods and Bayesian inference. The most appropriate model for nucleotide substitution for ITS1 (HKY+I+G) was determined using the AIC within jModelTest v.0.1.1. Due to a lack of published ITS1 data available for nudibranchs, the aeolid nudibranch verrucosa (AB180834) served as the outgroup but only approximately 100-bp of this outgroup sequence was alignable with Fiona pinnata ITS1 fragments. An ML tree was constructed using MEGA, and branch support was evaluated by bootstrapping (1000 replicates). Bayesian analysis (settings: nst=2, rates=invgamma) performed in Mr. Bayes v.3.1.2 ran for twenty million generations, and trees were sampled every 26

100 generations. The first 10,000 trees (i.e., 1,000,000 generations) were discarded as burn-in, and the remaining trees were used to estimate PP values. Due to the low variation observed within ITS2, a haplotype network for this gene was constructed in TCS 1.21 but no further phylogenetic analysis of this nuclear region was undertaken.

2.4 RESULTS

2.4.1 Mitochondrial COI variation Thirty-three rafts in New Zealand and Chile were sampled for Fiona pinnata, and one to three individuals from each raft were generally used in molecular analysis. COI fragments contained 104 variable sites, and a total of sixty-four COI sequences yielded thirty-four unique haplotypes overall, one of which was shared between New Zealand and Chile (see Appendices 3 and 4 for haplotype frequencies in all sampling locations). Bayesian inference produced a topology very similar to the ML tree, and both analyses were congruent in the recovery of two major clades (Fig. 2.3). Both phylogenetic analyses showed maximum support for the monophyly of all Fiona pinnata specimens (ML: 100%; Bayesian PP: 1.00), while two major clades emerged as sister groups. Clade A contained all Chilean specimens of Fiona pinnata as well as ~92% of all individuals sampled in the South Island NZ, and received very high support (ML: 95%; Bayesian PP: 1.00). Some slight phylogeographic structuring was evident within clade A, wherein haplotypes from South Island NZ and Chile each clustered into shallow groups. The grouping of South Island NZ specimens was moderately supported (ML: 72%; Bayesian PP: 0.96), while the majority of Chilean specimens formed a weakly supported group (ML: 39%; Bayesian PP: 0.63). However, one Chilean haplotype clustered intermediately and another haplotype was shared between these two countries. Furthermore, these two regional groupings were differentiated by only 1.2-1.9% COI sequence divergence. In contrast, Clade B contained all specimens collected in North Island NZ as well as three South Island NZ individuals, and received much weaker support (ML bootstrap 38%; Bayesian PP 0.60). A depiction of the geographical distribution of these two major clades can be seen in Figure 2.4. Mean divergence between these 27 two mitochondrial clades was 6.2%, while variation within both clade A (mean 1.01%) and clade B (mean 0.49%) was quite low (Table 2.4). Fiona pinnata specimens exhibited substantial genetic differentiation from both of the outgroup sequences (19.9-21.5%). The COI haplotype network generated in TCS (Fig. 2.5) also highlights the existence of two distinct clades, and illustrates that one of the haplotypes sampled in South Island NZ was very prevalent as it was found on eighteen of the twenty-one rafts sampled in Dunedin, and was shared among ~62% of the specimens sequenced from this area. This network also shows that one of the COI haplotypes was shared between two Chilean specimens and an individual from South Island NZ. Hierarchical analysis of COI diversity within and among Fiona pinnata populations using AMOVA revealed no basis for geographic partitioning among the rafts sampled in either Dunedin or Coquimbo (Table 2.5). The majority of genetic variation within both of these regions occurred within rafts, and was not partitioned among sampling sites or along local biogeographic boundaries. In contrast, significant subdivision was observed between the North Island and South Island of New Zealand, as molecular variance was geographically congruent with a North/South break. A regional comparison between Chile and South Island NZ found that the majority of COI variation was distributed between these two distant landmasses (74%; P = 0.000), while within-raft variation accounted for the remaining portion of genetic variation (P = 0.000).

Table 2.4 Minimum and maximum percent divergences among Fiona pinnata COI sequences calculated with the Kimura (1980) 2-parameter model of sequence evolution. Intra-clade divergence values are italicized.

Clade A Clade B

Clade A 0.1 - 1.9% Clade B 5.1 - 6.9% 0.1 - 0.9%

28

Figure 2.3 Maximum likelihood bootstrap consensus tree of Fiona pinnata haplotypes collected in New Zealand and Chile based on COI data. Bootstrap support values above branches are derived from ML analysis (1000 replicates), and values below branches are Bayesian posterior probabilities. The numbers in parentheses following some haplotypes indicate frequency of haplotype occurrence if greater than one. 29

Figure 2.4 Geographic distribution of Fiona pinnata COI clades among New Zealand and Chile. Sample sizes of individuals sequenced for COI are given, and a highly simplified representation of the overall phylogenetic structure is also included.

Table 2.5 Analyses of molecular variance (AMOVA) of Fiona pinnata populations based on COI data. Genetic distances between haplotypes were calculated according to the Kimura (1980) 2-parameter model. Geographic partitioning was tested in: (1) Dunedin, (2) North vs. South Island NZ, (3) South Island NZ vs. Chile, and (4) Coquimbo. Percent of Source of variation P value Fixation indices variation Analysis 1 Among regions -2.98 0.78690 Fct = - 0.02979 Among rafts, within regions 11.43 0.37732 Fsc = 0.11096 Within rafts 91.55 0.33431 Fst = 0.8448 Analysis 2 Among regions 81.81 0.01173 Fct = 0.81808 Among rafts, within regions 6.60 0.02346 Fsc = 0.36297 Within rafts 11.59 0.00000 Fst = 0.88411 Analysis 3 Among regions 74.37 0.00000 Fct = 0.74369 Among rafts, within regions -2.60 0.52004 Fsc = - 0.10142 Within rafts 28.23 0.00000 Fst = 0.71770 Analysis 4 Among regions -5.08 0.87195 Fct = - 0.05080 Among rafts, within regions -1.94 0.75269 Fsc = - 0.01842 Within rafts 107.02 0.75562 Fst = - 0.07016

30

Figure 2.5 Unrooted statistical parsimony network for Fiona pinnata. Each colored circle/pie represents a unique COI haplotype and sizes are drawn in proportion to relative haplotype frequencies. Small empty circles represent intermediate hypothesized haplotypes. Haplotypes are colored according to their geographic origin.

31

2.4.2 Nuclear ITS variation A total of fifteen ITS1 sequences obtained from ten different rafts in New Zealand and Chile contained sixteen variable sites and yielded twelve distinct haplotypes. ML and Bayesian analyses produced similar topologies, although the resulting tree showed relatively shallow overall phylogeographic structure (Fig. 2.6). Despite the fact that the phylogenetic relationships were not as well resolved in this nuclear region, ITS1 was broadly congruent with the COI results due to a break detected between North Island NZ samples and Chilean/South Island NZ haplotypes. Specimens sampled from North Island NZ formed a relatively well-supported clade (ML: 74%; Bayesian PP: 0.83), while all Chilean and South Island NZ individuals comprised a second clade with weaker bootstrap support (ML: 66%; Bayesian PP: 0.44). Most important to note however, is that the three South Island NZ specimens that clustered within clade B in COI analysis instead group with clade A specimens in ITS1 analysis (the placement of these three individuals is denoted in Fig. 2.6 with COI clade symbols taken from Fig. 2.3). Mean divergence between these two nuclear clades was 2.2%, and intra-clade divergences within both the North Island NZ group (0.2-0.8%) and the Chile/South Island NZ group (0.2-1.7%) were relatively low. The haplotype network generated for ITS1 (Fig. 2.7) also suggests the presence of two shallow clades and shows that two of the ITS1 haplotypes were present in both Chilean and South Island NZ samples. The ITS2 region exhibited very low variation (mean 0.7% K2P divergence) and only four closely related haplotypes, distinguished by four variable sites, were found overall (Fig. 2.8). The most common of these haplotypes, present in ~73% of the specimens sequenced for ITS2, was detected in samples from Chile and both islands of New Zealand.

32

Figure 2.6 Maximum likelihood bootstrap consensus tree of Fiona pinnata haplotypes collected in New Zealand and Chile based on ITS1 data. Bootstrap support values above branches are derived from ML analysis (1000 replicates), and values below branches are Bayesian posterior probabilities. The numbers in parentheses following some haplotypes indicate frequency of haplotype occurrence if greater than one. The circles/pies along the right margin are COI clade symbols taken from Fig, 2.3, wherein white represents a temperate mitochondrial haplotype (clade A) and black corresponds to a sub-tropical mitochondrial haplotype (clade B).

Figure 2.7 Unrooted statistical parsimony network for Fiona pinnata. Each colored circle/pie represents a unique ITS1 haplotype and sizes are drawn in proportion to relative haplotype frequencies. Small empty circles represent intermediate hypothesized haplotypes. Haplotypes are colored according to their geographic origin.

33

Figure 2.8 Unrooted statistical parsimony network for Fiona pinnata. Each colored circle/pie represents a unique ITS2 haplotype and sizes are drawn in proportion to relative haplotype frequencies. Small empty circles represent intermediate hypothesized haplotypes. Haplotypes are colored according to their geographic origin.

2.5 DISCUSSION

Phylogenetic analysis of mitochondrial COI sequence data revealed the presence of two divergent lineages of Fiona pinnata within the Southern Hemisphere of the Pacific Ocean. These two mitochondrial clades were separated by 5.1-6.9% sequence divergence, and given that dispersal in this nudibranch appears to be current- mediated, the prominent break in their geographic distributions is likely a reflection of contrasting oceanic circulation patterns. One clade was abundant and widespread throughout temperate coastal regions of Chile and the South Island of New Zealand, while the other clade was chiefly represented by specimens collected in the far north of New Zealand. In comparison, the phylogeographic structure observed in nuclear ITS1 data was shallower and not as well resolved. However, a distinction between temperate and sub-tropical lineages was still apparent in this nuclear region and thus 34 these two independent markers were in broad agreement over the major phylogeographic patterns.

2.5.1 Genetic structure of Dunedin populations Molecular variance in Fiona pinnata specimens collected from floating macroalgal rafts off the coast of Dunedin, New Zealand was not concordant with a biogeographic break across the Southland Front. Furthermore, an AMOVA also revealed that the majority of genetic variation observed in these nudibranchs occurred within rafts and there was no evidence for population subdivision at a regional scale. The high dispersal potential of this known rafter has evidently resulted in ongoing genetic connectivity within this coastal convergence zone (mean 2.5% divergence), and a common haplotype was shared among ~62% of all sequenced individuals and was present on ~86% of the sampled rafts. These findings imply that genetic diversity in this marine invertebrate is well distributed throughout its offshore range, and thus an alternative hypothesis of self-recruiting populations with high genetic structuring is rejected (Swearer et al. 2002). Given that planktotrophic larvae are assumed to have unrestricted dispersal (Shuto 1974; Strathmann 1990) and also that macroalgal rafting strongly promotes gene flow on smaller spatial scales (Rothausler 2012), these results are perhaps unsurprising for an animal with free-swimming veliger larvae and a propensity for passive rafting. Although the two major clades detected in COI analysis were largely allopatric, some intermixing was evident within the South Island of New Zealand. Dunedin populations were predominantly composed of haplotypes that grouped closely together with Chilean specimens in a well-supported temperate clade, but three Dunedin individuals sequenced for COI were found to possess mitochondrial haplotypes representative of the divergent North Island lineage. These three ‘sub- tropical’ specimens were found co-occurring in sympatry with ‘temperate’ individuals on two separate kelp holdfasts. However, nuclear ITS1 data did not corroborate the divergent mitochondrial haplotypes displayed by these three animals, and thus the discrepancies between their mitochondrial and nuclear loci could simply be due to incomplete lineage sorting in the slower-evolving ITS region. On the other hand, although the two major lineages of Fiona pinnata are largely allopatric, it is possible that they may not be reproductively isolated and thus freely interbreed when they do 35 co-occur (Bigelow 1965). Given that nudibranchs are simultaneous hermaphrodites (Gosliner et al. 2008) and that complete gamete compatibility may persist between sister species for as long as five million years (Ziegler et al. 2005), hybridization- mediated mitochondrial introgression represents another possible explanation for the minor proportion of sub-tropical haplotypes detected in Dunedin. Dunedin thus may represent a hybridization zone in which mitochondrial introgression has occurred (van Oppen et al. 2001; Schroth et al. 2002; Depraz et al. 2009), although whether this potential introgression occurred historically or is currently maintained is unknown. Given that hybridization appears limited (sub-tropical mitochondrial haplotypes accounted for only ~8% of the total individuals sampled for COI in Dunedin), it is possible that strong differential selection is operating to limit the number of successful hybrids. Alternatively, this prospective hybridization may be quite recent, possibly due to a southward range extension of tropical Fiona populations as a result of climate change, which would imply that hybridization in the higher latitudes will become increasingly prevalent in the years to come.

2.5.2 North/South New Zealand dynamics The geographic distribution of the two major lineages of Fiona pinnata, as well as the overall partitioning of genetic variation within New Zealand populations are consistent with a North/South split. An AMOVA showed that the majority (82%; P < 0.020) of the total variation observed within New Zealand occurred between the North and South Islands, and the fixation index value for this pairwise comparison also revealed significant differentation (FST: 0.884, P = 0.000). This North/South division in population structure is concordant with known marine biogeographic dynamics that act as strong barriers to dispersal and gene flow (Pawson 1965; Apte and Gardner 2002). Genetic discontinuities have been found to coincide in a number of taxa at 42°S, and these marine disjunctions are often attributed to converging water masses and coastal upwelling regimes in the Cook Strait region (Waters and Roy 2004b; Ayers and Waters 2005; Goldstien et al. 2006). The North Island of New Zealand is broadly characterized by sub-tropical, southward-flowing watermasses while the South Island is differentiated by sub-Antarctic, northward-flowing watermasses (Chiswell 2009), and thus it appears that the phylogeographic break 36 observed in Fiona pinnata populations along this coastline is consistent with these well-defined oceanic circulation patterns.

2.5.3 Genetic structure of Coquimbo populations All specimens of Fiona pinnata collected from floating macroalgal rafts in the Coquimbo region of Chile grouped closely together in both mitochondrial COI (mean 0.6% divergence) and nuclear ITS1 analyses (mean 0.5% divergence). Furthermore, although samples were obtained across a recognized biogeographic boundary at 30°S (Brattstrom and Johanssen 1983; Santelices and Meneses 2000), wherein sample sites 1, 2, and 6 occurred at latitudes greater than 30°S while all other rafts occurred at latitudes lower than 30°S, an AMOVA revealed no basis for genetic partitioning consistent with this biogeographic break. Instead, genetic diversity was well distributed throughout this coastal region and the majority of molecular variance occurred within rafts, which suggests that high levels of gene flow are maintained among populations of this rafting nudibranch.

2.5.4 Southern Hemisphere patterns Phylogenetic analyses of COI and ITS1 sequence data were concordant in the recovery of a monophyletic group composed of Chilean and South Island NZ specimens, and these trans-oceanic populations were differentiated by low sequence divergences (1.2-1.9%). Trans-Pacific haplotype sharing was also detected: one COI haplotype as well as two ITS1 haplotypes were sampled in both of these distant temperate localities. The observed levels of genetic diversity among these widespread populations of Fiona pinnata indicate that this nudibranch is capable of trans-oceanic dispersal. Long-distance rafting has previously been demonstrated to enable trans-oceanic gene flow in a variety of marine invertebrates and buoyant macroalgae in the Southern Hemisphere (Waters and Roy 2004a; Donald et al. 2005; Fraser et al. 2009; Horta 2010; Nikula et al. 2010), and the strong easterly flow of the West Wind Drift also plays an important role in shaping southern temperate marine biogeography (Waters 2008). The low temperatures of the sub-Antarctic enhance the longevity of floating seaweeds at the sea surface (Rothausler 2009), and indeed an estimated seventy million rafts of the bull kelp Durvillaea antarctica are afloat in the Southern Ocean at any given time (Smith 2002), which highlights the prevalence of 37 macroalgal rafting at high latitudes. The West Wind Drift provides long-distance dispersal opportunities for planktonic veligers as well as egg masses and post- metamophic individuals of Fiona pinnata that occur on kelp rafts, and the high degree of genetic similarity detected between New Zealand and Chilean specimens reveals that this powerful circumpolar current has been effective in mediating population connectivity between these two isolated coastlines. The ubiquitous distribution observed for Fiona pinnata within the temperate Southern Hemisphere is thus partially substantiated with the molecular data presented in this study, although the detection of a divergent North Island NZ lineage suggests that cryptic speciation has been driven on a broader geographic scale as a result of varied oceanic circulation regimes. This hypothesis remains to be tested for Fiona pinnata populations worldwide, and thus a global phylogeographic analysis is necessary to robustly assess the cosmopolitan status attributed to this nudibranch.

