Phylogeny and Population Genetics of Acorn in Family (Crustacea: Cirripedia)

WU, Tsz Huen

A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Philosophy in Biology

The Chinese University of Hong Kong July 2011

/ ‘.. 1 e SEP zotTji XJvtN JJBRAP.Y S\S\-my<^W Thesis/Assessment Committee

Professor WONG, Chong Kim (Chair) Professor CHU, Ka Hou (Thesis Supervisor) Professor ANG, Put O Jr (Committee Member) Professor BARBER, Paul H (External Examinor) Declaration

I declare that this thesis represents my own works and that it has not been previously included in a thesis, dissertation or degree, diploma or other qualification.

‘ . Abstract

Barnacles in the family Chthamalidae are shallow water inhabitants commonly found in temperate and tropical regions worldwide. Despite their high ecological values that make them suitable model organisms in environmental science, the phylogenetic relationship within this family is poorly known. In this study, I aim to investigate the intrafamilial relationship of barnacles in the family Chthamalidae using three molecular markers. In addition, the phylogeography of one of the species from the Chthamalidae, mow in the Northwest Pacific region is studied. Phylogenetic trees generated from the sequences of the three markers, the mitochondrial 12S ribosomal RNA gene, nuclear histone 3 gene and nuclear elongation factor 1 alpha region showed some inconsistencies from the current classification in the family Chthamalidae. The subfamily Euraphiinae was polyphyletic with members clustered with either Chthamalinae or Notochthamalinae. The traditional classification on subfamily Notochthamalinae was valid in general, although it was of polyphyletic origin. Apart from the exclusion of a Euraphiinae barnacle, one of the members in this subfamily, Pseudooctomeris sulcata, is distantly related to the rest of the Notochthamalinae barnacles. The relationship of the subfamily Chthamalinae inferred from genetic data matched the traditional taxonomic classification to a certain degree, except the inclusion of Microeuraphia from the Euraphiinae. Apart from revealing the dissimilarity in phylogenetic relationships inferred by traditional and molecular , this study does not support the hypothesis that there is a tendency of /eduction in the number of shell plates in the evolution of acom

i barnacles. The phylogenetic tree shows that the species with different number of shell plates intermix and position randomly. It is believed that the evolution of the number of shell plates is likely to be a random event. For the study on phylogeography of C. mow, analysis of the mitochondrial cytochrome c oxidase subunit I sequences of over 240 individuals from 13 localities in Northwest Pacific region has revealed three genetically distinct lineages, namely Taiwan lineage, Ogasawara lineage and Ryukyu lineage, with sequence divergence of 4-9%. Using the molecular clock of closely related barnacles in the genus Euraphia, the time of divergence of the three lineages dated back roughly to 1.3-2.9 million years ago, suggesting they originated during Pleistocene glaciations, when lowering of sea level might have caused geographical isolation and diversification. The Pleistocene glaciations might also lead to population decline and rapid postglacial population expansion in the three lineages, as indicated by the mismatch distribution analysis. Bayesian skyline analysis shows that these island inhabiting lineages have experienced demographic expansion at -450,000-250,000 years ago, which is much earlier than the expansion time for the continental barnacle species. Despite the role of Pleistocene glaciations in causing diversifications and affecting the demographic histories in C. mow, ocean circulation patterns and geography in the region have probably shaped the contemporary distribution range of the lineages. The Kuroshio Current starts near northern Philippines and flows along the eastern coast of Taiwan is likely to maintain the genetic connectivity between populations in the Philippines and the majority of the Taiwan lineage. The Kuroshio Current that flows along the Ryukyu Island chain and mainland Japan may homogenize the genetic structure of both the Ryukyu lineage and

ii Ogasawara lineage among different localities in the Ryukyu Islands and Ogasawara Islands. While the Kuroshio Current promotes gene flow between different localities, the complicated hydrography in the Kerama Gap in the Ryukyu Islands prevents complete homogenizing between the Taiwan lineage and the other two lineages, and hence maintains a rather discrete boundary between them. Seasonal eddy currents near Ogasawara Islands probably play a role in producing distinct population structure observed in Ogasawara Island, probably retaining larval supply of C. mow there, and reduce the connectivity between Ogasawara Islands and the other localities.