38

CHAPTER 3 Global phylogeography, evolutionary history, and systematic status of the aeolid nudibranch Fiona pinnata

3.1 ABSTRACT

The pelagic aeolid nudibranch Fiona pinnata (Mollusca: Gastropoda) is considered a single cosmopolitan species that occurs throughout tropical and temperate regions in all oceans. This invertebrate’s systematic status is based entirely on morphological characters and has been disputed throughout its taxonomic history, and its ubiquitous distribution has been attributed to its planktotrophic larval stage and propensity for passive rafting. In order to examine Fiona pinnata in a molecular phylogenetic framework, DNA sequences of mitochondrial COI and nuclear ITS were obtained from 84 specimens collected from 45 rafts worldwide. Substantial genetic diversity was observed within Fiona pinnata populations and the global COI molecular phylogeny uncovered three distinct clades separated by 6.1-12.4% mean sequence divergence. Two of these clades were widespread and generally allopatric, exhibiting circum-equatorial and anti-tropical distributions, although they were found to co-occur in the South Island of New Zealand. The third clade was represented by a single Indonesian sample collected in SE Sulawesi that showed marked divergence within several mitochondrial and nuclear genes from all other ingroup taxa. Phylogenetic relationships were not as well resolved in the nuclear ITS1 phylogeny, which showed shallower overall phylogeographic structure and slight incongruence with COI data. A molecular clock suggests that these clades diverged approximately 4.9-12.8 millon years ago, corresponding to widespread cooling and glaciation events in the late Miocene that may have driven cryptic speciation. Overall these findings indicate that this obligate rafter is highly capable of long-distance dispersal and that intra-clade gene flow has occurred on trans-oceanic scales. However, the regional partitioning of genetic variation is consistent with oceanic circulation patterns and the distributions of the three clades are linked to prominent breaks along marine biogeographic zones. Deep-rooted phylogenies constructed for several mitochondrial and nuclear markers strongly supported the monophyly of all sampled haplotypes, and 39 showed that Fiona pinnata was highly divergent from even its presumed closest relatives in the superfamily Fionoidea, thereby providing support for the systematic standing of the historically monotypic family Fionidae.

3.2 INTRODUCTION

Due to a lack of conspicuous barriers to gene flow in the marine environment, as well as the observation that many organisms display morphological uniformity throughout their broad geographic ranges, genetic connectivity among populations has historically been considered much higher in the oceans than in terrestrial environments (Mayr 1954; Palumbi 1992; de Vargas et al. 2003). Furthermore, dispersal potential in the sea has generally been coupled with larval mode and duration (Scheltema 1971; Strathmann 1987; Hedgecock 1986; Jablonski 1986), whereby animals exhibiting long-lived planktonic larvae are presumed to have suppressed opportunities for allopatric speciation (Shuto 1974; Hellberg 1996; Collin 2001). However, recent molecular evidence derived from a variety of marine taxa has revealed that cryptic speciation is quite common in the sea, and several ‘widespread’ pelagic organisms have been demonstrated to comprise cryptic species complexes (Knowlton 1993; Lessios et al. 2001; Wilson et al. 2007; Schwaninger 2008). Furthermore, it is becoming increasingly clear that developmental strategy and larval lifespan are not always good predictors of species distribution and levels of gene flow (Sponer 2002; Waters and Roy 2004; Nikula et al. 2010). Biogeographical studies employing molecular tools have frequently debunked the cosmopolitan status of marine invertebrates, showing instead that these seemingly ubiquitous species are actually composed of multiple cryptic lineages with regionally restricted distributions (Klatau et al. 1999; Schroth et al. 2002; Zulliger and Lessios 2010). Cryptic diversity has been uncovered in several widespread pleustonic cnidarians including Aurelia aurita (Dawson and Jacobs 2001) and Physalia physalis (Pontin and Cruickshank 2012), as well as in noted rafters such as Membranipora bryozoans (Schwaninger 2008), Rhabditis polychaetes (Derycke et al. 2008), and Amphipolis sea stars (Sponer and Roy 2002). The aeolid nudibranch Fiona pinnata occurs worldwide over a broad latitudinal range and is thus recognized as cosmopolitan, although its distribution is based solely 40 on the morphological similarities observed between far-flung populations. Given that this animal has planktotrophic larvae and is frequently observed rafting on the high seas (Bushing 1994; Aliani and Molcard 2003; Thiel and Gutow 2005b), both of which are processes promoting long-distance dispersal, it is conceivable that Fiona pinnata populations worldwide are so well mixed that they can be legitimately pooled into a single cosmopolitan species. However, this assumption has never before been robustly tested with molecular phylogenetics. Given the enormous distribution attributed to this nudibranch, it could be predicted that local and regional populations would show high genetic structuring, as has been demonstrated in a number of widespread marine invertebrates (Lee 2000; Sponer 2002; Wilson et al. 2009). This study therefore entails a global genetic analysis of specimens collected from different rafts worldwide in order to elucidate the genetic structure across widespread populations of this ubiquitous nudibranch. Specimens were sampled from the Pacific, Atlantic, and Indian Oceans from as far north as 58 °N (Cross Sound, Alaska) and as far south as 45 °S (Dunedin, New Zealand) in order to test for cryptic speciation on a global scale. Using sequence analysis of mitochondrial and nuclear DNA (COI, ITS), I tested whether Fiona pinnata populations worldwide are indeed genetically homogeneous, warranting single species status, or if this nudibranch instead conceals a cryptic species complex. Fiona pinnata has historically been assigned to the monotypic family Fionidae (Gray 1857), as many early authors determined that this aeolid nudibranch “presents peculiarities that forbid its being associated generically with any known form [of aeolid]” (Alder and Hancock 1855). The Fionidae in turn belongs to the superfamily Fionoidea (Gray, 1857), a group containing four speciose other families: the Tergipedidae (Bergh, 1889), the Eubranchidae (Odhner, 1934), the Pseudovermidae (Thiele, 1931), and the Calmidae (Iredale and O’Donoghue, 1923). Fiona pinnata exhibits several morphological characteristics unique among aeolid nudibranchs, and some of the most conspicuous features relate to the cerata, which lack cnidosacs but have a distinctive sailshape edge (Behrens 1992; Rudman 2002). Indeed, the membranous frill running along the cerata is often used as the most reliable basic morphological character to identify this species (e.g., Scarabino 2004). However, the Fionidae’s long-standing designation within the superfamily Fionoidea has been disputed in recent years, as Gosliner et al. (2008) recently revised the Tergipedidae to 41 include Fiona pinnata, although the basis for this change in classification was not given. A molecular phylogeny is required to evaluate the phylogenetic relationships between Fiona pinnata and other members of the Fionoidea, and thus DNA sequence data was generated from several independent markers with different mutational rates in order to assess the systematic standing of the Fionidae. The aims of this study were to test the cosmopolitan status of Fiona pinnata, to infer a possible scenario for its evolutionary history, and to clarify its systematics within the superfamily Fionoidea. I sought to resolve these questions by assessing the levels of genetic connectivity in this nudibranch over various geographic scales, and also by examining its phylogenetic relationships at higher taxonomic levels. By investigating these global patterns we can explore the extent to which passive rafting maintains gene flow over trans-oceanic scales and thereby further our understanding of the evolutionary significance of this biological process.

3.3 METHODS

3.3.1 Sample collection A total of eighty-four individuals of Fiona pinnata were used in molecular analyses, including both recently collected samples as well as preserved museum material (Table 3.1). The specimens were obtained from fifteen sampling locations worldwide (Fig. 3.1), and were preserved in 70-96% ethanol during this study.

3.3.2 DNA extraction, PCR amplification, and sequencing A small portion (approximately 2 mm3) of foot tissue was dissected from each individual. This tissue was transferred to 300-500 µl of 5% Chelex solution (Walsh et al. 1991) and after adding 3 µL of proteinase-K (20 mg/ml), the solution was incubated at 60°C overnight. The extraction was resuspended by vortexing, raised to 90°C for ten minutes, vortexed again, and finally centrifuged at 14,000 rpm for ten minutes. The resulting supernatant was used as the DNA template in subsequent polymerase chain reaction (PCR) amplifications. In order to explore the global phylogeography of Fiona pinnata, a portion of mitochondrial cytochrome c oxidase I (COI) as well as two fragments of the nuclear rRNA gene-cluster were investigated. All PCR amplifications were performed in 20- 42

µl volume reactions containing 1 U of Taq DNA polymerase (Bioline), 1 µl of each primer (10 µM), 4 µl of 5x MyTaq Red Buffer (Bioline), and 1 µl of extracted DNA. Thermal cycling for all reactions was carried out using Eppendorf Mastercycler ep gradient proS machines. A portion of COI was amplified for eighty-four specimens using various combinations of primers (Table 3.2). Most individuals amplified using the primers LCO1490 (Folmer et al. 1994) and H7005 (Hafner et al. 1994), yielding 988 base pairs (bp) of data following sequencing. However, a fraction of the samples, particularly older museum material, failed to amplify with these primers and thus various other primer pairs were used (see Chapter 2), resulting in 454-759 bp fragments (typically > 600-bp). Thermal cycling conditions for COI were 94°C for 3 mins, followed by 40 cycles of 45 s at 94°C, 45 s at 45-48°C, 1.25 mins at 72°C, and a final extension for 10 mins at 72°C. An approximate 560-bp fragment of the nuclear internal transcribed spacer one (ITS1) region, including the 3’ end of 18S rRNA, complete ITS1, and the 5’ end of 5.8S rRNA, was successfully amplified for twenty-six individuals using the primers 18S and 5.8S (Armbruster et al. 2002). Several specimens failed to amplify for this region, and hence only a subset of sequences was obtained. Thermal cycling for ITS1 entailed an initial denaturation step of 2 mins at 95°C, followed by 40 cycles of 30 sec at 93°C, 30 sec at either 52-55°C, 1 min at 72°C, and a final extension for 5 mins at 72°C. An approximate 714-bp fragment of the nuclear internal transcribed spacer two (ITS2) region, including the 3’ end of the 5.8S rRNA gene, complete ITS2, and the 5’ end of the 28S rRNA gene, was successfully amplified for thirteen individuals using the primers LSU-1 and LSU-3 (Wade et al. 2006). However, some specimens failed to amplify with these universal primers and thus two novel primers (Fiona_ITS2_F and Fiona_ITS2_R) were designed to specifically target a 327-bp fragment nested within this region. Sequence data from this 327-bp portion was generated in eleven additional individuals, resulting in twenty-four ITS2 sequences altogether. Thermal cycling for ITS2 involved an initial denaturation step of 2 mins at 96°C, followed by 45 cycles of 30 sec at 93°C, 30 sec at either 48°C (LSU-1/LSU-3 primer pair) or 59°C (novel primers), 1 min at 72°C, and a final extension for 7 mins at 72°C. 43

Three additional genes were examined in a subset of samples in order to investigate the systematic standing of Fiona pinnata at higher taxonomic levels. A 421-bp fragment of the mitochondrial 16S rRNA gene was amplified and sequenced for thirty-seven individuals using the universal primers 16Sar and 16Sbr (Palumbi et al. 1996). Thermal cycling for 16S consisted of the following conditions: an initial denaturation step at 94°C for 3 mins, followed by 40 cycles of 30 s denaturation at 94°C, 30 s annealing at 45°C, 1 min extension at 72°C, and a final extension step for 7 mins at 72°C. Fragments of the nuclear 18S rRNA (18S) and histone protein H3 (H3) genes were investigated in a small subset of the most divergent (based on COI findings) specimens. A 328-bp fragment of H3 was amplified and sequenced for seven individuals using the primers H3aF and H3aR (Colgan et al. 1998). Thermal cycling parameters for H3 involved an initial denaturation step of 3 mins at 94°C, 40 cycles of 45 s at 94 °C, 45 s at 50 °C, and 1.75 mins at 72°C, followed by a final extension step of 10 mins at 72°C. Lastly, a 1305-bp fragment of 18S was amplified and sequenced for seven individuals using the primers 18S1F (Okusu et al. 2003) and 18Sbi (Whiting et al. 1997). Thermal cycling conditions for 18S consisted of an initial denaturation step of 3 min at 94°C, 40 cycles of 45 sec at 94°C, 45 sec at 50°C, and 1.75 mins at 72°C, followed by a final extension step of 10 min at 72°C. Two µl aliquots of all PCR products were visualized by agarose gel electrophoresis in order to confirm that amplifications were successful. ITS1 and ITS2 occasionally resulted in multiple amplification products, in which case gel purification was performed in order to isolate the target fragment. All PCR products were purified using an Ultra-Sep Gel Extraction Kit (Omega) according to the manufacturer’s instructions, and then sequenced in both forward and reverse directions using the original PCR primers at the University of Otago Genetic Analysis Services facility. Sequencher v.5.0 (Gene Codes Corporation, Ann Arbor, MI) was used to initially assemble and align all sequences by eye. Sequences were then imported into MEGA v.5.05 (Tamura et al. 2011) and alignment was further optimized by using ClustalW (Higgins et al. 1994), and then re-checked by eye. Protein-coding regions (i.e., COI and H3) were translated into amino acids to ensure no premature stop codons, indicative of sequencing errors, were present. COI, 16S, H3, and 18S were aligned 44 easily due to a lack of insertions and deletions (indels), while indels occurred throughout the ITS1 and ITS2 regions. GenBank “blast” searches (Benson et al. 2005) were performed on the earliest obtained sequences for all genes, in order to check for contamination, verify that the target gene had been amplified, and inspect how genetically similar Fiona pinnata samples were to other nudibranchs for which molecular data were available. All DNA sequences generated from the Fiona pinnata specimens used in this study will be deposited in the NCBI GenBank database.

3.3.3 Analysis of global mitochondrial DNA variation Pairwise COI sequence divergences among Fiona pinnata samples were calculated using the Kimura (1980) 2-parameter model (K2P) of sequence evolution. TCS 1.21 (Clement et al. 2000) was used to identify all unique COI haplotypes, and a haplotype network was subsequently constructed based on complete sequences (i.e., 988-bp fragments). Six sequences were excluded from TCS network analysis due to missing data (i.e., < 988-bp fragments), which would have skewed statistical parsimony analysis because all variation outside of the incomplete sequence data available for these samples would have been ignored. These six sequences included samples from Madeira Island, the Galapagos, the French Frigate Shoals, and both islands of New Zealand. Phylogenetic COI analyses consisted of maximum likelihood (ML) and Bayesian inference methods. The Akaike information criterion (AIC) (Akaike 1974) implemented in jModelTest v.0.1.1 (Guindon and Gascuel 2003; Posada and Buckley 2004; Posada 2008) determined the best-fit model for COI to be HKY+I (Hasegawa et al. 1985). COI sequences obtained from GenBank for two members of the Fionoidea (Cuthona ocellata, AY345043; Tergipes tergipes, AY345032) were included as outgroups. MEGA was used to construct an ML tree in which branch support was evaluated with 1000 bootstrap replicates (Felsenstein 1985). Mr. Bayes v.3.1.2 (Huelsenbeck and Ronquist 2001; Ronquist and Huelsenbeck 2003) was used to perform Bayesian analysis on COI data (settings: nst=2, rates=inv), in which the Markov chain Monte Carlo search (Geyer 1991) ran for twenty million generations and trees were sampled every 100 generations. The first 10,000 trees (i.e., 1,000,000 generations) were discarded as burn-in, and the remaining trees were used to estimate Bayesian posterior probability (PP) values. Incomplete COI sequences were included 45 in all model-based analyses, as the incorporation of samples with missing data is not only preferable to excluding them, but presumably improves overall phylogenetic resolution (Philippe et al. 2004; Wiens 2006). A hierarchical approach was used to test for genetic partitioning among separate ocean basins. An analysis of molecular variance (AMOVA) (Excoffier et al. 1992) was performed in ARLEQUIN v.3.1 (Excoffier et al. 2005) in order to examine the large-scale distribution of variation based on genetic distances among mitochondrial haplotypes (Kimura 1980).

3.3.4 Analysis of global nuclear DNA variation Pairwise ITS1 sequence divergences among Fiona pinnata samples were calculated using the K2P model of sequence evolution. Unique ITS1 haplotypes were identified using TCS 1.21 (Clement et al. 2000), and a haplotype network was subsequently constructed after the removal of redundant sequences. Comparisons of network analyses in which all sequence data was examined (i.e., alignment gaps were treated as a 5th base) versus analyses in which gaps were ignored revealed only minor differences, and thus alignment gaps were excluded from analysis in order to produce the following ITS haplotype networks. The model selected for ITS1 phylogenetic analysis, HKY+I+G (Hasegawa et al. 1985), was determined using the AIC implemented in jModelTest v.0.1.1. Owing to a lack of published nudibranch ITS1 data, the aeolid nudibranch Flabellina verrucosa (AB180834) was used as the outgroup but only approximately 100-bp of this outgroup sequence was alignable with Fiona pinnata ITS1 fragments. An ML tree was constructed in MEGA, in which 1000 boostrap replicates were performed. Mr. Bayes v.3.1.2 was used to carry out Bayesian analysis (settings: nst=2, rates=invgamma), which ran for twenty million generations. Trees were sampled every 100 generations, and the first 10,000 trees (i.e., 1,000,000 generations) were discarded as burn-in. ITS2 contained relatively little variation and thus a haplotype network was produced in TCS 1.21 but no other phylogenetic analysis was performed on this nuclear region.

3.3.5 Phylogenetic analysis of the family Fionidae 46

In order to examine the higher taxonomic standing of Fiona pinnata in a phylogenetic context, several nudibranchs (Baeolidia nodosa, HQ616769; Cuthona ocellata, AY345043; , HM162734; rustyus, GQ292065; minor, DQ417310; Tergipes tergipes, AY345032) and an Aplysiidae sea hare (Aplysia juliana, AF343425) were used as outgroups in order to construct deep- rooted molecular phylogenies for a combined analysis of several mitochondrial and nuclear genes. The Eubranchidae and Tergipedidae are members of the superfamily Fionoidea, and some authors (Gosliner et al. 2008) have suggested that the latter family should include Fiona pinnata. A sea hare of the genus Aplysia was used to root all trees; although not a nudibranch, this taxon represents an ancestral euthyneuran that is basal to all other species examined herein (Wade et al. 2006). All outgroup sequence data was drawn from GenBank. MEGA was used to calculate K2P divergences among DNA sequence data and to construct ML trees for both COI and 16S, and branch support was evaluated with 1000 bootstrap replicates. The model HKY+I was selected for both of these mitochondrial genes using the AIC in jModelTest v.0.1.1. Neighbor-joining (NJ) bootstrapped consensus trees (1000 replicates) based upon the K2P model were also constructed for comparison. Nearly zero variation was detected in H3 and 18S, due to their more conservative nature. Accordingly, neighbor-joining trees based upon the K2P model (1000 bootstrap replicates) were constructed in MEGA but no other phylogenetic analyses of these nuclear genes were undertaken.

47

Table 3.1 Collection details for Fiona pinnata samples collected worldwide. Museum codes: CASIZ=California Academy of Sciences, MMF= Museu de Historia Natural do Funchal, AM=Australian Museum, UF=Florida Museum of Natural History, AK= Auckland War Memorial Museum, ZSM=Zoologische Staatssammlung Munchen.

Collection locality Collection Voucher Raft substrate date Zavora Beach, 14/3/2012 AM C.476792 Beach-cast driftwood Mozambique Dunedin, Otago, 2/2009- AM C.476807 Floating and beach-cast kelp New Zealand 5/2012 - 476827 (Durvillaea & Macrocystis) Ohui Beach, Opoutere, 1/2/2008 AK118314 Beach-cast storm debris New Zealand Cape Rodney, Goat 20/2/2012 AM C.476793 Floating chilly bin lid Island, New Zealand Cylinder Beach, North 13/6/2003 AM C.477132 Beach-cast cuttlebone Stradbroke Island, Australia Heron Island, 24/8/1999 AM C.477130 Floating plastic chair Australia Heron Island, 2/8/2003 AM C.477131 N/A (collected as veligers) Australia SE Sulawesi, 22/8/2002 AM C.477133 Rompong (deep-water Indonesia FAD) Wakayama, 8/6/2012 AM C.476794 Beach-cast bamboo shoot Shirahama, Japan North Pacific Gyre 15/6/2012 AM C.476795 Floating plastic cup (32° 03' N, 153° 10' E) French Frigate Shoals, 21/10/200 UF415673, Floating plastic debris Hawaii 6 UF427012 Georges Island, Cross 6/7/1993 CASIZ 088586 Not recorded Sound, Alaska Santa Cruz Island, 27/7/2003 CASIZ 172044 Floating buoy Galapagos Coquimbo, Chile 1/2012 AM C.476797 Floating kelp (Durvillaea & - 476806 Macrocystis) Roca Mar, Madeira 11/9/1996 MMF 29835 Floating buoy Island, Portugal Ilha de Pico, Azores, 29/5/2012 ZSM Floating bucket Portugal Mol20120145

48

Table 3.2 Primers used in PCR amplification of mitochondrial and nuclear genes.

Region Primer name Direction Primer sequence (5’-3’) Reference COI LCO1490 forward GGTCAACAAATCATAAAGAT ATTGG Folmer et HCO2198 reverse TAAACTTCAGGGTGACCAAA al. (1994) AAATCA H7005 reverse CCGGATCCACNACRTARTAN Hafner et al. GTRTCRTG (1994) CoxAF forward CWAATCAYAAAGATATTGG AAC Colgan et CoxAR reverse AATATAWACTTCWGGGTGA al. (2003) CC Mtd6 forward GGAGGATTTGGAAATTGATT Simon et al. AGTTCC (1994) ITS1 18S forward TAACAAGGTTTCCGTATGTG Armbruster AA et al. (2002) 5.8S reverse GCGTTCTTCATCGATGC ITS2 LSU-1 forward CTAGCTGCGAGAATTAATGT Wade et al. GA (2006) LSU-3 reverse ACTTTCCCTCACGGTACTTG Fiona_ITS2_F forward GTCTGAGGGTCGGACGATAC This study Fiona_ITS2_R reverse ATCCCGGTTGGTTTCTTTTC 16S 16Sar forward CGCCTGTTTAACAAAAACAT Palumbi et 16Sbr reverse CCGGTCTGAACTCAGATCAC al. (1996) GT H3 H3aF forward ATGGCTCGTACCAAGCAGAC VGC Colgan et H3aR reverse ATATCCTTRGGCATRATRGT al. (1998) GAC 18S 18S1F forward TACCTGGTTGATCCTGCCAG Okusu et al. TAG (2003) 18Sbi reverse GAGTCTCGTTCGTTATCGGA Whiting et al. (1997)

3.4 RESULTS

3.4.1 Global mitochondrial DNA variation Eighty-four Fiona pinnata COI sequences, distinguished by 178 variable nucleotide positions, yielded fifty-three distinct haplotypes worldwide. Overall K2P divergences ranged from 0.1% to 12.8% (Table 3.3), and all Fiona pinnata sequences 49

Figure 3.1 Global sampling map and geographic distribution of Fiona pinnata clades based on COI data. COI clade symbols are taken from Fig. 3.2 and a highly simplified representation of the overall phylogenetic structure is included in the lower right corner. The sampling locations and sample sizes of individuals sequenced for COI are as follows: 1-Mozambique (n=1), 2-Dunedin (n=39), 3-Opoutere (n=2), 4- Goat Island (n=3), 5-Stradbroke Island (n=2), 6-Heron Island (n=1), 7-SE Sulawesi (n=1), 8-Japan (n=3), 9-North Pacific Gyre (n=2), 10-French Frigate Shoals (n=2), 11-Alaska (n=1), 12-Galapagos (n=1), 13-Coquimbo (n=20), 14-Madeira Island (n=1), 15-Azores (n=5).

were highly distinct from those of the Tergipedidae outgroups (mean 19.7-21.2% divergence). ML analysis and Bayesian inference produced similar tree topologies, and both analyses were congruent in the recovery of three distinct clades (Fig. 3.2) and gave strong support to the monophyly of all ingroup taxa (ML: 100%; Bayesian PP: 1.00). Most samples clustered into two major clades (mean 6.1% K2P divergence), and the node connecting these two groups (herein called clades A and B) was reasonably well supported (ML: 78%; Bayesian PP: 0.87). The only sampling site in which individuals from both clades A and B were found in sympatry was the South Island of New Zealand (Fig. 3.1). The single Indonesian specimen formed a third lineage that was considerably divergent from all other ingroup taxa (mean 11.8%). 50

Clade A was comprised of twenty-eight haplotypes from the temperate zones in both hemispheres of the Pacific Ocean (Chile, the South Island of New Zealand, and Alaska), and received very high bootstrap support from ML analysis (96%). The Bayesian posterior probability at this node was similarly very high (1.00). The relationships among these temperate samples showed some slight structuring between New Zealand and Chilean populations (see Chapter 2), although the overall phylogeographic structure within this clade was quite shallow (mean 1.03% K2P divergence). Additionally, one haplotype was present in samples from both Chile and the South Island of New Zealand. Clade B included specimens from the tropical and sub-tropical regions of the Pacific, Indian, and Atlantic Oceans, and did not receive high bootstrap support from ML analysis (41%) or Bayesian inference (0.38). Phylogeographic structure within this clade was very shallow (mean 0.83% K2P divergence) and the relationships among taxa were poorly resolved, with no internal groupings receiving high node support from either of the phylogenetic analyses. Although most of the Portuguese specimens collected in Madeira Island and the Azores clustered together, this grouping received weak bootstrap support (ML 56%; Bayesian PP 0.48). Furthermore, one haplotype was shared between the northeast Atlantic and southwest Pacific Ocean. Clade C was represented by a single individual collected in Indonesia. This specimen showed substantial genetic distinction from clade A (mean 12.4% divergence) and clade B (mean 11.1% divergence). The COI haplotype network (Fig. 3.3) also supports the existence of three clades, as it shows three major groupings of specimens that are separated by high numbers of mutational steps. Within clade A, the Alaskan specimen grouped closely with the Chilean haplotypes, and the Chilean samples contained more genetic diversity (sixteen haplotypes) than the samples collected in the South Island of New Zealand (twelve haplotypes). One haplotype was shared between individuals collected in Chile and the South Island of New Zealand, and intra-clade genetic variation was regionally partitioned between these two coastlines. The most frequently encountered haplotype in the South Island of New Zealand was sampled in ~62% of the individuals collected there. Clade B was genetically more heterogenous than clade A, and identical haplotypes were generally not shared among individuals. The exception 51 to this was a haplotype that was sampled in separate ocean basins, as it was found in one Portuguese specimen from the Azores as well as two individuals collected in the North Island of New Zealand. Hierarchical analysis of COI diversity using AMOVA revealed that only 22% of the total variation occurred between ocean basins, while the majority (61%) was distributed among regional sampling sites within oceans (Table 3.4). Genetic variation within sampling sites accounted for the remaining 17% of worldwide variation. All three levels of molecular variance partitioning were significant (P <

0.03), and the global FST value was 0.83187 (P = 0.000).

Table 3.3 Minimum and maximum percent divergences of Fiona pinnata COI sequences calculated using the Kimura (1980) 2-parameter model of sequence evolution. Intra-clade divergence values are italicized (clade C is represented by a single specimen collected in Indonesia). Clade A Clade B Clade C Clade A 0.1 - 2.0% Clade B 4.9 - 7.1% 0.1 - 1.9% Clade C 11.9 - 12.7% 10.2 - 12.8% 0% (n=1)

52

Figure 3.2 Maximum likelihood bootstrap consensus tree of Fiona pinnata haplotypes collected worldwide based on COI data. Bootstrap support values above branches are derived from ML analysis (1000 replicates), and values below branches are Bayesian posterior probabilities. The numbers in parentheses following some haplotypes indicate frequency of haplotype occurrence if greater than one.

53

Figure 3.3 Unrooted statistical parsimony networks for Fiona pinnata. Each colored circle/pie represents a unique COI haplotype and sizes are drawn in proportion to relative haplotype frequencies. Small empty circles represent intermediate hypothesized haplotypes. Haplotypes are colored according to their geographic origin. High numbers of mutational steps between the clades are excluded for clarity. 54

Table 3.4 Analysis of molecular variance (AMOVA) of Fiona pinnata global populations partitioned among ocean basins. Genetic distances between haplotypes were calculated according to the Kimura (1980) 2-parameter model.

Sum of Percent of Source of variation d.f. P value Fixation indices squares variation Among oceans 2 116.641 22.34 0.028348 Fct = 0.22345 Among locations, 10 521.855 60.84 0.00000 Fsc = 0.78349 within oceans Within locations 71 183.553 16.81 0.00000 Fst = 0.83187

3.4.2 Global nuclear DNA variation Twenty-six Fiona pinnata ITS1 sequences resulted in twenty-two unique haplotypes worldwide. These haplotypes were distinguished by fifty-three variable nucleotide positions, and indels also occurred throughout this nuclear region. Overall sequence divergences ranged from 0.2% to 8.5%, and the Indonesian specimen again showed marked diversity (5.7-8.5%, mean 7.4% divergence) and was separated from all other ingroup taxa by a long branch length. The topologies resulting from ML and Bayesian analyses were similar and thus only the ML tree is presented (Fig. 3.4). All other Fiona pinnata ITS1 haplotypes formed a monophyletic group (ML: 96%; Bayesian PP: 0.52), and a distinction between temperate and tropical haplotypes was apparent in the resulting phylogeographic structure. The grouping of Alaskan, Chilean, and South Island New Zealand specimens received moderate bootstrap support (ML: 75%; PP: 0.84) while all tropical/sub-tropical samples clustered together (ML: 70%; PP: 0.72). Thus, the phylogenetic relationships based on nuclear ITS1 were broadly congruent with mitochondrial COI data, although nuclear differentiation was much shallower (0.2-3.0%, mean 1.6% K2P divergence) and some inconsistencies were detected: three specimens collected in the South Island of New Zealand grouped closely with tropical specimens in COI analysis, but ITS1 analysis placed them within the temperate group (the placement of these three individuals is denoted in Fig. 3.4 with COI clade symbols taken from Fig. 3.2). 55

The statistical parsimony network of global ITS1 haplotypes (Fig. 3.5) found that connections generally occurred between haplotypes sampled within broad geographic regions (i.e., ocean basins), although the northeast Atlantic and southwest Indian Ocean specimens grouped closely with Indo-Pacific samples. Two haplotypes were shared between Chile and the South Island of New Zealand, and one of these was also present in the Alaskan individual. This network analysis also highlights the considerable divergence between the Indonesian specimen and all other samples of Fiona pinnata. The ITS2 region contained relatively little diversity overall, containing twelve variable nucleotide positions, and the highly divergent SE Sulawesi specimen accounted for most of the observed variation (2.2-3.2% divergence). All other ingroup taxa showed rather shallow levels of divergence (0.3-1.3%; mean 0.7%). As shown in the haplotype network (Fig. 3.6), eight unique haplotypes were found, and three of these were shared across broad geographic ranges. The most common haplotype, present in ~54% of the total specimens sequenced for ITS2, was found in New Zealand, Chile, and Eastern Australia. A second haplotype was shared between the French Frigate Shoals and Chile, while another widespread haplotype was present in the Pacific (Eastern Australia), Atlantic (Madeira Island), and Indian (Mozambique) Oceans.

56

Figure 3.4 Maximum likelihood bootstrap consensus tree of Fiona pinnata haplotypes collected worldwide based on ITS1 data. Bootstrap support values above branches are derived from ML analysis (1000 replicates), and values below branches are Bayesian posterior probabilities. The numbers in parentheses following some haplotypes indicate frequency of haplotype occurrence if greater than one. The Indonesian specimen was separated from all other ingroup taxa by a long branch length (top), and thus the subtree containing all other haplotypes is shown separately for clarity (bottom). K2P divergence values associated with major phylogenetic groupings are indicated by arrows. The circles/pies shown along the subtree are COI clade symbols are taken from Fig. 3.2, wherein white represents temperate mitochondrial haplotypes and black corresponds to tropical mitochondrial haplotypes.

57

Figure 3.5 Unrooted statistical parsimony network for Fiona pinnata. Each colored circle/pie represents a unique ITS1 haplotype and sizes are drawn in proportion to relative haplotype frequencies. Small empty circles represent intermediate hypothesized haplotypes. Haplotypes are colored according to their geographic origin.

58

Figure 3.6 Unrooted statistical parsimony network for Fiona pinnata. Each colored circle/pie represents a unique ITS2 haplotype and sizes are drawn in proportion to relative haplotype frequencies. Small empty circles represent intermediate hypothesized haplotypes. Haplotypes are colored according to their geographic origin.

3.4.3 Phylogenetic analysis of the family Fionidae A deep-rooted tree based on COI data was constructed in order to assess the phylogenetic relationships of Fiona pinnata at higher taxonomic levels (Fig. 3.7). ML and NJ methods produced extremely similar tree topologies, and all Fiona pinnata specimens formed a monophyletic group that received very strong bootstrap confidence levels in both analyses (ML: 98%; NJ: 100%). Overall ingroup-outgroup sequence divergences ranged between 17.4% and 29.1%, and of all the outgroup sequences Cuthona ocellata displayed the highest level of genetic similarity to the sampled Fiona pinnata specimens (mean 19.7% divergence). These COI findings indicate that Fiona pinnata is substantially divergent from even its closest relatives in 59

Figure 3.7 Maximum likelihood bootstrap consensus tree of Fiona pinnata haplotypes collected worldwide based on COI sequence data. Bootstrap support values above branches are derived from ML analysis (1000 replicates), and values below branches are neighbour joining confidence values based on the K2P model (1000 iterations). The numbers in parentheses following some haplotypes indicate frequency of haplotype occurrence if greater than one.

60 the superfamily Fionoidea, with a weakly supported bootstrap value occurring at the node connecting Fiona pinnata and the Fionoidea outgroups (ML: 25%; NJ: 31%). For mitochondrial 16S rRNA, the alignment of thirty-seven sequences resulted in six unique haplotypes. Only five sites in the 421-bp fragments were variable within the Fiona pinnata samples, and thus 16S haplotypes showed low overall divergence (0.2-0.7%; mean 0.4%). Haplotypes were commonly shared within and between regions, and the most common haplotype, shared by thirty of the thirty-seven total specimens included in analysis, was present in samples from all three oceans. ML and NJ analyses produced congruent tree topologies and yielded maximum bootstrap support (100%) for the monophyly of all Fiona pinnata haplotypes (Fig. 3.8). Ingroup/outgroup sequence divergences for 16S ranged between 10.0% and 37.8%. Fiona pinnata clustered with the four other Fionoidea members (three Tergipedidae sequences, one Eubranchidae sequence) and this group received strong bootstrap support in both ML (94%) and NJ (99%) analyses. The taxon to group most closely with Fiona pinnata sequences was (10.0-10.3% divergence), although node support for this grouping was much weaker in ML (59%) than NJ (85%) analysis. The alignment of seven nuclear H3 fragments, obtained from a subset of Fiona pinnata specimens collected in Alaska (n=1), New Zealand (n=3), Mozambique (n=1), the French Frigate Shoals (n=1), and SE Sulawesi (n=1), yielded two distinct haplotypes. Only a single nucleotide position out of the 316-bp fragments was variable between the SE Sulawesi specimen and all other samples, resulting in very low overall sequence divergence (maximum 0.3%). NJ analysis based on the K2P model yielded 100% bootstrap support for the monophyly of these ingroup sequences (Fig. 3.9), although the outgroups used are not close relatives of Fiona pinnata due to the lack of published H3 sequences for Fionoidea members. The aeolid Baeolidia nodosa was the closest genetic match for Fiona pinnata H3 sequences (bootstrap 85%), and overall ingroup/outgroup divergences ranged between 12.1% and 29.1%. Zero variation was observed in the 1407-bp fragments of nuclear 18S obtained for a subset of seven Fiona pinnata individuals collected in Alaska (n=1), New Zealand (n=4), Madeira Island (n=1), and Mozambique (n=1). NJ analysis based on the K2P model gave 100% support to the grouping of Fiona pinnata with the three Fionoidea

61

Figure 3.8 Maximum likelihood bootstrap consensus tree of Fiona pinnata haplotypes collected worldwide based on 16S data. Bootstrap support values above branches are derived from ML analysis (1000 replicates), and values below branches are neighbour joining confidence values based on the K2P model (1000 iterations). The numbers in parentheses following some haplotypes indicate frequency of haplotype occurrence if greater than one.

Figure 3.9 Neighbor joining bootstrapped consensus tree (1000 iterations) based on the K2P model of Fiona pinnata histone H3 haplotypes sampled worldwide.