iii 摘要

小藤壺科(Chthamalidae)的成員廣泛分佈於全世界之溫帶和熱帶地區的潮間

帶。由於它們擁有很高的生態價值，使它們經常應用於各項環境科學研究中，但是

人們對小藤壺科的親緣關係所知甚少。是次研究利用了三個分子標記去探討小藤壺

科內的親緣關係。此外，是次研究亦會探討直背小藤壺…fhamqlus moro, family

Chthamalidae)於西北太平洋地區的親緣地理關係。

在是次研究中，由三個分子標記的基因序列（線粒體的12S核糖體RNA基因，

細胞核的組蛋白亞單位3基因和細胞核的延伸因子la基因）所產生的系統演化樹出

現了一些與傳統小藤壺科分類學上不一致的地方。首先，地藤壺亞科(subfamily

Euraphiinae)為多源系，這亞科下的兩個藤壶品種分別歸於小藤壺亞科（subfamily

Chthamalinae)或樂都小藤壶亞科（subfamily Notochthamalinae)�此外’在是次硏究

中顯示擬肋筋藤壺sulcata (subfamily Notochthamalinae)並不是樂都

小藤壺亞科的近親，所以樂都小藤壺亞科亦為多源糸。對於小藤壺亞科內的親緣關

係，除了系統演化樹所顯示的其中一種地藤壺與它有密切親緣關係外，是次基因排

序的結果與傳統的小藤壶亞科的分類大致上吻合。

是次研究除了揭示與傳統分類學相異的親緣關係之外，亦明顯不支持藤壺穀板.

數目在進化過程中有減少趨勢的假設。糸統演化樹說明，藤壶穀板數目與小藤壺科

的演化史並無明確之關係，而小藤壺科穀板數量的演化可能是一個隨機事件。

iv

>! ‘ 關於直背小藤壺的親緣地理學硏究，這次硏究對來自西北太平洋的13個地

區，共超過240個藤壺樣本的線粒體細胞色素C氧化酶亞基I (COI)序列進行分

析，並發現了三個不同的基因族群，分別定名為台灣族群、小笠原族群及琉球族群，

而各族群之間的序列差異為4-9%�跟據與直背小藤壺同源的地藤壺的分子時鐘，

三個直背小藤壺族群的分化時間大致可追溯至1.3-2.9萬年前，即更新世冰期。更新

世冰期時的惡劣氣候，及因為冰冠擴張所引致的海平面下降，都可能是造成族群分

化的主因。另外，硏究中的不配對分佈分析顯示更新世冰期的惡劣氣候導致小笠原

族群和琉球族群數量大幅減少，及冰期完結後族群的迅速擴張。’

直背小藤壺的族群結構除了受到更新世冰期的影響外，還會受到西北太平洋地

區的海洋環流模式及地理因素影響。例如，流經菲律賓北部及台灣東部沿海的黑潮

維持了菲律賓和其他台灣族群統群間的基因聯繫。另外，流經琉球列島和小笠原群

島的黑潮亦分別混和了小笠原族群及琉球族群在北菲律賓海各個地點的遺傳結

構。相反地，位於琉球群島間的慶良間海槽，因為有著複雜的水流’形成了一個屏

障，阻止了琉球列島以南的台灣族群，與其以北的兩個族群之間的基因交流，因此

形成了它們之間的分佈邊界。另外，小笠原群島附近的季節性過流可能減少小笠原

群島上的直背小藤壺族群和其他地方族群之間的連通性，從而導致小笠原群島族群

與別的族群結構上出現明顯的差異。

V Acknowledgements

First and foremost I would like to express my deepest gratitude to my supervisor, Prof Ka Hou Chu, for his warm encouragements and exceptional patience in guiding me throughout this thesis and the master study. I very much appreciate the freedom he grants me in exploring the field of marine biology, and also his generosity in providing me opportunities to broaden my horizon. Being a member in Prof Chu's lab is one of the most thankful and fascinating things I have ever experienced, as it brings me the key to the awesome world of biology. For this I am indebted to him. My appreciation also goes to Dr Benny Chan from Academia Sinica, Taiwan for providing specimens for the research project, and more importantly, fundamental training in barnacle taxonomy and constructive comments on my thesis. I would like to give thanks to my thesis committee, Prof Chong Kim Wong and Prof Put O ANG, Jr. for their insightful questions and valuable advice on my research work, and Prof Paul Barber, for devoting his time serving as my external examiner. It is a pleasure to thank all the staff and colleagues in the Simon FS Li Marine Science Laboratory for providing me with technical support and lots of joy. Special thanks are given to Dr LM Tsang and Ms KY Ma, for teaching me all the lab skills and • assisting in my research work. They are not just good teachers, but also companies who are willing to share my burdens and support me always. It is really a blessing to work with such a friendly and wonderful research team. Finally, I want to extend my gratitude to my beloved family and friends. I must thank my parents for having confidence in me, in all my pursuits, and telling me that no

vi matter what I want to do, they are with me and love me always. Thanks go to my brother as well, who always delights me when I am frustrated. I am never a tough girl; without the love, support and understanding from my family and the warm encouragements from friends, I can never go this far, and it would be impossible for me to finish my thesis. For these, I would like to dedicate my thesis to them.

vii Content Chapter 1 General Introduction to thesis 1 Chapter 2 Literature review 3 2.1 Introduction 3 2.1.1 General introduction to barnacles 3 2.1.2 Classification of barnacles 4 2.1.3 Importance of barnacles 4 2.2 Molecular phylogenetics of barnacles 6 2.2.1 What is phylogenetics 6 2.2.2 Phylogenetic studies on barnacles 7 2.2.3 Choices on characters for barnacle phylogenetics: Morphological characters molecular characters 11 2.2.4 Choices of molecular markers in phylogenetic and population genetic studies on barnacles 14 2.3 The use of barnacles as model organism in population genetic studies 21 2.5.1 Pleistocene glaciations 22 2.5.2 Oceanographic pattern and habitat availability 23 Chapter 3 Phylogenetic relationship of barnacles in family Chthamalidae 26 3.1 Introduction 26 3.2 Materials and methods 31 3.2.1 Sample collection, DNA extraction and amplification 31 3.2.2 Phylogenetic analyses 34 3.3 Results 35 3.4 Discussion 40 3.4.1 Subfamily Notochthamalinae 40 3.4.2 Subfamily Chthamalinae 42 3.4.3 Subfamily Euraphiinae 42 3.4.4 Phylogenetic relationship in the family Chthamalidae 43 3.4.5 Suggestions on taxonomy of Chthamalidae 45 Chapter 4 Cryptic Diversity and Genetic Structure of the Acorn Barnacle Chthamalus moro in the Northwest Pacific 47 4.1 Introduction 47 4.2 Materials and methods 53 4.2.1 Sample collection, DNA extraction and amplification 53 4.2.2 Phylogenetic analysis 55 4.2.3 Population genetic analysis 55 4.3 Results 57 , 4.3.1 Phylogenetic analysis 57 4.3.2 Population genetics analyses, demographic history and neutrality 61 4.4 Discussion 72 4.4.1 Origin and the systematic status of the three lineages 72

viii 4.2,2 Demographic history of the three lineages 81 4.3.3 Contemporary distribution and genetic connectivity of the three lineages... Chapter 5 Concluding remarks 87 References 89

ix List of tables Table 3.1. Classification, sampling localities and molecular markers used in Chapter 3 32 Table 3.2. Summary of parsimony results 37 Table 4.1. Sampling localities, abbreviation and genetic diversity indices for chapter 4 62-63 Table 4.2. Pairwise Ost values for the three lineages 64 Table 4.3. Pairwise Ost values for populations of North West Pacific 65 Table 4.4a. Pairwise Ost values for populations of Ogasawara lineage 67 Table 4.4b. Pairwise Ost values for populations of Ryukyu lineage 68 Table 4.4c. Pairwise Ost values for populations of Taiwan lineage 69 Table 4.5. Analysis of Molecular Variance (AMOVA) for genetic structure between different localities 70

V List of figures Fig 3.1 Changes in the classification scheme of the family Chthamalidae over time.....30 Fig 3.2 Maximum likelihood tree from the combined 12S, H3 and EFla analysis under the best-fitting model CTR+I+G 38 Fig. 3.3 Bayesian inference tree from combined 12S, H3 and EFla analysis 39 Fig. 4.1 Geographical differences in the Northwest Pacific region between the present day and the Last Glacial Maximum during the Pleistocene 48 Fig. 4.2 Main path of the North Equatorial Current and Kuroshio Current in the Northwest Pacific region 50 Fig. 4.3 Main path of Kuroshio Current together and the positions of the three straits, the Luzon Strait, Ilan Strait and Tokara Strait in the Northwest Pacific region 51 Fig. 4.4 Sampling localities of Chthamalus mow in the Northwest Pacific 54 Fig. 4.5 Neighbor-joining tree of mitochondrial COI haplotypes 58 Fig. 4.6 Relative abundance of the three lineages of Chthamalus mow 60 Fig. 4.7a Mismatch distribution analysis of Ogasawara lineage 73 Fig. 4.7b Mismatch distribution analysis of Ryukyu lineage 74 Fig. 4.7c Mismatch distribution analysis of Taiwan lineage 75 Fig. 4.8 Bayesian skyline plots of Ogasawara, Ryukyu and Taiwan lineage 76 Fig. 4.9 Relative distribution (in %) of Ryukyu and Ogasawara lineages in different shore level respectively in Okinawa Island 77