Figure 3.10 Neighbor joining bootstrapped consensus tree (1000 iterations) based on the K2P model of the only 18S haplotype sampled in Fiona pinnata populations worldwide. 62 outgroups (1.6-4.7% divergence), and of these was the closest genetic match for the solitary Fiona pinnata sequence (Fig. 3.10).

3.5 DISCUSSION

3.5.1 Global phylogeography and evolutionary history Three major clades were identified from phylogenetic analysis of COI sequence data. Both clades A and B were abundant and widespread, showing a prominent break in geographic distribution along temperate and tropical/sub-tropical zones. In contrast, clade C was represented by a single specimen from Indonesia and consequently represents a comparatively rare localized lineage. Clade A individuals were found on macroalgals rafts throughout the temperate regions of the Southern Hemisphere as well as in Alaska, thus demonstrating an anti-tropical distribution in the Pacific Ocean (Hubbs 1952; Briggs 1977). Anti-tropical distributions are commonly observed in the sea (for a review see Briggs 1995) and are exhibited by a diverse range of marine taxa including fish (Bowen and Grant 1997; Burridge and White 2000; Grant et al. 2005), algae (Peters and Breeman 1992), (Nations 1979), molluscs (Koufopanou et al. 1999; Hilbish et al. 2000), and seals (Fyler et al. 2005). Within the Southern Hemisphere, genetic variation was regionally partitioned among Chilean and New Zealand populations but this temperate clade nonetheless exhibited shallow overall phylogeographic structure (maximum 2% divergence) and a common haplotype was shared between these two landmasses. It thus appears that trans-oceanic gene flow between these two distant coastlines via passive rafting has been facilitated by the West Wind Drift (see Chapter 2). This strong surface current has previously been demonstrated to function as a powerful dispersal agent in the Southern Ocean for a variety of buoyant seaweeds and their associated faunal holdfast communities (Helmuth et al. 1984; Smith 2002; Waters 2008; Nikula et al. 2010). Clade B individuals occurred worldwide throughout tropical and sub-tropical seas, thereby exhibiting a circum-equatorial distribution, and were generally found on driftwood or plastic debris. Marine invertebrates generally occur exclusively in either temperate or tropical zones (Hyman 1955), and the geographic distributions of the two widespread clades of Fiona pinnata were indeed correlated with prominent 63 biogeographic boundaries. However, this nudibranch’s dispersal patterns appear to be largely current-mediated rather than influenced strictly by latitude alone. This point is illustrated by the fact that the Chilean specimens were collected at 29°S but grouped firmly with the Alaskan and South Island NZ samples due to the temperate influence exerted by the northward flowing Humboldt Current and coastal upwelling events in the Coquimbo region (Brattstrom and Johanssen 1983; Meneses and Santelices 2000). In comparison, sampling sites in the North Island of New Zealand lay at 36°S in a region characterized by the warm south-flowing waters of the East Auckland Current. The Fiona pinnata specimens collected here showed close phylogenetic relationships with tropical/sub-tropical samples obtained at low latitudes, which is perhaps unsurprising given that some biota in the North Island of New Zealand are known to display Indo-Pacific affinities (Pawson 1961). Despite the fact that the Atlantic and Pacific Ocean basins are semi-closed systems and inter-oceanic exchange between them is believed to occur relatively rarely among marine invertebrates (Hyman 1955), an AMOVA showed that only 22% of the overall genetic variation observed in global Fiona pinnata populations was partitioned between oceans. In addition, one COI haplotype was shared between individuals sampled in New Zealand and the Azores. Although anthropogenic translocation cannot be unequivocally discarded as a possibility, there are currently no known records of Fiona pinnata occurring on ship hulls colonized by Lepas spp. barnacles. Instead, this nudibranch is continuously encountered on floating seaweeds and other buoyant substrates (see Thiel and Gutow 2005b for a review), and thus passive rafting (in combination with a planktonic larval stage) is more likely to be the dominant dispersal process that has shaped this invertebrate’s widespread distribution patterns. As no molecular clock rate estimates exist specifically for opisthobranchs, many nudibranch studies employ a generalized molluscan clock in order to estimate divergence times. Proposed COI divergence rates for molluscs generally range from 0.4% to 3.1% per million years (MY) (Hellberg and Vacquier 1999; Marko 2002; Frey and Vermeij 2008; Lessios 2008). For the present study, a rate of 1%/MY was used to roughly date divergence times based on COI data, following the example of previous work on nudibranch phylogenetics (Marko 2002; Frey and Vermeij 2008; Shields 2010). This relatively conservative rate is also consistent with findings that protein-coding mitochondrial DNA evolves at a rate of 0.5-1.5%/MY (Ho et al. 64

2005). However, caution must be used when inferring divergence times from a single molecular marker because timelines may be confounded by the presence of an undetermined level of ancestral polymorphism (Edwards and Beerli 2000; Hare et al. 2002). For this reason, when interpreting divergence times based on COI data the minimum time since divergence is likely to be more accurate than the calculated maximum time (Schwaninger 2008). The relatively deep COI divergences detected between the Indonesian specimen and all other Fiona pinnata haplotypes (10.2-12.8%) dated this divergence event to 10.2-12.8 million years ago, roughly corresponding to the mid-Miocene. In contrast, the COI divergences between the two major clades of Fiona pinnata (4.9-7.1%) imply that this temperate-tropical split may have occurred as recently as 4.9 million years ago, during the early Pliocene. The Miocene (approximately 22-5 MYA) was characterized by relatively warm oceans that were subject to ongoing widespread cooling at mid– and high-latitudes as well as glacial events as a result of the expansion of the Antarctic ice sheet (Mercer and Sutter 1982; Hodell and Kennett 1986; Woodruff and Savin 1989). This gradual global cooling of the sea surfaces continued into the Pliocene (approximately 5-2.5 MYA), and was potentially aided by the formation of the Isthmus of Panama approximately 3.1 MYA, which severed the last trace of a circum-equatorial current (Keigwin 1982; Duque-Caro 1990; Coates and Obando 1996). The abundance and broad distributions of the two major clades detected in COI analysis reflect the high dispersal potential of Fiona pinnata and also confound a definite interpretation of this animal’s speciation history. Timing of divergence events also involves an inherent amount of speculation and thus makes it difficult to pinpoint the historical processes that led to the divergence of lineages within Fiona pinnata. Consequently, the present data cannot provide an unequivocal account for this nudibranch’s evolutionary history and the following scenario is merely suggested as one possibility. A hypothetical evolutionary scenario for Fiona pinnata entails a formerly circum-tropical population distributed throughout the world’s warm ancient seas, which gradually diverged in concordance with widespread cooling of the oceans and glacial events during the Miocene-Pliocene. Sympatric speciation of the three major lineages cannot be ruled out completely (Higashi et al. 1999; Via 2001; Munday et al. 2004), but cryptic divergence is generally believed to occur in allopatry 65

(Barraclough and Vogler 2000; Meyer 2003; Williams and Reid 2004). Accordingly, it is possible that historical bottlenecking due to ice age dynamics and sea surface cooling resulted in a break within the previously continuous distribution of Fiona pinnata. While the ancestral warm-water population may have continued to prevail within the equatorial region, an isolated lineage may have radiated into the temperate Southern Hemisphere and subsequently diversified into a similarly widespread West Wind Drift population. In essence, clade B may represent the tropical parent population, while clade A may be the result of a founder event within southern temperate zones. Subsequent divergence may have been driven in accordance with changing oceanic circulation patterns (Vermeij 1987). If Fiona pinnata did originate in the tropics, clades B and C have maintained the warm-water temperature affinities of the ancestral state, while clade A has adapted to the cooler climates of higher latitudes and thus resulted in a range expansion. The Indonesian specimen could possibly represent a more benthic ancestral state of Fiona pinnata that has simply maintained ancestral polymorphisms, while the pelagic circum-tropical and anti- tropical clades have undergone extensive divergence over the course of adapting a rafting life strategy. On the other hand, the Indonesian haplotype may represent a highly derived lineage undergoing local adaptation in the oceanographically isolated Indo-Pacific. Interestingly, the rompong (deep-water fish attracting device) on which this specimen was found had been placed in the water only two weeks prior to its collection at 20 m depth (N.G. Wilson, pers. comm.), which shows remarkable colonization by the barnacles and nudibranchs encountered during sampling in SE Sulawesi. These pelagic invertebrate communities have clearly evolved to capitalize on the exceptionally ephemeral habitats on which they carry out their life cycles. The finding of this specimen at 20 m depth rather than rafting at the sea surface is unusual, but nothing else is known of the life history of Fiona nudibranchs in this Indo-Pacific region. The anti-tropical distribution observed for clade A is based on the inclusion of the Alaskan specimen with the grouping of samples from Chile and New Zealand. Given the close relationship between this North Pacific haplotype with samples from the temperate Southern Hemisphere, it appears that clade A’s disjunction across the tropics is reflective of recent trans-equatorial dispersal, rather than a vicariance event. Although a “relict theory” in which formerly widespread species become extinct in 66 the tropics has been used to explain how anti-tropical distributions could arise (Briggs 1987), dispersal is often considered the more plausible mechanism is establishing North-South disjunctions across the tropics (Vermeij 1989; Lindberg 1991; Peters and Breeman; Hilbish et al. 2000). Within clade A, the Alaskan haplotype groups more closely with the Chilean haplotypes (0.6-1.8% divergence) than the New Zealand samples (1.8-2.0% divergence), which suggests that this Alaskan lineage dispersed northward from the southeast Pacific by crossing the tropics and reaching the higher latitudes of the northeast Pacific. Application of the 1%/MY COI molecular clock dates this trans-equatorial dispersal to 0.6-1.8 million years ago, possibly corresponding to a cooling trend in the tropics during the Pleistocene (Burridge 2002). The distribution of temperate and tropical water masses has altered radically over the past million years (Savin et al. 1975; Dansgaard et al. 1993; Graves 1998), so there have certainly been historical opportunities for hemispheric exchange of biota and North-South gene flow. The tropics generally represent a strong biogeographic barrier (Stepien and Rosenblatt 1996), but a proposed dispersal route between the hemispheres along the eastern rim of the Pacific is supported by the upwelling of cold water along these coastlines, which acts to diminish the width of the tropical belt and thereby serve as a series of stepping stones in the migrations of temperate organisms (Lindberg 1991; Briggs 1995). Horta (2010) similarly suggested a Southern Hemisphere origin for the widespread kelp Macrocystis pyrifera, which subsequently dispersed into the Northern Hemisphere via trans-tropical migration of floating sporophytes. This hypothesis is supported by the observation that macroalgal rafts in the southeastern Pacific are generally carried north along the Chilean coastline by way of the Humboldt Current (Macaya et al. 2005). Fiona pinnata is frequently observed rafting on detached Macrocystis pyrifera plants within the eastern Pacific Ocean (Bushing 1994; Hobday 2000a), and thus the anti-tropical pattern observed within the temperate lineages of this nudibranch is probably linked to the colonization history and geographic distribution of its macroalgal hosts. Despite the fact that clade C is represented by only a single specimen, it clearly shows marked genetic differentiation from all other ingroup sequences. The deep COI sequence divergence detected between the Indonesian specimen and all other Fiona pinnata samples (10.2-12.8%) was corroborated with nuclear data from the ITS1 (5.7-8.5% divergence) and ITS2 regions (2.2-3.2% divergence), which suggests 67 that this Indo-Pacific lineage represents a distinct biological species (Moritz 1994; Morando et al. 2003). Furthermore, this individual was collected at 20 m depth on a rompong (deep-water fish attraction device), a somewhat unusual habitat for an animal that generally occurs on floating objects at the sea surface. Although this Indonesian specimen does not group into either of the two major clades of Fiona pinnata, phylogenetic analyses of several mitochondrial and nuclear genes yielded extremely high support for its inclusion in monophyletic groupings with all other Fiona samples. Furthermore, almost zero variation was detected within sequence data generated from nuclear H3, which is known to be a conservative gene within opisthobranchs (Dinapoli et al. 2006). These results suggest that while the Indonesian specimen represents a highly divergent lineage, it nonetheless appears to be a congeneric species that is sister to all other sampled Fiona pinnata individuals. In an attempt to establish genetic divergence thresholds among sister species of molluscs, some authors have proposed that a mean COI distance of ~10% be used as a species delimitation approach (Hebert et al. 2003; Malaquias and Reid 2009). However, these sorts of threshold references are confounded due to the high variance in rates of molecular evolution among even closely related taxa, and thus barcoding methods cannot unequivocally identify cryptic species. Indeed, reviews of the literature have uncovered COI distances as low as 3.3-5.1% for well-supported sister species of nudibranchs (Pola et al. 2007; Turner and Wilson 2008). With this in mind, it is reasonable to suggest that the three clades detected within Fiona pinnata may very well represent a cryptic species complex. In addition to the levels of sequence divergence detected between the clades, the well-defined geographic partitioning of these major lineages lends further support to the notion that the observed genetic variation is shaped by oceanic current dynamics and also that long- distance dispersal in this nudibranch is achieved by a planktotrophic larval stage and passive rafting. Thus, “Fiona pinnata” appears to be composed of at least three diverse lineages and the cosmopolitan distribution attributed to this invertebrate is in fact misleading. Although clades A and B appeared mostly allopatric, exhibiting anti-tropical and circum-tropical distributions respectively, they were found in sympatry in the Dunedin region of New Zealand’s South Island. Furthermore, individuals from divergent mitochondrial lineages co-occurred not only in the general region but on the 68 same rafting substrates (i.e., on the same kelp holdfasts), although they were present in uneven abundances (specimens belonging to the tropical clade accounted for only ~8% of the total individuals sampled in Dunedin). However, these discrepancies were only observed in COI data and were not reflected in the nuclear ITS region, which may suggest that this is due to incomplete lineage sorting in the slower- evolving nuclear region. It is alternatively possible that southern New Zealand represents a hybridization zone, although if this is the case, it remains unknown whether the distributions of the two clades consistently overlap there or if tropical specimens only sporadically disperse there. A small degree of mitochondrial introgression within Dunedin populations could be maintained by a recent influx of tropical haplotypes, possibly as a result of climate change (van Oppen et al. 2001; Schroth et al. 2002; Depraz et al. 2009). Conversely, this evidence of hybridization could simply be the reflection of historical introgression. Knowlton (1993) suggested that a minimum of 3-3.5 million years without secondary contact might be necessary for the development of strong reproductive isolation between allopatric populations. The use of a molecular clock revealed that these two major clades of Fiona pinnata might have diverged as recently as 4.9 million years ago, and since complete gamete compatibility can persist between sister species for as long as five million years (Ziegler et al. 2005), it is possible that no reproductive barriers exist between sympatric Fiona individuals. Furthermore, given that nudibranchs are simultaneous hermaphrodites, all conspecifics presumably represent prospective mates and overlapping populations could potentially be considered panmictic (Thorpe and Sole- Cava 1994; Thollesson 1998). While all three Fiona clades were unambiguously differentiated by mitochondrial COI variation, phylogenetic relationships were not as well resolved within the slower- evolving nuclear ITS1 region. The Indonesian specimen showed pronounced divergence from all other sampled ITS1 haplotypes (minimum 5.7% divergence), and temperate specimens did cluster together, albeit in a moderately supported internal group, but the overall phylogeographic structure was much shallower in ITS1 than in COI. The slight incongruence between these two independent markers thus may be the result of incomplete lineage sorting within ITS1 (Kemppainen et al. 2009; Sandoval-Castellanos et al. 2010).

69

3.5.2 Systematic status of the family Fionidae Phylogenetic analyses of COI, 16S, H3, and 18S all yielded strong support for a monophyletic grouping of all sampled Fiona pinnata haplotypes. These independent mitochondrial and nuclear genes, which are subject to different rates of sequence evolution, all produced congruent topologies in which Fiona pinnata consistently grouped with fellow members of the superfamily Fionoidea, as expected. However, long branch lengths often separated Fiona pinnata samples from members of the Tergipedidae and Eubranchidae, and these nodes were typically weakly supported. Furthermore, relatively large sequence divergences separated Fiona pinnata from all outgroup taxa, which again indicates that this rafting nudibranch is highly divergent from even its closest relatives (see Fig. 3.11 for a simplified taxonomy of the outgroups used). Consequently, these molecular data do not support the inclusion of Fiona pinnata within the Tergipedidae, as proposed by Gosliner et al. (2008), and instead make the case for the retention of a separate classification. The systematic standing of the Fionidae, historically a monotypic family that was established to reflect the extensive morphological differentiation of this animal from all other known aeolid nudibranchs, has now been corroborated with molecular evidence that confirms the pronounced divergence of Fiona pinnata from all other related taxa sequenced thus far.

Kingdom: Animalia Phylum: Mollusca Class: Gastropoda (unranked): clade [Aplysia juliana] (unranked): clade [Doto coronata] (unranked): clade [Baeolida nodosa] Superfamily: FIONOIDEA Family: Calmidae Family: Eubranchidae [Eubranchus rustyus] Family: Fionidae [Fiona pinnata] Family: Pseudovermidae Family: Tergipedidae [Cuthona ocellata, , Tergipes tergipes]

Figure 3.11 A simplified taxonomy of Fiona pinnata. The outgroups used in phylogenetic analysis of the family Fionidae are shown in brackets.