xi Chapter 1

General introduction to Thesis

This master thesis aims at studying the phylogenetic relationships of barnacles in the family Chthamalidae, and also the phylogeography of one of the species from the Chthamalidae, Chthamalus mow using a molecular approach. The family Chthamalidae is a species-rich and cosmopolitan family of barnacles, which encompasses more than 40 species. This family plays an important role in barnacle evolution, as it appears to be the transition between stalked barnacles and acom barnacles. However, even though this family is of great evolutionary significance, the very fundamental knowledge of the Chthamalidae, that is, the phylogenetic relationship within this group have remained obscured for 150 years, since Darwin first attempted to investigate the family thoroughly (Darwin, 1854). Although many revisions have been made based on various morphological features and larval morphology, the systematics of Chthamalidae is controversial (Spears et al.，1994; Harris et al, 2000). Thus, in this thesis, I aim to study the intrafamilial relationships of family Chthamalidae from a new perspective, that is, by molecular means. Apart from inferring the phylogeny of the Chthamalidae, this study also enables us to answer some of the questions in barnacle evolution, such as the evolution of partite plates in acom barnacles. Members in the family Chthamalidae are important to intertidal ecology and have been frequently used as model in phylogeography. Studying the genetic population structure of these intertidal barnacles helps show how different factors, such as

1 Pleistocene glaciations and ocean currents affect the evolutionary history of intertidal organisms. The population genetic study on Chthamalus malayensis (Tsang et al, 2008) in the Indo-West Pacific region has highlighted the role of Pleistocene glaciations in leading to genetic differentiation between lineages in this species complex, and this result is concordant to other findings in the region (Lessios et al., 2003). However, in the Northwest Pacific (NWP) region, even though there are well documented geological records and detailed oceanographic studies, the role of Pleistocene glaciations or other environmental factors on the evolutionary history of marine organisms are largely unknown, owning to a general lack of population genetic studies in this region, particularly on island fauna. Considering the geography of the Northwest Pacific region, which comprises numerous oceanic islands, Chthamalus moro appears to be a suitable model as an island species in studying the phylogeography in NWP. C. moro is a widespread barnacle species that is commonly found in tropical and subtropical islands of NWP. By correlating the oceanographic patterns and historical events to the population structure of this island inhabitant, this study is one of the very few works that demonstrate the roles of environmental factors in shaping the population structure of intertidal species in the island chain of NWP. This thesis comprises five chapters. Chapter 1 is an overview of this thesis, which includes short introduction to each chapter. Chapter 2 is a literature review that is composed of a brief introduction of barnacles, and a review on barnacle phylogenetics and population genetics. Chapters 3 and 4 present the research that I have done on the phylogeny of the Chthamalidae and the phylogeography of Chthamalus moro respectively. Chapter, 5 is a general conclusion on the findings from the two studies.

2 Chapter 2

Literature review

2.1 Introduction 2.1.1 General introduction to barnacles Barnacles are a group of organisms that live exclusively in marine habitats. They are an ancient group of and the oldest barnacle fossil can be dated back to Mid- epoch, approximately 530 million years ago (see P^rez- Losada et al.，2004). There are more than 800 thoracican barnacle species recorded worldwide (Foster and Buckeridge, 1987)，and they are observed in very diverse habitats in the marine realm, ranging from habitats of low salinity such as brackish waters in estuaries (Gomes-Filho et al, 2010), to habitats with very extreme temperature and salinity fluctuation and strong wave exposure, such as the intertidal shores (Crisp et al, 1981). Barnacles can be found in the deep sea 5000 m below the sea surface (Buhl-Mortensen and Hoeg, 2006), or next to hydrothermal vents splashing liquid over 100°C (Newman 1979; Buckeridge, 2000). Thoracican barnacles can also be found on a wide variety of substrata, such as shells of mollusks, the carapace of crabs or even spines of the sea urchin. Some barnacle species tend to live on corals (Ross and Newman, 2000)，sea turtles (Frick et al, 1998)，on the skin of whales and dolphins (Bane and Zullo, 1980)，or even man-made products like glass and plastic bottles. The high flexibility on habitat choice allows barnacles to occupy a wide range of ecological niches. 2.1.2 Classification of Barnacles Barnacles are under the subphylum Crustacea. It is classified into the class , subclass Thecostraca and the infraclass Cirripedia. The word Cirripedia means 'curly' 'legs', referring to a very distinct characteristic of a typical barnacle: the curly cirri for feeding. Barnacles are further divided into three superorders: Rhizocephala, Acrothoracica and . There are 40 families of Cirripedia, of which 28 belong to the Thoracica, three families are under Acrothoracica and nine families are under Rhizocephala (Martin and Davis, 2001). The superorders Rhizocephala and Acrothoracica are composed of parasitic barnacles and burrowing barnacles respectively, and they are not discussed thoroughly here. The superorder Thoracica harbors the majority of barnacle species, including those commonly found on intertidal rocky shores (Guerry et al., 2009)，mangroves (Perry, 1988) and estuaries (Dean and Hurd, 1980). Thoracican barnacles can be further classified into two orders according to their external morphologies. Acom barnacles (order ) usually bear conical shells (flattened in some cases), and encrust firmly on the substratum. Stalked barnacles (order Pedunculata) attach to the substratum with a fleshy stalk, at the top of which is the parities covering the cirri of the barnacles.