70

CHAPTER 4 Summary and future research

4.1 Overall conclusions The increasing use of molecular tools to test marine biogeographic hypotheses has shown that the taxonomy of few seemingly widespread marine invertebrates holds up under molecular scrutiny. Instead, these “ubiquitous” organisms are often found to conceal substantial levels of molecular divergence, and are thus revealed to comprise cryptic species complexes (e.g., Knowlton 1993; Dawson and Jacobs 2001; Cox and Moore 2005; Blanquer and Uriz 2007). Phylogenetic analyses across a wide variety of marine taxa have revealed that, despite the lack of obvious barriers to gene flow, cryptic speciation is quite common in the sea and truly cosmopolitan marine invertebrates are now presumed to be quite rare. However, some species with seemingly limited dispersal potential have been demonstrated to show genetic homogeneity over very broad geographic ranges, possibly as a result of passive rafting (Waters and Roy 2004a; Horta 2010; Nikula 2010). The aeolid nudibranch Fiona pinnata (Eschscholtz, 1831) is considered a single cosmopolitan species, but the enormous distribution attributed to this invertebrate has been based entirely on the uniform morphological characters shared between isolated populations. Despite the fact that this aeolid nudibranch cannot swim or even float by itself, it has adapted a pelagic lifestyle by colonizing floating objects and thus apparently achieves long-distance dispersal by the rafting of egg masses and post- metamorphic individuals. Although Fiona pinnata has planktotrophic larvae and is prone to long-distance rafting, and thus exhibits very high dispersal ability, this study identified three divergent lineages within this “cosmopolitan” aeolid nudibranch. The geographic distributions of these genetic disjunctions appear to be in concordance with oceanic circulation regimes. One of the lineages detected in this study was restricted to the tropical Indo- Pacific, while the other two clades were widespread and abundant, showing a prominent split between temperate and tropical/sub-tropical zones. The circum- equatorial clade possibly represents the ancestral tropical state of Fiona pinnata, while the temperate clade may have arisen as a result of a founder event into the 71

Southern Hemisphere during the Miocene-Pliocene. Furthermore, the anti-tropical distribution observed within the temperate lineage suggests that this nudibranch has successfully undergone trans-equatorial migration relatively recently in its evolutionary history, and this dispersal event was likely due to Fiona pinnata‘s association with Macrocystis pyrifera kelp rafts. Long-distance dispersal in this nudibranch is apparently current-mediated, and although these two major clades appeared to be largely allopatric, they were found to co-occur in uneven abundances in the South Island of New Zealand. This apparent sympatry, in which individuals of differing mitochondrial clades were found together on the same kelp holdfast, was not corroborated in the nuclear markers. This suggests that mitochondrial introgression may have occurred, and it is therefore plausible that these two sister groups have not developed significant reproductive isolation and thus can either freely hybridize or did so in the past. Although these two major clades were most clearly differentiated by COI data, whereas the nuclear ITS1 region showed much shallower resolution, these two independent loci yielded relatively congruent phylogeographic structures. Despite the fact that phylogenetic relationships were not as well resolved in the nuclear ITS1 data, both markers yielded support for a distinction between temperate and tropical specimens. The observed levels of genetic divergence in combination with the geographic partitioning of these two widespread clades suggest that these lineages may represent separate species. The Indonesian specimen in particular repeatedly showed pronounced levels of divergence in all sampled genes (with the exception of the highly conservative, slow-evolving nuclear H3), which may indicate that it represents a reproductively isolated species of Fiona that has not yet been described. Overall, these molecular data indicate that Fiona pinnata populations are dispersed via the main oceanic currents and can travel across vast reaches of the open ocean, as demonstrated by the close phylogenetic relationships between trans-oceanic populations. An identical haplotype was shared between populations of Fiona pinnata sampled in New Zealand and Chile, and overall genetic differentiation between these trans-Pacific coastlines was very low. Previous molecular research has demonstrated that passive oceanic rafting is a powerful agent of genetic connectivity among populations of various marine invertebrates, particularly in the Southern Hemisphere where the West Wind Drift strongly facilitates long-distance dispersal 72

(Waters and Roy 2004a; Donald et al. 2005; Nikula et al. 2010; Fraser et al. 2011). The high abundance of robust floating seaweeds in the Southern Ocean (Smith 2002) further supports the notion that rafting of buoyant macroalgae and associated invertebrate communities presumably occurs very frequently in the higher latitudes. Fiona pinnata has historically been classified as the sole member of the monotypic Fionidae. This family was created to reflect the many peculiar morphological features displayed by this animal that make it unique among all aeolid nudibranchs (e.g., Alder and Hancock 1851). However, some authors have suggested that Fiona pinnata is closely related to the Tergipedidae and thus recommend a revision of this long- standing classification (Gosliner et al. 2008). In this study, the monophyly of all sampled Fiona specimens was recovered in phylogenetic analyses of several mitochondrial and nuclear genes. Long branch lengths often separated all Fiona haplotypes from this nudibranch’s presumed closest relatives in the superfamily Fionoidea, and the node values for these branches were weakly supported in the less conservative mitochondrial genes. This indicates that Fiona is quite genetically distinct from its closest living relatives and may not have a close sister group. Based on these molecular data, Fiona pinnata should not be absorbed into any other families in the Fionoidea, and the family Fionidae should maintain its systematic status. Fiona pinnata is commonly found in association with its Lepas spp. goose barnacle prey, and given that Lepas anatifera has been documented on ship hulls (Zvyagintsev and Mikhajlov 1985; Farrapeira et al. 2007), human-mediated dispersal cannot be entirely dismissed as a possibility. However, there are no reports of Fiona pinnata having been inadvertently transported via human translocation and thus anthropogenic introduction of this invertebebrate is not the most parsimonious explanation for its distribution patterns. In contrast, this invertebrate is known to have planktotrophic larvae and large aggregations of reproductive adults are frequently observed rafting on the high seas, so it is more likely that Fiona pinnata achieves long-distance dispersal via these natural mechanisms.

4.2 Future work and recommendations Morphological examination was outside the scope of this study but would be useful in order to explore whether the observed molecular divergences are coupled with morphological differentiation. Although many widespread marine invertebrates 73 appear to exhibit morphological uniformity throughout their vast geographic ranges, subsequent morphometric analysis upon the discovery of cryptic speciation often detects subtle yet discernible differences (Knowlton 1993; Fujita et al. 2012). An investigation of reproductive biology was not included in this study, so it remains unknown whether these clades are reproductively isolated. Given that some specimens in Dunedin potentially showed evidence for mitochondrial introgression, it is certainly possible that the two widespread clades can freely hybridize in sympatry, although the extent to which these largely allopatric clades overlap remains to be seen. Despite the fact that Dunedin was the only study site in which divergent mitochondrial lineages were found to co-occur, several of the global sampling sites were represented by only one or two museum specimens and thus larger sample sizes would have been desired to more thoroughly test for sympatry. Laboratory experiments examining variance in reproductive parameters within Fiona pinnata would therefore be useful in determining whether reproductive barriers exist between the divergent lineages identified in this thesis. Although Fiona pinnata has free-swimming planktotrophic larvae, larval dispersal alone presumably cannot explain the trans-oceanic distributions and close phylogenetic relationships detected within the widespread clades identified in this study. Consequently, passive rafting appears to be a very important mechanism of long-distance dispersal for this nudibranch. The challenge of finding floating substrate colonized by suitable prey in the open ocean presumably limits settlement success, particularly for an animal with an extremely rapid development rate. Willan (1979) suggested that veligers of Fiona pinnata can possibly delay metamorphosis until chemically triggered by prey, but this hypothesis has never been tested. Some authors have demonstrated that nudibranch veligers can metamorphose in response to a water-borne cue (Hadfield and Scheuer 1985; Lambert and Todd 1994), while other studies have shown that direct contact with live prey is required to induce metamorphosis (Perron and Turner 1977; Chia and Koss 1978; Todd 1981). Lab experiments have shown that Fiona pinnata veligers in the presence of a food resource hatch in approximately five days (Hollemann 1972), but larval duration in the open ocean is unknown for this nudibranch. Planktonic larvae are already recognized as an important determinant in the geographic ranges and levels of gene flow displayed by benthic marine invertebrates (Shuto 1974; Valentine and Jablonski 74

1983; Hedgecock 1986), and thus delayed metamorphosis in this nudibranch could very well result in extremely long-lived planktonic larvae. If Fiona pinnata is truly able to spend long stretches of time in the until proper food is located, then it is possible that larval duration in this animal could be a more important element in its dispersal patterns than we currently realize. Detailed laboratory observations of larval mode parameters would thus be necessary to test this hypothesis. The Indonesian specimen appears to represent an undescribed species, and thus its morphology, life history, and developmental strategy remain to be explored. Because long-distance dispersal in Fiona pinnata appears to be current-mediated, the high divergence detected within the Indonesian individual is probably a result of local adaptation due to being cut off from the world’s major oceanic circulation regimes. Accordingly, samples from similiarly oceanographically isolated locations could very well yield other highly divergent lineages of this nudibranch. One such possibility is the Mediterranean, as Fiona pinnata is known to occur there (Costa 1866; Aliani and Molcard 2003), and rafters in this semi-closed marine system may be largely allopatric from Atlantic populations. Although this study sampled extensively from the Pacific Ocean, only three sample lots could be acquired from elsewhere. An extended molecular phylogeny of Fiona using specimens from a greater variety of locales worldwide may reveal intermediate haplotypes between the divergent lineages identified in this study. The discovery of undetected lineages would be useful in reconstructing this nudibranch’s speciation history and would help to refine estimates of divergence times. In particular, temperate Atlantic specimens would shed further light on the evolutionary history of this organism, as the current interpretation is based largely on the geographic distribution of clades within the Pacific Ocean. Although the anti-tropical distribution observed for clade A was only observed within the Pacific Ocean and was supported by a single Alaskan specimen, it is possible that this pattern could have been demonstrated more conclusively had more specimens from the temperate Northern Hemisphere been available. Given that oceanic circulation regimes, rather than latitude, appear to exert the greatest influence on the geographic partitioning of genetic diversity within Fiona pinnata, populations occurring within the Alaska and Oyashio current systems would presumably show close phylogenetic affinities with the Alaskan individual examined in this study, and would also provide an opportunity 75 to examine the dispersal dynamics of this nudibranch within the temperate Northern Hemisphere (Ohtani 1970; Tabata 1975). However, given that the samples from Alaska and Indonesia are each represented by a single specimen, it cannot be entirely ruled out that these are anomalous individuals rather than haplotypes characteristic of the lineages in these locations. Increased sample sizes per location thus would have been ideal in order to determine how representative these individuals are of their respective locations and thus further clarify some of the interpretations (e.g., the apparent anti-tropical distribution observed in clade A). Passive rafting has frequently been invoked to speculatively explain the distributions of some widespread nudibranchs (e.g., Carmona et al. 2011; Martynov and Schrödl 2011), but this is the first study to present molecular data confirming the role of long-distance rafting in shaping gene flow among trans-oceanic populations of a nudibranch. Despite a paucity of direct evidence for rafting maintaining large-scale genetic connectivity in opisthobranchs, over forty species of nudibranch have been reported rafting (Thiel and Gutow 2005b), and this list is certainly not exhaustive. Accordingly, future phylogeographic analyses of wide-ranging nudibranchs that have been observed rafting on the high seas may very well provide evidence for the strong influence of passive rafting on their far-flung distributions and colonization histories. Rafting research generally focuses on macroalgal substrata, and thus most studies concentrate on the higher latitude systems where temperate seaweeds predominate (Rothausler 2012). However, it is clear that passive rafting still occurs in the tropics due to the presence of other floating objects such as driftwood, plastic debris, and pumice, and thus future work should more thoroughly examine the potential of these substrata for long-distance dispersal. A comparison of rafting dynamics in the temperate and tropical zones would help to shed light on the different abiotic and biotic factors regulating the persistence of buoyant objects at the sea surface and would thereby provide information on the genetic diversification and structuring of faunal rafting communities within these contrasting biogeographic zones.

76

REFERENCES

Aguilera-Munoz, F., F. Lafarga-Cruz, and C. Gallardo-Escarate. 2009. Molecular analysis in Chilean commercial gastropods based on 16S rRNA, COI and ITS1-5.8S rDNA-ITS2 sequences. Gayana 73:17-27. Akaike, H. 1974. A new look at the statistical model identification. IEEE Transactions on Automatic Control 19:716-723. Alder, J., and A. Hancock. 1851a. Including the families of Gasteropoda from Neritidae and Elysiadae. In: Forbes, E., and S. Hanley (Eds.), A history of British Mollusca and their shells. Volume 3. Van Voorst, London, 616 pp. Alder, J., and A. Hancock. 1851b. Descriptions of two new species of nudibranchiate Mollusca, one of them forming the type of a new genus. Annals and Magazine of Natural History (series 2)8(46):290-302, pls. 9-10. Alder, J., and A. Hancock. 1855. A Monograph of the British Nudibranchiate Mollusca with Figures of all the Species 7, fam. 1, pls. 21a, 27; fam. 2, pls. 1.2; fam. 3, pls. 38a, 45-48. Appendix, p. I-XXVIII. Aliani, S., and A. Molcard. 2003. Hitch-hiking on floating marine debris: macrobenthic species in the Western Mediterranean Sea. Hydrobiologia 503:59-67. Apte, S., and P. A. Gardner. 2002. Population genetic subdivision in New Zealand greenshell mussel (Perna canaliculus) inferred from single-stranded conformation polymorphism analysis of mitochondrial DNA. Molecular Ecology 11:1617-1628. Armbruster, G. F. J., C. H. M. van Moorsel, and E. Gittenberger. 2000. Conserved sequence patterns in the non-coding ribosomal ITS-1 of distantly related snail taxa. Journal of Molluscan Studies 66:570-573. Avise, J. C. 2000. Phylogeography: The History and Formation of Species. Harvard University Press, Cambridge, MA. Ayers, K. L., and J. M. Waters. 2005. Marine biogeographic disjunction in central New Zealand. Marine Biology 147:1045-1052. Bayer, F. M. 1963. Observations on pelagic molluscs associated with the siphonophores Velella velella and Physalia. Bulletin of Marine Science of the Gulf and Caribbean, University of Miami 13:454-466. Beeman, R. D. 1977. Gastropoda: Opishtobranchia. In: Gise, A.C., J. S. Pearse (Eds.), Reproduction of Marine Invertebrates, Volume 4. Academic Press, New York, pp. 115-179 Behrens, D. W. 1980. Pacific Coast Nudibranchs: A Guide to the Opisthobranchs of the northeastern Pacific. Sea Challengers Natural History Books, Danville. Behrens, D. W. 1992. Pacific Coast Nudibranchs: A Guide to the Opisthobranchs from Alaska to Baja California. 2nd ed. Sea Challengers. 112 pp. Bennett, I. 1966. Some pelagic molluscs and associated animals in South-Eastern Australia. Journal of the Malacological Society of Australia 9:40-51. Benson, D. A., I. Karsch-Mizrachi, D. J. Lipman, J. Ostell, and D. L. Wheeler. GenBank. Nucleic Acids Research 33:D34-D38. Benzie, J. A. H., and S. T. Williams. 1997. Genetic structure of giant clam (Tridacna maxima) populations in the West Pacific is not consistent with dispersal by present-day ocean currents. Evolution 51:768-783. 77

Bergh, L. S. R. 1857. Anatomisk Untersogelse af Fiona atlantica, Bgh:273-335. Bergh, L. S. R. 1892. Opisthobranches provenant des campagnes du yacht l’Hirondelle. Resultats des Campagnes Scientifiques accomplies sur son yacht par Albert I, Prince Souverain de Monaco 4:35 pp. Bergh, L. S. R. 1894. Die Opisthobranchien. Reports on the dredging operations off the west coast of Central America to the Galapagos, to the west coast of Mexico, and in the Gulf of California, in charge of Alexander Agassiz, carried on by the U. S. Fish Commission steamer “Albatross” during 1891. Bulletin of the Museum of Comparative Zoology, Harvard 25:125-233. Bergh, L. S. R. 1900. Ergebnisse einer Reise nach dem Pacific (Schauinsland 1896- 1897). Die Opisthobranchier. Zoologische Jahrbucher 13:207-246. Bieri, R. 1966. Feeding preferences and rates of the snail, Janthina prolongata, the barnacle, Lepas anserifera, the nudibranchs and Fiona pinnata, and the food web in the marine neuston. Publications of the Seto Marine Biological Laboratory 14: 161-170. Bierne, N., P. Borsa, C. Daguin, D. Jollivet, F. Viard, F. Bonhomme, and P. David. 2003a. Introgression patterns in the mosaic hybrid zone between edulis and M. galloprovincialis. Molecular Ecology 12:447-461. Bierne, N., F. Bonhomme, and P. David. 2003b. Habitat preference and the marine- speciation paradox. Proceedings of the Royal Society of London B 270:1399- 1406. Bigelow, R. S. 1965. Hybrid zones and reproductive isolation. Evolution 19:449-458. Blanquer, A., and M. J. Uriz. 2007. Cryptic speciation in marine sponges evidenced by mitochondrial and nuclear genes: a phylogenetic approach. Molecular Phylogenetics and Evolution 45:392-397. Bleidorn, C., I. Kruse, S. Albrecht, and T. Bartolomaeus. 2006. Mitochondrial sequence data expose the putative cosmopolitan polychaete Scoloplos armiger (Annelida, Orbiniidae) as a species complex. BMC Evolutionary Biology 6:47. Bowen, B. W., and W. S. Grant. 1997. Phylogeography of the sardines (Sardinops spp.): assessing biogeographic models and population histories in temperate upwelling zones. Evolution 51:1601-1610. Brattstrom, H., and A. Johanssen. 1983. Ecological and regional zoogeography of the marine benthic fauna of Chile. Sarsia 68. Briggs, J. C. 1987. Antitropical distribution and evolution in the Indo-West Pacific Ocean. Systematic Zoology 36:237-247. Briggs, J. C. 1995. Global Biogeography. Elsevier, New York. Buckely, T. R., C. Simon, and G. K. Chambers. 2001. Phylogeography of the New Zealand cicada Maoricicada campbelli based on mitochondrial DNA sequences: ancienct clades associated with Cenozoic environmental change. Evolution 7:1395-1407. Burridge, C. P., and R. W. G. White. 2000. Molecular phylogeny of the antitropical subgenus Goniistius (Perciformes: Cheilodactylidae: Cheilodactylus): evidence for multiple transequatorial divergences and non-monophyly. Biological Journal of the Linnean Society 70:435-458. Burridge, C. P. 2002. Antitropicality of Pacific fishes: molecular insights. Environmental Biology of Fishes 65:151-164. Bushing, W. W. 1994. Biogeographic and ecological implications of kelp rafting as a dispersal vector for marine invertebrates. In: Proceedings of the Fourth 78