2.13 Importance of barnacles Although barnacles are less well known to human than some other crustaceans, they play important roles in different aspects. Ecologically, barnacles are crucial to coastal marine ecosystem, especially in . intertidal habitats. Acom barnacles are regarded as one of the key occupiers on intertidal shores and they interact with other intertidal communities in various ways.

4 , Despite the fact that they facilitate some communities by serving as preys of predatory organisms such as snail (Jarrett, 2009), barnacles also benefit other intertidal organisms by creating microhabitats; empty barnacle shells, together with the gaps between densely populated barnacles with water trapped in between are oasis and shelters for tiny organisms such as gastropods which live on high shore levels by protecting them from desiccation and overheating. Yet barnacles can also impose negative effect on other organisms. For example, barnacles compete with other intertidal settlers for space (Zabin, 2009). Occasionally, barnacles act as ectoparasite and grow on the shells or surface of whales, sea turtles or crabs, which may add load on the host. Barnacles in superorder Rhizocephala are parasites of crabs, will certainly harm the host by getting nutrients from them (Alvarez et al., 1995). Economically, barnacles are the most common biofouling organisms, and can cause billion-dollar loss per year. Some barnacle species, such as Balanus amphitrite, can settle and grow on man-made structure such as ship hulls or fences in harbors (Callow and Callow, 2002). Biofouling of barnacles can induce serious problems in marine transportation, as they increase the roughness of ship's hull and hence, increase water resistance, slow down vessels, and boost up the expense on ftiel. Also, biofouling by barnacles can speed up rusting of ships' hull. Vast amount of money is input every year in shipping sector for extra fuel expense, ship hull cleaning and repairing due to biofouling (Callow and Callow, 2002). And the US Navy spends more than 1 billion US dollars per year to tackle with biofouling problem (Callow and Callow, 2002). . Barnacles are for long popular experimental models in the field of biology. Barnacles were first thoroughly studied by Charles Darwin in 19 century. His work

5 on barnacles, summarized in four volumes of monographs on extant and fossil Cirripedia, has established the basic classification scheme of this group (Darwin 1851; Darwin, 1854). More important than barnacle taxonomy is that Darwin's eight-year study on barnacles contributes a lot to his later and the very influential publication 'On the Origin of Species' (Darwin, 1859). As he mentioned in his autobiography, the classification work on highly variable barnacles is useful in discussing the principle of natural classification in the seminal book. Barnacles are also used as model organism in demonstrating ecological theories and environmental problems. For instance, barnacles in the genus Chthamalus have been used to demonstrate the effect of interspecific competition (Connell, 1961). Barnacles are also used to study ions uptake and are potential bioindicators for heavy metal pollution (Rainbow and White, 1990). Apart from basic scientific studies, research on barnacles becomes more technical and practical. Many efforts have been made in studying anti-biofouling of barnacles, including researches on the mechanism of barnacle larval settlement (Bielecki et al, 2009), the chemical nature of cement secreted by barnacles during settlement (Wiegemaim, 2005) and potential anti-fouling coatings (Murosaki et al, 2009; Zhou et al, 2009). The cement of barnacles draws scientists' attention, as it is known to be the strongest and the most durable natural adhesive in aquatic environment (Wiegemann, 2005). Studies have been carried out to seek the potential technical application of this natural adhesive.