California Islands Symposium: Update on the Status of Resources, March 22- 25, 1994, Halvorson, W. L., G. J. Maender (eds.). Santa Barbara Museum of Natural History, pp. 103-110. Camus, P. A. 2001. Marine biogeography of continental Chile. Revista Chilena de Historia Natural 74:587-617. Cardenas, L., J. C. Castilla, and F. Viard. 2009. A phylogeographical analysis across three biogeographical provinces of the south-eastern Pacific: the case of the marine gastropod Concholepas concholepas. Journal of Biogeography 36:969- 981. Carmona, L., T. M. Gosliner, M. Pola, and J. L. Cervera. 2011. A molecular approach to the phylogenetic status of the aeolid genus Roller, 1973 (Nudibranchia). Journal of Molluscan Studies 77:417-422. Carmona, L., M. A. E. Malaquias, T. M. Gosliner, M. Pola, and J. L. Cervera. 2011. Amphi-Atlantic distributions and cryptic species in Sacoglossan sea slugs. Journal of Molluscan Studies 77:401-412. Carpenter, E. J., and K. L. Smith. 1972. Plastics on the Sargasso Sea surface. Science 175:1240-1241. Casteel, D. B. 1904. The cell-lineage and early larval development of Fiona marina, a nudibranchiate mollusk. Proceedings of the Academy of Natural Sciences of Philadelphia 56:325-405. Chan, B. K. K., L. M. Tsang, and F. Shih. 2009. Morphological and genetic differentiations of the stalked barnacle Heteralepas japonica Aurivillius, 1892, with description of a new species of Heteralepas Pilsbry, 1907, from the Philippines. The Raffles Bulletin of Zoology 20:83-95. Chia, F. S., and R. Koss. 1978. Development and metamorphosis of the planktotrophic larvae of Rostana pulchra (Mollusca: Nudibrancha). Marine Biology 46:109-119. Chiswell, S. M. 1996. Variabiliity in the Southland Current, New Zealand. New Zealand Journal of Marine and Freshwater Research 30:1-17. Chiswell, S. M. 2009. Colonisation and connectivity by intertidal limpets among New Zealand, Chatham and Sub-Antarctic Islands. II. Oceanographic connections. Marine Ecology Progress Series 388:121-135. Clement, M., D. Posada, and K. Crandall. 2000. TCS: a computer program to estimate gene genealogies. Molecular Evolution 9:1657-1660. Coates, A. G., and J. A. Obando. 1996. The geologic evolution of the Central American Isthmus. In: Evolution and Environment in Tropical America, Jackson, J. B. C., A. F. Budd (eds.). University Chicago Press, Chicago, pp. 21-56. Colborn, J., R. E. Crabtree, J. B. Shaklee, E. Pfeiler, and B. W. Bowen. 2001. The evolutionary enigma of bonefishes (Albula spp.): cryptic species and ancient separations in a globally distributed shorefish. Evolution 55:807-820. Coleman, A. W., and V. D. Vacquier. 2002. Exploring the phylogenetic utility of ITS sequences for animals: a test case for abalone (Haliotis). Journal of Molecular Evolution 54:246-257. Colgan, D. J., W. F. Ponder, E. Beacham, and J. M. Macaranas. 2003. Gastropod phylogeny based on six fragments from four genes representing coding or non-coding and mitochondrial or nuclear DNA. Molluscan Research 23:123- 148. 79

Collin, R. 2001. The effects of mode of development on phylogeography and population structure of North Atlantic Crepidula (Gastropoda: Calyptraeidae). Molecular Ecology 10:2249-2262. Coombs, D. S., and C. S. Landis. 1966. Pumice from the South Sandwich eruption of March 1962 reaches New Zealand. Nature 209:289-290. Costa, A. 1866. Acquisti fatti durante l’anno 1863. Annuario del Museo Zoologico della Universita di Napoli 3 (2):13-41. Costa, A. 1867. Acquisti fatti durante l’anno 1864. Annuario del Museo Zoologico della Universita di Napoli 4 29-30. Table 1, figure 1-3. Costello, D. P. 1938. Notes on the breeding habits of the nudibranchs of Monterey Bay and vicinity. Journal of Morphology 63:319-343. Coston-Clements, L. L. R. Dettle, D. E. Hoss, and F. A. Cross. 1991. Utilization of the Sargassum habitat by marine invertebrates and vertebrates- a review. NOAA Technical Memorandum NMFS-SEFSC-296. Cox, C. B., and P. D. Moore. 2005. Biogeography: an ecological and evolutionary approach. 7th ed. Blackwell Publishing, Malden, Massachusetts, USA. Dansgaard, W., S. H. Johnsen, H. B. Clausen, D. Dahl-Jensen, N. S. Gundestrup, C. U. Hammer, C. S. Hvidberg, J. P. Steffensen, Sveinbjornsdottir, J. Jouzel, and G. Bond. 1993. Evidence for general instability of past climate from a 250kyr ice-core record. Nature 364:218-220. Darwin, C. R. 1872. On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. 6th edition (John Murray), London, UK. Dawson, E. W. 1965. Oceanography and marine zoology of the New Zealand Subantarctic. Proceedings of the New Zealand Ecological Society 12:44-57. Dawson, M. N., and D. K. Jacobs. 2001. Molecular evidence for cryptic species of Aurelia aurita (, Scyphozoa). Biological Bulletin 200:92-96. Depraz, A., J. Hausser, and M. Pfenninger. 2009. A species delimitation approach in the Trochulus sericeus/hispidus complex reveals two cryptic species within a sharp contact zone. BMC Evolutionary Biology 9:171. Derycke, S., T. Remerie, T. Backeljau, A. Vierstraete, J Vanfleteren, M. Vincx, and T. Moens. 2008. Phylogeography of the Rhabditis (Pellioditis) marina species complex: evidence for long-distance dispersal, and for range expansions and restricted gene flow in the northeast Atlantic. Molecular Ecology 17:3306- 3322. de Queiroz, A. 2005. The resurrection of oceanic dispersal in historical biogeography. Trends in Ecology and Evolution 20:68-73. de Vargas, C., R. Norris, L. Zaninetti, S. W. Gibb, and J. Pawlowski. 2003. Molecular evidence of cryptic speciation in planktonic foraminifera and their relation to oceanic provinces. Proceedings of the National Academy of Sciences USA 96:2864-2868. Dinapoli, A., C. Tamer, S. Frassen, L. Naduvilezhath, and A. Klussmann-Kolb. 2006. Utitility of H3 gene sequences for phylogenetic reconstruction- a case study of heterobranch Gastropoda. Bonner Zoologische Beitrage 55:191-202. Donald, K. M., M. Kennedy, and H. G. Spencer. 2005. Cladogenesis as the result of long-distance rafting events in South Pacific topshells (Gastropoda, Trochidae). Evolution 59:1701-1711. 80

Donlan, C. J., and P. A. Nelson. 2003. Observations of invertebrate colonized flotsam in the eastern tropical Pacific with a discussion of rafting. Bulletin of Marine Science 72:231-240. Duque-Caro, H. 1990. Neogene stratigraphy, paleoceanography and paleobiology in northwest South America and the evolution of the Panama Seaway. Palaeogeography, Palaeoclimatology, Palaeoecology 77:203-234. Edgar, G. J. 1987. Dispersal of faunal and floral propagules associated with drifting Macrocystis pyrifera plants. Marine Biology 95:599-610. Edmands, S., and D. C. Potts. 1997. Population genetic structure in brooding sea anemones (Epiactis spp.) with contrasting reproductive modes. Marine Biology 127:485-498. Edmunds, M. 1977. Larval development, oceanic currents, and origins of the Opisthobranch fauna of Ghana. Journal of Molluscan Studies 43:301-308. Eschscholtz, F. 1831. Zoologischer Atlas 4. 19 pp., pls. 16-20. G. Reimer, Berlin. Page 14, plate 19, figure 1. Excoffier, L. G. Laval, and S. Schneider. 2005. Arlequin ver. 3.0: An integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1:47-50. Farrapeira, C. M. R., A. V. de Oliveira, M. de Melo, D. F. Barbosa, and K. M. E. da Silva. 2007. Ship hull fouling in the port of Recife, Pernambuco. Brazilian Journal of Oceanography 55:207-221. Faucci, A., R. J. Toonen, and M. G. Hadfield. 2007. Host shift and speciation in a coral –feeding nudibranch. Proceedings of the Royal Society B 274:111-119. Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783-791. Fine, M. L. 1970. Faunal variation on pelagic Sargassum. Marine Biology 7:112-122. Finlay, H. J. 1927. A further commentary on New Zealand Molluscan Systematics. Transactions and Proceedings of the Royal Society of New Zealand 57:320- 485. Folmer, O., M. Black, W. Hoeh, R. Lutz, and R. Vrijenhoek. 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology 3:294- 299. Forskal, P. 1775. Vermes.1.Mollusca. In: Descriptiones animalium, avium, amphiborum, piscorum, insectorum, vermum quae in itinere orientali observavit Petrus Forskal. Hauniae:99-100. (Fauna Arabica). Foster, B. A. 1978. The marine fauna of New Zealand: barnacles (Cirripedia: Thoracica). New Zealand Oceanic Institute Memoirs, volume 69. Wellington, New Zealand, 190 pp. Francis, M. P. 1996. Geograhic distribution of marine reef fishes in the New Zealand region. New Zealand Journal of Marine and Freshwater Research 30:35-55. Fraser, C. I., R. Nikula, H. G. Spencer, and J. M. Waters. 2009. Kelp genes reveal effects of subantarctic sea ice during the Last Glacial Maximum. Proceedings of the National Academy of Science of the USA 106:3249-3253. Fraser, C. I., R. Nikula, and J. M. Waters. 2011. Oceanic rafting by a coastal community. Proceedings of the Royal Society B 278:649-655. Fujita, M. K., A. D. Leache, F. T. Burbrink, J. A. McGuire, and C. Moritz. 2012. Coalescent-based species delimitation in an integrative taxonomy. Trends in Ecology and Evolution 27:480-488. 81

Fyler, C. A., T. W. Reeder, A. Berta, G. Antonelis, A. Aguilar, and E. Androukaki. Historical biogeography and phylogeny of monachine seals (Pinnipedia: Phocidae) based on mitochondrial and nuclear DNA data. Journal of Biogeography 32:1267-1279. Gallardo, M. H., and J. I. Carrasco. 1996. Genetic cohesiveness among populations of Concholepas concholepas (Gastropoda, Muricidae) in southern Chile. Journal of Experimental Marine Biology and Ecology 197:237-249. Gerber, L. R., S. S. Heppell, F. Ballantyne, and E. Sala. 2005. The role of dispersal and demography in determining the efficacy of marine reserves. Canadian Journal of Fisheries and Aquatic Sciences 62:863-871. Geyer, C. J. 1991. Markov chain Monte Carlo maximum likelihood. Computing Science and Statistics: Proceedings of the 23rd Symposium on the Interface pp.156-163. Goetze, E. 2003. Cryptic speciation on the high seas; global phylogenetics of the copepod family Eucalanidae. Proceedings of the Royal Society B 270:2321- 2331. Goldstien, S. J., D. R. Schiel, and N. J. Gemmell. 2006. Comparative phylogeography of coastal limpets across a marine disjunction in New Zealand. Molecular Ecology 15:3259-3268. Gomez-Uchida, D., D. Weetman, L. Hauser, R. Galleguillos, and M. Retamal. 2003. Allozyme and AFLP analyses of genetic population structure in the hairy edible crab Cancer setosus from the Chilean coast. Journal of Biology 23:486-494. Gordillo, S. 2006. The presence of Tawera gayi (Hupe in Gay, 1854) (Veneridae, Bivalvia) in southern South America: did Tawera achieve a Late Cenozoic circumpolar traverse? Palaeogeography Palaeoclimatology Palaeoecology 240:587-601. Gosliner, T. M., D. W. Behrens, and A. Valdes. 2008. Indo-Pacific nudibranchs and sea slugs: a field guide to the world’s most diverse fauna. California Academy of Sciences/Sea Challengers Natural History Books: San Francisco. 426 pp. Grande, C., J. Templado, J. L. Cervera, and R. Zardoya. 2004. Phylogenetic relationships among Opisthobranchia (Mollusca: Gastropoda) based on mitochondrial cox1, trnV, and rrnL genes. Molecular Phylogenetics and Evolution 33:378-388. Grant, W. S., and F. M. da Silva-Tatley. 1997. Lack of genetically-subdivided population structure in Bulla digitalis, a southern African marine gastropod with lecithotrophic development. Marine Biology 129:123-137. Grant, W. S., R. W. Leslie, and B. W. Bowen. 2005. Molecular genetic assessment of bipolarity in the anchovy genus Engraulis. Journal of Fish Biology 67:1242- 1265. Grantham, B. A., G. L. Eckert, and A. L. Shanks. 2003. Dispersal potential of marine invertebrates in diverse habitats. Ecological Applications 13:108-116. Graves, J. E. 1998. Molecular insights into the population structures of cosmopolitan marine fishes. The Journal of Heredity 89:427-437. Gray, J. E. 1857. Guide to the systematic distribution of Mollusca in the British Museum. Part I. Taylor and Francis, London, xii + 230 pp, page 227. Greenwood, P. G., and R. N. Mariscal. 1984. The utilization of cnidarian nematocysts by aeolid nudibranchs: nematocyst maintenance and release in . Tissue and Cell 16:719-730. 82

Guindon, S., and O. Gascuel. 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52:696-704. Hadfield, M. G., and H. Karlson. 1969. Externally induced metamorphosis in a marine gastropod. American Zoologist 9:1122. Hadfield, M. G., and D. Scheuer. 1985. Evidence for a soluble metamorphic inducer in Phestilla: ecological, chemical, and biological dta. Bulletin of Marine Science 37:556-566. Hafner, M. S., P. D. Sudman, F. X. Villablance, T. A. Spradling, J. D. Demastes, and S. A. Nadler. 1994. Disparate rates of molecular evolution in cospeciating hosts and parasites. Science 265:1087-1090. Harrold, C., and S. Lisin. 1989. Radio-tracking rafts of giant kelp: local production and regional transport. Journal of Experimental Marine Biology and Ecology 130:237-251. Hasegawa, M., H. Kishino, and T. Yano. 1985. Dating the human-ape split by a molecular clock of mitochondrial DNA. Journal of Molecular Evolution 22:160-174. Heath, R. A. 1985. A review of the physical oceanography of the seas around New Zealand- 1982. New Zealand Journal of Marine and Freshwater Research 19:79-124. Hebert, P. D. N., S. Ratnasingham, and J. R. D. Waard. 2003. Barcoding animal life: cytochrome c oxidase subunit 1 divergences among closely related species. Proceedings of the Royal Society 270:S96-S99. Hedgecock, D. 1986. Is gene flow from pelagic larval dispersal important in the adaptation and evolution of marine invertebrates? Bulletin of Marine Science 39:550-564. Helmuth, B. R. R. Veit, and R. Holbertson. 1994. Long-distance dispersal of a subantarctic brooding bivalve (Gaimardia trapesina) by kelp-rafting. Marine Biology 120:421-442. Hernandez-Carmona, G., B. Hughes, and M. H. Graham. 2006. Reproductive longevity of drifting kelp Macrocystis pyrifera (Phaeophyceae) in Monterey Bay, USA. Journal of Phycology 42:1199-1207. Higashi, M., G. Takimoto, and N. Yamamura. 1999. Sympatric speciation by sexual selection. Nature 402:523-526. Higgins, D., J. Thompson, T. Gibson, J. D. Thompson, D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Research 22:4673- 4680. Highsmith, R. C. 1985. Floating and algal rafting as potential dispersal mechanisms in brooding invertebrates. Marine Ecology Progress Series 25:169-179. Hilbish, J. T., A. Mullinax, S. I. Dolven, A. Meyer, R. K. Koehn, and P. D. Rawson. 2000. Origin of the antitropical distribution pattern in marine mussels (Mytilus spp.): routes and timing of trans-equatorial migration. Marine Biology 136:69- 77. Hinojosa, I., S. Boltana, D. Lancellotti, E. Macayo, P. Ugalde, N. Valdivia, N. Vasquez, W. A. Newman, and M. Thiel. 2006. Geographic distribution and description of four pelagic barnacles along the south east Pacific coast of Chile- a zoogeographical approximation. Revista Chilena de Historia Natural 79:13-27. 83

Ho, S. Y. W., M. J. Phillips, A. Cooper, and A. J. Drummond. 2005. Time dependency of molecular rate estimates and systematic overestimation of recent divergence times. Molecular Biology and Evolution 22:1561-1568. Hobday, A. J. 2000a. Age of drifting Macrocystis pyrifera (L.) C. Agardh rafts in the Southern California Bight. Journal of Experimental Marine Biology and Ecology 253:97-114. Hobday, A. J. 2000b. Persistence and transport of fauna on drifting kelp (Macrocystis pyrifera (L.) C. Agardh) rafts in the Southern California Bight. Journal of Experimantal Marine Biology and Ecology 253:75-96. Hodell, D. A., and J. P. Kennett. 1986. Late Miocene-Early Pliocene stratigraphy and paleoceanography of the South Atlantic and southwest Pacific oceans: a synthesis. Paleoceanography 1:285-311. Holleman, J. J. 1972. Observations on growth, feeding, reproduction, and development in the opisthobranch Fiona pinnata (Eschscholtz). Veliger 15:142-146. Hopkins, J., A. G. P. Shaw, and P. Challenor. The Southland Front, New Zealand: variability and ENSO correlations. Continental Shelf Research 30:1535-1548. Horta, E. C. M. 2010. Phylogeny, connectivity and dispersal patterns of the giant kelp Macrocystis pyrifera (Phaeophyceae). PhD Thesis, Victoria University of Wellington. Huelsenbeck, J. P., and F. Ronquist. 2001. MRBAYES: Bayesian inference of phylogeny. Bioinformatics 17:754-755. Hurr, K. A., P. J. Lockhart, P. B. Heenan, and D. Penny. 1999. Evidence for the recent dispersal of Sophora (Leguminosae) around the Southern Oceans: molecular data. Journal of Biogeography 26:565-577. Hutton, F. W. 1882. Notes on Branchiate Mollusca. Transactions of the Royal Society of New Zealand 14:162-167. Hyman, L. H. 1955. The Invertebrates IV. Echinodermata. McGraw-Hill Book Company Inc., New York. Inatsuchi, A., S. Yamato, and Y. Yusa. 2010. Effects of temperature and food availability on growth and reproduction in the newstonic pedunculate barnacle Lepas anserifera. Marine Biology 157:899-905. Ingolfsson, A. 1995. Floating clumps of seaweed around Iceland: natural microcosms and a means of dispersal for shore fauna. Marine Biology 122:13-21. Jablonski, D. 1986. Larval ecology and macroevolution inmarine invertebrates. Bulletin of Marine Science 39:565-587. Jensen, K. 2005. Distribution and zoogeographic affinities of the nudibranch fauna (Mollusca, Opisthobranchia, Nudibranchia) of the Faroe Islands. BIOFAR Proceedings 2005:109-124. Johannesson, K. 1988. The paradox of Rockall- why is a brooding gastropod (Littorina saxatilis) more widespread than one having a planktonic larval dispersal stage (L. littorea)? Marine Biology 99:507-513. Jokiel, P. L. 1984. Long distance dispersal of reef corals by rafting. Coral Reefs 3:113-116. Keigwin, L. D. 1982. Isotopic paleocanography of the Caribbean and east Pacific: role of Panama uplift in late Neogene time. Science 217:350-353. Kemppainen, P., M. Panova, J. Hollander, and K. Johannesson. 2009. Complete lack of mitochondrial divergence between two species of NE Atlantic marine intertidal gastropods. Journal of Evolutionary Biology 22:2000-2011. 84

Kimura, M. 1980. A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16:111-120. Kingsford, M. J. 1992. Drift algae and small fish in coastal waters of northeastern New Zealand. Marine Ecology Progress Series 80:41-55. Kingsford, M. J., and J. H. Choat. 1985. The fauna associated with drift algae captured with a plankton-mesh purse seine net. Limnology and Oceanography 30:618-630. Kinlan, B. P., and S. D. Gaines. 2003. Propagule dispersal in marine and terrestrial environments: a community perspective. Ecology 84:2007-2020. Klatau, M., C. A. M. Russo, C. Lazoski, N. Boury-Esnault, J. P. Thorpe, and A. M. Sole-Cava. 1999. Does cosmopolitanism result from overconservative systematics? A case study using the marine sponge Chondrilla nucula. Evolution 53:1414-1422. Knox, G. A. 1960. Littoral ecology and biogeography of the southern ocean. In: Pantin, C. F. A. (Ed.), A discussion of the biology of the southern cold temperate zone. Proceedings of the Royal Society 152B, London, pp. 577-624. Knowlton, N. 1993. Sibling species in the sea. Annual Review of Ecology and Systematics 24:189-216. Knowlton, N. 2000. Molecular genetic analyses of species boundaries in the sea. Hydrobiologia 420:73-90. Koufopanou, V., D. G. Reid, S. A. Ridgeway, and R. H. Thomas. 1999. A molecular phylogeny of the patellid limpets (Gastropoda: Patellidae) and its implications for the origins of their antitropical distribution. Molecular Phylogenetics and Evolution 11:138-156. Kropp, B. 1931. The pigments of Velella spirans and Fiona marina. Biological Bulletin, Marine Biological Laboratory, Woods Hole 60:120-123. Krug, P. J., M. S. Morlet, J. Asif, L. L. Hellyar, and W. M. Blom. 2008. Molecular confirmation of species status for the rare cephalaspidean Melanochlamys lorrainae (Rudman, 1968), and comparison with its sister species M. cylindrical Cheesman, 1881. Journal of Molluscan Studies 74:267-276. Krug, P. J. 2011. Patterns of speciation in marine gastropods: a review of the phylogenetic evidence for localized radiations in the sea. American Malacological Bulletin 29:169-186. Ladoukakis, E. D., C. Saavedra, A. Magoulas, and E. Zouros. 2002. Mitochondrial DNA variation in a species with two mitcochondrial genomes: the case of Mytilus galloprovincialis from the Atlantic, the Mediterranean, and the Black Sea. Molecular Ecology 11:755-769. Lambert, W. J., and C. D. Todd. 1994. Evidence for a water-borne cue inducing metamorphosis in the dorid nudibranch mollusc Adalaria proxima (Gastropoda: Nudibranchia). Marine Biology 120:265-271. Lancellotti, D. A., and J. A. Vasquez. 1999. Biogeographical patterns of benthic macroinvertebrates in the southeastern Pacific littoral. Journal of Biogeography 26:1001-1006. Lazoski, C., A. M. Sole-Cava, N. Boury-Esnault, M. Klatau, and C. A. M. Russo. 2001. Cryptic speciation in a high gene flow scenario in the oviparous marine sponge Chondrosia reniformis. Marine Biology 139:421-429. Le Gac, M., Jp Feral, E. Poulin, M. Veyret, and A. Chenuil. Identification of allopatric clades in the cosmopolitan opiuroid species complex Amphipholis 85

squamata (Echinodermata). The end of a paradox? Marine Ecology Progress Series 278:171-178. Lee, C. E. 2000. Global phylogeography of a cryptic copepod species complex and reproductive isolation between genetically proximate ‘populations’. Evolution 54:2014-2027. Lessios, H. A., B. D. Kessing, and S. J. Pearse. 2001. Population structure and speciation in tropical seas: global phylogeography of the sea urchin Diadema. Evolution 55:955-975. Levin, L. A. 2006. Recent progress in understanding larval dispersal: new directions and digressions. Integr. Comp. Biology 46:282-297. Lindberg, D. R. 1991. Marine biotic interchange between the northern and southern hemispheres. Palaeobiology 7:308-324. Loza, A. L. and L. F. Lopez-Jurado. 2008. Comparative study of the epibionts on the pelagic and mature female loggerhead turtles on the Canary and Cape Verde Islands. Proceedings of the Twenty-Fourth Annual Symposium on Sea Turtle Biology and Conservation. NOAA Technical Memorandum, p.100. Luttikhuizen, P. C., J. Drent, and A. J. Baker. 2003. Disjunct distribution of highly diverged mitochondrial lineage clade and population subdivision in a marine bivalve with pelagic larval dispersal. Molecular Ecology 12:2215-2229. Macaya, E. C., S. Boltana, I. A. Hinojosa, J. E. Macchiavello, N. A. Valdivia, and N. R. Vasquez. 2005. Presence of Sporophylls in floating kelp rafts of Macrocystis spp. (Phaeophyceae) along the Chilean Pacific coast. Journal of Phycology 41:913-922. MacFarland, F. M. 1966. Studies of opisthobranchiate mollusks of the Pacific Coast of North America. Memoirs of the California Academy of Sciences 6:546pp. Malaquias, M. A. E., and D. G. Reid. 2008. Systematic revision of the recent speies of Bullidae (Mollusca: Gastropoda: Cephalaspidea), with a molecular phylogenetic analysis. Zoological Journal of the Linnean Society 153:453- 543. Malaquias, M. A. E., G. P. Calado, V. Padula, G. Villani, and J. L. Cervera. 2009. Molluscan diversity in the North Atlantic Ocean: new records of opisthobranch gastropods from the Archipelago of Azores. Marine Biodiversity Records 2:1-9. Malaquias, M. A. E., and D. G. Reid. 2009. Tethyan vicariance, relictism and speciation: evidence from a global molecular phylogeny of the opisthobranch genus Bulla. Journal of Biogeography 36:1760-1777. Martynov, A., and M. Schrödl. 2011. Phylogeny and evolution of corambid nudibranchs (Mollusca: Gastropoda). Zoological Journal of the Linnean Society 163:585-604. Mayr, E. 1954. Geographic speciation in tropical echinoids. Evolution 8:1-18. McDonald, G. R., and J. W. Nybakken. 1978. Additional notes on the food of some California nudibranchs with a summary of known food habits of California species. Veliger 21:110-119. McDonald, G. R. 2006. Nudibranch Systematic Index. UC Santa Cruz: Institute of Marine Sciences. Retrieved from: http://escholarship.org/uc/item/0hb5d87j. Meneses, I., and B. Santelices. 2000. Patterns and breaking points in the distribution of benthic algae along the temperate Pacific coast of South America. Revista Chilena de Historia Natural 73:615-623. 86

Mercer, J. H., and J. F. Sutter. 1982. Late Miocene-earliest Pliocene glaciation in southern Argentina: implications for global ice-sheet history. Palaeogeography, Paleoclimatology, Palaeoecology 38:185-206. Merritt, T. J. S., L Shi, M. C. Chase, M. A. Rex, R. J. Etter, and J. M. Quattro. 1998. Universal cytochrome b primers facilitate intraspecific studies in molluscan taxa. Molecular Marine Biology and Biotechnology 7:7-11. Meyer, C. P. 2003. Molecular systematics of cowries (Gastropoda: Cypraeidae) and diversification patterns in the tropics. Biological Journal of the Linnean Society 79:401-459. Miglietta, M. P., A. Faucci, and F. Santini. 2011. Speciation in the sea: overview of the symposium and discussion of future directions. Integrative and Comparative Biology 51:449-455. Miller, M. C. 1974. Aeolid nudibranchs (Gastropoda: Opisthobranchia) of the family Glaucidae from New Zealand waters. Zoological Journal of the Linnean Society 54:31-61. Miller, M. C. 1977. Aeolid nudibranchs (Gastropoda: Opisthobranchia) of the family Tergipedidae from New Zealand waters. Zoological Journal of the Linnean Society 60:197-222. Morando, M., L. J. Avila, and J. W. Sites. 2003. Sampling strategies for delimiting species: genes, individuals, and populations in the Liolaemus elongates- kriegeri complex (Squamata: Liolaemidae) in Andean-Patagonian South America. Systematic Biology 52:159-185. Moritz, C. 1994. Defining evolutionarily-significant-units for conservation. Trends in Ecology and Evolution 9:373-375. Morrow, C. C., J. P. Thorpe, and B. E. Picton. 1992. Genetic divergence and cryptic speciation in two morphs of the common subtidal nudibranch Doto coronata (Opisthobranchia: Dendronotacea: ) from the northern Irish Sea. Marine Ecology Progress Series 84:53-61. Munday, P. L., L. van Herwerden, and C. L. Dudgeon. Evidence for sympatric speciation by host shift in the sea. Current Biology 14:1498-1504. Nations, D. 1979. The genus Cancer and its distribution in time and space. Bulletin of the Biological Society of Washington 3:153-187. Nikula, R., C. I. Fraser, H. G. Spencer, and J. M. Waters. 2010. Circumpolar dispersal by rafting in two subantarctic kelp-dwelling crustaceans. Marine Ecology Progress Series 405:221-230. Nikulina, E. A., H. Hanel, and P. Schafer. 2007. Cryptic speciation and paraphyly in the cosmopolitan bryozoan Electra pilosa- impact of the closing on species evolution. Molecular Phylogenetics and Evolution 45:765-776. Nilsson-Cantell, C. A. 1930. Thoracic cirriped collected in 1925-1927. Discovery Reports 2:223-260. O’Foighil, D., B. A. Marshall, T. J. Hilbish, and M. A. Pino. 1999. Trans-Pacific range extension by rafting is inferred for the flat oyster Ostrea chilensis. Biological Bulletin 196:122-126. O’Foighil, D., and C. J. Jozefowicz. 1999. Amphi-Atlantic phylogeography of direct- developing lineages of Lasaea, a genus of brooding bivalves. Marine Biology 135:115-122. Ohtani, K. 1970. Relative transport in the Alaskan Stream in winter. Journal of Oceanography 26:271-282. 87

Oliverio, M., M. Cervelli, and P. Mariottini. 2002. ITS2 rRNA evolution and its congruence with the phylogeny of muricid neogastropods (Caenogastropoda, Muricoidea). Molecular Phylogenetics and Evolution 25:63-69. Oosthuizen, A., M. Jiwaji, and P. W. Shaw. 2004. Genetic analysis of the Octopus vulgarisi population on the coast of South Africa. South African Journal of Science 100:603-607. Palumbi, S. R. 1992. Marine speciation on a small planet. Trends in Ecology and Evolution 7:114-118. Palumbi, S. R. 2003. Population genetics, demographic connectivity, and the design of marine reserves. Ecological Applications 13:S146-S158. Palumbi, S. R., G. Grabowsky, T. Duda, L. Geyer, and N. Tachino. 1997. Speciation and population genetic structure in tropical Pacific sea urchins. Evolution 51:1506-1517. Parker, T., and V. Tunnicliffe. 1994. Dispersal strategies of the biota on an oceanic seamount- implications for ecology and biogeography. Biological Bulletin USA 187:336-345. Pawson, D. L. 1961. Distribution patterns of New Zealand echinoderms. Tuatara 9:9- 18. Pawson, D. L. 1965. The distribution of echinoderms alogn the east coast of New Zealand. Transactions of the Royal Society of New Zealand 6:245-252. Perron, F. E., and R. D. Turner. 1977. Develoment, metamorphosis and natural history of the nudibranch Doridella obscura Verrill (Corambidae: Opisthobranchia). Journal of Experimental Marine Biology and Ecology 27:171-185. Peters, A. F., and A. M. Breeman. 1992. Temperature response of discjunct temperate brown algae indicated long-distance dispersal of microthalli across the tropics. Journal of Phycology 28:428-438. Philippe, H., E. A. Snell, E. Bapteste, P. Lopez, P. W. H. Holland, and D. Casane. 2004. Phylogenomics of eukaryotes: impact of missing data on large alignments. Molecular Biology and Evolution 21:1740-1752. Pola, M., J. L. Cervera, and T. M. Gosliner. 2007. Phylogenetic relationships of Nembrothinae (Mollusca: Doridacea: Polyceridae) inferred from morphology and mitochondrial DNA. Molecular Phylogenetics and Evolution 43:726-742. Pontin, D. R., and R. H. Cruickshank. 2012. Molecular phylogenetics of the genus Physalia (Cnidaria: Siphonophora) in New Zealand coastal waters reveals cryptic diversity. Hydrobiologia 686:91-105. Posada, D., and T. R. Buckley. 2004. Model selection and model averaging in phylogenetics: advantages of the AIC and Bayesian approaches over likelihood ratio tests. Systematic Biology 53:793-808. Posada, D. 2008. jModelTest: Phylogenetic model averaging. Molecular Biology and Evolution 25:1253-1256. Pruvot-Fol, A. 1954. Mollusques Opisthobranches. Faune de France 58, 460pp. Quattro, J. M., M. R. Chase, M. A. Rex, T. W. Greig, and R. J. Etter. 2001. Extreme mitochondrial DNA divergence within populations of the deep-sea gastropod Frigidoalvania brychia. Marine Biology 139:1107-1113. Raahauge, P., and T. K. Kristenson. 2000. A comparison of Bulinus africanus group species (Planorbidae; Gastropoda) by use of internal transcribed spacer region 1 combined by morphological and anatomical characters. Acta Tropica 75:85- 94. 88

Reed, D. C., C. D. Amsler, and A. W. Ebeling. 1992. Dispersal in kelps: factors affecting spore swimming and competence. Ecology 73:1577-1585. Reid, D. G., E. Rumbak, and R. H. Thomas. 1996. DNA, morphology and fossils: phylogeny and evolutionary rates of the gastropod genus Littorina. Philosophical Transactions of the Royal Society of London B 351:877-895. Ronquist, F., and J. P. Huelsenbeck. 2003. MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572-1574. Rothausler, E., L. Gutow, and M. Thiel. 2012. Floating Seaweeds and Their Communities. In: Wiencke, C., K. Bischof (eds.), Seaweed Biology, Volume 219. Springer Berlin Heidelberg, pp.359-380. Rudman, W. B. 2002. Fiona pinnata (Eschscholtz, 1831). In: Sea Slug Forum, Australian Museum, Sydney. 18 June 2002. . Rudman, W. B., and P. R. Bergquist. 2007. A review of feeding specificity in the sponge-feeding Chromodorididae (Nudibranchia: Mollusca). Molluscan Research 27:60-88. Sandoval-Castellanos, E., M. Uribe-Alcocer, and P. Diaz-Jaimes. 2010. Population genetic structure of the Humboldt squid (Dosidicus gigas d’Orbigny, 1835) inferred by mitochondrial DNA analysis. Journal of Experimental Marine Biology and Ecology 385:73-78. Savin, S. M., R. G. Douglas, and F. G. Stehli. 1975. Tertiary marine paleotemperatures. Geological Society of America Bulletin 86:1499-1510. Scarabino, F. 2004. Two gastropods associated with floating objects from the Uruguayan coast. Comunicaciones de la Sociedad Malacologica del Uruguay 8:275-277. Schaefer, K. 1996. Review of data on cephalaspid reproduction, with special reference to the genus Haminaea (Gastropoda: Opisthobranchia). Ophelia 45:17-37. Scheltema, R. S. 1971. Larval dispersal as a means of genetic exchange between geographically separated populations of shallow-water benthic marine gastropods. Biological Bulletin 140:284-322. Scheltema, R. S. 1986. On dispersal and planktonic larvae of benthic invertebrates: an eclectic overview and summary of problems. Bulletin of Marine Science 39:290-322. Schrödl, M. 2003. Sea Slugs of Southern South America. ConchBooks, Hackenheim. Schrödl, M., K. M. Jörger, A. Klussmann-Kolb, and N. G. Wilson. 2011. Bye bye “Opisthobranchia”! A review on the contribution of mesopsammic sea slugs to euthyneuran systematics. Proceedings of the 4th International Opisthobranch Workshop, Thalassas 27:101-112. Schroth, W., G. Jarms, B. Streit, and B. Schierwater. 2002. Speciation and phylogeography in the cosmopolitan marine moon jelly, Aurelia sp. BMC Evolutionary Biology 2:1-10. Schwaninger, H. R. 2008. Global mitochondrial DNA phylogeography and biogeographic history of the antitropically and longitudinally disjunct marine bryozoan Membranipora membranacea L. (Cheilostomata): Another cryptic marine sibling species complex? Molecular Phylogenetics and Evolution 49:893-908. 89

Shanks, A. L., B. A. Grantham, and M. H. Carr. 2003. Propagule disperal distance and the size and spacing of marine reserves. Ecological Applications 13:159- 169. Shields, C. C. 2009. Nudibranchs of the Ross Sea, Antarctica: phylogeny, diversity, and divergence. MSc Thesis, Clemson University. Shuto, T. 1974. Larval ecology of prosobranch gastropods and its bearing on biogeography and paleontology. Lethaia 7:239-256. Skerman, T. M. 1958. Rates of growth in two species of Lepas (Cirripedia). New Zealand Journal of Science 1:402-411. Smissen, R. D., P. J. Garnock-Jones, and G. K. Chambes. 2003. Phylogenetic analysis of ITS sequences suggests a Pliocene origin for the bipolar distribution of Scleranthus (Caryophyllaceae). Australian Systematic Botany 16:301-315. Smith, S. D. A. 2002. Kelp rafts in the Southern Ocean. Global Ecology and Biogeography 11:67-69. Sole-Cava, A. M., J. P. Thorpe, and C. D. Todd. 1994. High genetic similarity between geographically distant populations in a sea anemone with low dispersal capabilities. Journal of the Marine Biological Association UK 74:895-902. Sponer, R. 2002. Phylogeography and evolutionary history of the cosmopolitan brooding brittle star Amphipolis squamata (Delle Chiaje, 1828; Echinodermata: Ophiuroidea). PhD Thesis, University of Otago. Stepien, C. A., and R. H. Rosenblatt. 1996. Genetic divergence in anti-tropical pelagic marine fishes (Trachurus, Merluccius, and Scomber) between North and South America. Copeia 1996:586-598. Stout, C. C., M. Pola, and A. Valdes. 2010. Phylogenetic analysis of nudibranchs with emphasis on northeastern Pacific species. Journal of Molluscan Studies 76:367-375. Strathmann, R. R. 1985. Feeding and nonfeeding larval development and life-history evolution in marine invertebrates. Annual Review of Ecology and Systematics 16:339-361. Strathmann, R. R. 1990. Why do life histories evolve differently in the sea? American Zoologist 30:197-207. Suter, H. 1913. Manual of the New Zealand Mollusca. Government Printer, Wellington, pp. 586-587. 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. Tabuta, S. 1975. The general circulation of the Pacific Ocean and a brief account of the North Pacific Ocean. Part I: Circulation and Volume Transports. Atmosphere 13:133-168. 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. Tapia, F. J., J. Pineda, F. J. Ocampo-Torres, H. L. Fuchs, P. E. Parnell, P. Montero, and S. Ramos. 2004. High-frequency observations of wind-forced onshore transport at a coastal site in Baja California. Contintinental Shelf Research 24:1573-1585. 90

Teske, P. R., I. Papadopoulos, and G. I. Zardi. 2007. Implications of life history for genetic structure and migration rates of southern African coastal invertebrates: planktonic, abbreviated, and direct development. Marine Biology 152:697- 711. Thiel, M. 2002. The biogeography of littoral algal-associated peracarids along the Pacific coast of Chile. Journal of Biogeography 29:999-1008. Thiel, M., and L. Gutow. 2005a. The ecology of rafting in the marine environment I. The floating substrata. Oceanography and Marine Biology: An Annual Review 42:181-263. Thiel, M., and L. Gutow. 2005b. The ecology of rafting in the marine environment II. The rafting organisms and community. Oceanography and Marine Biology: An Annual Review 43:279-418. Thiel, M., and P. A. Haye. 2006. The ecology of rafting in the marine environment. III. Biogeographical and evolutionary consequences. Oceanography and Marine Biology: an Annual Review 44:323-429. Thollesson, M. 1998. Discrimination of two Dendronotus species by allozyme electrophoresis and the reinstatement of Dendronotus lacteus (Thompson, 1840) (Nudibranchia, Dendronotoidea). Zoologica Scripta 27:189-195. Thollesson, M. 2000. Increasing fidelity in parsimony analysis of dorid nudibranchs by differential weighting, or a tale of two genes. Molecular Phylogenetics and Evolution 16:161-172. Thompson, T. E. 1958. The natural history, embryology, larval biology, and post- larval development of Adalaria proxima (Alder and Hancock) (Gastropoda: Opisthobranchia). Philosophical Transactions of the Royal Society 242B:1-58. Thompson, T. E. 1962. Studies on the ontogeny of hombergi Cuvier (Opisthobranchia). Philosophical Transactions of the Royal Society of London 24:171-218. Thorpe, J. P., and A. M. Sole-Cava. 1994. The use of electrophoresis in invertebrate systematics. Zoologica Scripta 23:1-18. Todd, C. D. 1981. The ecology of nudibranch mollusks. Oceanography and Marine Biology Annual Review 19:141-234. Turner, L. M., and N. G. Wilson. 2008. Polyphyly across oceans: a molecular phylogeny of the Chromodorididae (Mollusca, Nudibranchia). Zoologica Scripta 37:23-42. Valdes, A. and O. A. Campillo. 2004. Systematics of pelagic aeolid nudibranchs of the family Glaucidae (Mollusca, Gastropoda). Bulletin of Marine Science 75:381-389. Valentine, J. W., and D. Jablonski. 1983. Speciation in the shallow sea: general patterns and biogeographic controls. In: Evolution, Time and Space: The Emergence of the Biosphere. R. W. Sims, J. H. Price, and P. E. S. Whalles (eds.). Academic, New York, pp. 201-226. van Oppen, M. J. H., B. J. McDonald, B. Willis, and D. J. Miller. 2001. The evolutionary history of the coral genus Acropora (Scleractinia, Cnidaria) based on a mitochondrial and a nuclear marker: reticulation, incomplete lineage sorting, or morphological convergence? Molecular Biology and Evolution 18:1315-1329. Vasquez, J. A., and A. Vega. 2007. The Humboldt Current System of northern-central Chile: Oceanographic processes, ecological interactions and socioeconomic feedback. Oceanography and Marine Biology: an Annual Review 45:195-344. 91

Vermeij, G. J. 1987. The dispersal barrier in the tropical Pacific: implications for molluscan speciation and extinction. Evolution 41:1046-1058. Vermeij, G. J. 1989. Geographical restriction as a guide to the causes of extinction: the case of the cold northern oceans during the Neogene. Paleobiology 15:335-356. Via, S. 2001. Sympatric speciation in animals: the ugly duckling grows up. Trends in Ecology and Evolution 16:381-390. Vidal, R., I. Meneses, and M. Smith. 2008. Phylogeography of the genus Spongites (Corallinales, Rhodophyta) from Chile. Journal of Phycology 44:173-182. Wade, C. M., P. B. Mordan, and F. Naggs. 2006. Evolutionary relationships among the Pulmonate land snails and slugs (Pulmonata, Stylommatophora). Biological Journal of the Linnean Society 87:593-610. Walsh, P. S., D. A. Metzger, and R. Higuchi. 1991. Chelex 100 as a medium for simple extraction of DNA for PCR-based typing from forensic material. Biotechniques 10:506-513. Waters, J. M. 2008. Driven by the West Wind Drift? A synthesis of southern temperate marine biogeography, with new directions of dispersalism. Journal of Biogeography 35:417-427. Waters, J. M., and M. S. Roy. 2003a. Global phylogeography of the fissiparous sea- star genus Coscinasterias. Marine Biology 142:185-191. Waters, J. M., and M. S. Roy. 2003b. Marine biogeography of southern Australia: phylogeographical structure in a temperate sea-star. Journal of Biogeography 30:1787-1796. Waters, J. M., and M. S. Roy. 2004a. Out of Africa: the slow train to Australasia. Systematic Biology 53:18-24. Waters, J. M., and M. S. Roy. 2004b. Phylogeography of a high dispersal New Zealand sea-star: does upwelling block gene flow? Molecular Ecology 13:2797-2806. Westheide, W., E. Ha ß-Cordes, M. Krabusch, and M. Muller. 2003. Ctenodrilus serratus (Polychaeta: Ctenodrilidae) is a truly amphi-Atlantic meiofauna species- evidence from molecular data. Marine Biology 142:637-642. Whittaker, R. J., and J. M. Fernandez-Palacios. 2007. Island biogeography: ecology, evolution, and conservation. Oxford University Press, UK. Wiens, J. J. 2006. Missing data and the design of phylogenetic analyses. Journal of Biomedical Informatics 39:34-42. Willan, R. C. 1979. New Zealand locality records for the aeolid nudibranch Fiona pinnata (Eschscholtz). Tane (Journal of the Auckland University Field Club) 25:141-147. Williams, M. N. 1978. Buccal glands of some aeolid nudibranchs (ultrastructure and biochemistry). MSc Thesis, University of Auckland. Williams, S. T. 2000. Species boundaries in the starfish genus Linckia. Marine Biology 136:137-148. Williams, S. T., and D. G. Reid. 2004. Speciation and diversity on tropical rocky shores: a global phylogeny of snails of the genus Echinolittorina. Evolution 58:2227-2251. Wilson, N. G. 2002. Egg masses of chromodorid nudibranchs (Mollusca: Gastropoda: Opisthobranchia). Malacologia 44:289-305. Wilson, N. G., R. L. Hunter, S. J. Lockhart, and K. M. Halanych. 2007. Multiple lineages and absence of panmixia in the “circumpolar” crinoid 92

Promachocrinus kerguelensis from the Atlantic sector of Antarctica. Marine Biology 152:895-904. Wilson, N. G., M. Schrodl, and K. M. Halanych. 2009. Ocean barriers and glaciation: evidence for explosive radiation of mitochondrial lineages in the Antarctic sea slug kerguelensis (Mollusca, Nudibranchia). Molecular Ecology 18:965- 984. Wollscheid-Lengeling, E., J. Boore, W. Brown, and H. Wagele. 2001. The phylogeny of Nudibranchia (Opisthobranchia, Gastropoda, Mollusca) reconstructed by three molecular markers. Organisms Diversity and Evolution 1:241-256. Wood, A. R., S. Apte, E. S. MacAvoy, and J. P. A. Gardner. 2007. A molecular phylogeny of the marine mussel genus Perna (Bivalvia: Mytilidae) based on nuclear (ITS1&2) and mitochondrial (COI) DNA sequences. Molecular Phylogenetics and Evolution 44:685-698. Woodruff, F., and S. M. Savin. 1989. Micoene deepwater oceanography. Paleoceanography 4:87-140. Yorifuji, M., H. Takeshima, K. Mabuchi, and M. Nishida. 2012. Hidden diversity in a reef-dwelling sea slug, ianthina (Nudibranchia, Aeolidina), in the northwestern Pacific. Zoological Science 29:359-367. Zeppilli, D., A. Vanreusel, and R. Danovaro. 2011. Cosmopolitan and biogeography of the genus Manganonema (Nematoda: Monhysterida) in the deep sea. Animals 1:291-305. Ziegler, K. S., M. A. McCartney, D. R. Levitan, and H. A. Lessios. 2005. Sea urchin bindin divergence predicts gamete compatibility. Evolution 59:2399-2404. Zulliger, D. E., and H. A. Lessios. 2010. Phylogenetic relationships in the genus Astropecten Gray (Paxillosida: Astropectinidae) on a global scale: molecular evidence for morphological convergence, species-complexes and possible cryptic speciation. Zootaxa 2504:1-19. Zvyagintsev, A. Y., and S. R. Mikhajlov. 1985. The formation of fouling communities on a long-distance ship in tropical waters. 4:16-20.

93

Appendix 1: A taxonomic overview of Fiona pinnata

The following table briefly summarizes the taxonomic history of Fiona pinnata, and is based on McDonald’s (2006) ‘Nudibranch Systematic Index’, which pooled all known synonyms of this ubiquitous aeolid nudibranch into the monotypic family Fionidae (Alder & Hancock, 1855). . Author Year Classification established Forsskal 1775 Limax marinus Hasselt 1824 Eolida alba Eschscholtz 1831 Eolidia pinnata Quoy & Gaimard 1832 Eolidia longicauda Alder & Hancock 1851 Fiona nobilis Alder & Hancock 1851 nobilis Alder & Hancock 1855 Family Fionidae Gray 1857 Superfamily Fionoidea Bergh 1857 Fiona atlantica Costa 1866 Hymenaeolis elegantissima Hutton 1881 Aeolis plicata Bergh 1879 Fiona pacifica Suter 1913 Fiona marina Iredale & O’Donoghue 1923 Fiona pinnata Finlay 1927 Dolicheolis longicauda

94

Appendix 2: Mitochondrial cytochrome b sequence data

A portion of the 3’ region of mitochondrial cytochrome b was amplified using the primer Cytb_1032R (5’-GGACAAGCTCCTARYCAWGT-3’) paired with either Cytb_241F (5’-YGCTAATGGGGCWTCTTT-3’) or Cytb_780F (5’- CCCTATAGTKACKCCKGTTC-3’). These novel primers were designed for this study using data drawn from GenBank based on the cytochrome b sequences of three nudibranch species (Chromodoris magnifica, Notodoris gardineri, Roboastra europaea) and a sea hare (Aplysia californica). PCR amplifications were performed in 20-µl volumes and contained 1 U of MyTaq DNA polymerase (Bioline), 1 µl of each primer (10 µM), 4 µl of 5X MyTaq Red Buffer (Bioline), and 1 µl of extracted DNA. Thermal cycling for cytochrome b fragments was carried out using Eppendorf Mastercycler ep gradient proS machines and involved an initial denaturation step of 4 mins at 94°C, 40 cycles of 45 s at 94°C, 45 s at 45°C, and 1.5 mins at 72°C, followed by a final extension step of 6 min at 72 °C. The cytochrome b sequence data generated for a subset of thirteen Fiona pinnata individuals was concordant with the phylogeographic diversity and structuring observed in portions of mitochondrial COI. Cytochrome b haplotypes clustered into two major clades, one temperate in distribution and the other tropical/sub-tropical. These two groupings were separated by 4.5-6% sequence divergence, as illustrated in the neighbor-joining tree shown below.

95

Appendix 3: COI haplotype frequencies in New Zealand Sample Collection locality Latitude, n COI haplotype frequencies site longitude 1 Ohui Beach, Opoutere --- 2 NZNorth1=0.5; (AK118314) NZNorth2=0.5 2 St. Clair Beach, --- 2 NZSouth1=1.0 Dunedin, Otago 3 Dunedin, Otago 45° 49' 26" S, 9 NZSouth1=0.67; 171° 05' 02" E NZSouth2=0.11; NZSouth3=0.11; NZSouth4=0.11 4 Dunedin, Otago 45° 50' 02" S, 4 NZSouth1=0.75; 171° 29' 58" E NZSouth5=0.25 5 Sandfly Bay, Dunedin, --- 3 NZSouth12=0.33; Otago NZSouth13=0.33; NZSouth14=0.33 7 Dunedin, Otago 45° 48' 51" S, 2 NZSouth1=0.5; 171° 11' 42" E NZSouth/Chile=0.5 8 Dunedin, Otago 45° 48' 49" S, 4 NZSouth1=1.0 171° 12' 31" E 9 Dunedin, Otago 45° 48' 48" S, 2 NZSouth1=0.5; 171° 14' 01" E NZSouth7=0.5 10 Dunedin, Otago 46° 11' 31" S, 3 NZSouth1=0.33; 170° 14' 34" E NZSouth4=0.33; NZSouth8=0.33 11 Dunedin, Otago 45° 47' 23" S, 2 NZSouth1=0.5; 170° 58' 33" E NZSouth9=0.5 12 Cape Rodney, Goat --- 3 NZNorth1=0.33; Island, Leigh NZNorth3=0.33; NZNorth4=0.33 13 Dunedin, Otago 45° 47' 23" S, 6 NZSouth1=0.67; 170° 58' 33" E NZSouth10=0.16; NZSouth11=0.16

96

Appendix 4: COI haplotype frequencies in Coquimbo, Chile Sample Collection site (latitude, longitude) n COI haplotype frequencies site 1 30° 17' 15" S, 71° 49' 35" W 2 Chile1=0.5; NZSouth/Chile=0.5 2 30° 6' 18" S, 71° 55' 17" W 2 Chile2=0.5; Chile3=0.5 3 29° 36' 33" S, 71° 33' 46" W 2 Chile2=0.5; Chile4=0.5 4 29° 47' 00" S, 71° 47' 20" W 2 Chile5=0.5; Chile6=0.5 5 29° 47' 38" S, 71° 46' 15" W 2 Chile7=0.5; Chile8=0.5 6 30° 10' 03" S, 71° 29' 41" W 2 Chile2=0.5; Chile9=0.5 7 29° 49' 23" S, 71° 42' 07" W 2 Chile10=0.5; Chile11=0.5 8 29° 49' 52" S, 71° 44' 19" W 2 Chile12=0.5; Chile13=0.5 9 29° 51' 04" S, 71° 46' 31" W 2 Chile14=0.5; Chile15=0.5 10 29° 22' 48" S, 72° 01' 60" W 2 Chile8=0.5; NZSouth/Chile=0.5