2.2 Molecular phylogenetics of barnacles 2.2.1 What is phylogenetics There are millions of organisms on earth, ranging from bacteria to mammals,

6 each possessing very unique features. It is almost impossible to believe that all lives arise from a common ancestor. Understanding the evolution of life is one of the ultimate goals, yet it is also one of the most challenging issues in biology. The desire in uncovering the connection between species and tracing the history of life give birth to phylogenetics, the study of evolutionary relationship between organisms. Unlike traditional classification, which aims at naming and cataloging different organisms, phylogenetics bears a deeper meaning behind: to elucidate the similarities and differences between taxa，in order to understand the origin of different species and their evolutionary history. The history of phylogenetics can be traced back to late century, when Haeckel first mentioned the word 'phylogeny', i.e. the history of a race, in his work The Riddle of the Universe at the Close of the Nineteenth Century (Haeckel, 1901). Since then, this term and its derivative, the phylogenetics, have been modified, elaborated, and widely used in the field of biology. Phylogenetics is employed to describe the studies which aim at understanding the evolutionarily relationship of various taxa, such as insects (Kjer, 2004) and mammals (Murphy et al” 2001)，and more important to our kind, human being (Goldstein et al, 1995). These studies help us understand the linkages between different human races, and the evolutionary history of human.

2.2.2 Phylogenetic studies on barnacles The studies on the phylogenetic relationship of barnacles can be traced back to the beginning of 20"^ century, when Ruedemann (1918) published the article ‘The “ phylogeny of the acom barnacle' and discussed the evolutionary relationship between barnacles using shell plate configuration. However, lacking strong evidence

7

V ‘ and precise inference make the phylogenetic study at that time speculative and superficial. Despite that several studies after Ruedemann (1918) attempted to reveal the evolutionary relationships of cirripeds, most of them can barely be regarded as phylogenetics, as they mainly focused on taxonomy and classification only (Newman and Ross, 1976; Klepal 1985; Zullo, 1992). More importantly, some of these studies allowed paraphyletic groupings in the classification scheme (e.g. Anderson, 1994). Thus, the investigation on cirriped phylogeny has actually come to a comma after Ruedemann's publication. Important breakthroughs in phylogenetics of barnacles were made in 1990s, when workers such as Glenner and Spears tried to reveal barnacle phylogeny using morphological and molecular means respectively (Glenner et al., 1995; Spears et al, 1994). These two studies are hallmarks to recent phylogenetic studies on barnacles, since they were the first two analyses followed a cladistic approach. Though with very divergent methods, both studies tried to infer barnacle phylogeny by novel means. Glenner et al. (1995) incorporated 32 morphological characters from different aspects, instead of including solely shell plate configuration in morphological phylogenetic analysis and Spears's team was the first group to study barnacle phylogeny using DNA sequences and the inferred secondary structures of nuclear 18S RNA gene region (Spears et al., 1994). These two analyses, when compared with the previous phylogenetic studies, were more conscientious and detailed, and thus provided more promising as well as compelling results. While both morphology and molecular characters were used to reconstruct • the phylogeny of barnacles in 1990s, the role of morphological characters has become less important in recent years (the pros and cons of using morphological

8 characters in phylogenetic studies will be discussed in the next session), and they usually serve as supplementary information in recent phylogenetic studies of barnacles (Glenner and Hebsgaard, 2006; Perez-Losada et al, 2004). In contrast, the incorporation of molecular markers, the nuclear 18S RNA regions in particular, becomes a common practice in reconstructing the phylogeny in Cirripedia. For instance, analyses using 18S RNA region on thoracican barnacles have shown that the order Pedunculata, commonly known as stalked barnacles, is actually an artificial assemblage with two groups of independently evolved barnacles (Harris et al., 2000). The 18S RNA gene was also applied to elucidate the relationship in lower taxonomic levels, i.e. the phylogenetic relationship between suborder and Verrucomorpha (Perl-Treves et al., 2000). Although these studies have contributed a lot to our understanding to barnacle phylogeny, they are not without limitations. The results obtained from these earlier studies were all based on a single molecular marker, and more importantly, the taxon coverage is not adequate, which may lead to a misleading conclusion. For example, Spears et al. (1994) attempted to study the thoracican phylogeny using 18S RNA by including only eight taxa, and suggested that the suborder Pedunculata is a monophyletic group. Yet, when Harris et al. (2000) reexamined the thoracican phylogeny using the same gene with twice the number of taxa, the result contradicted to the Spears's study, as the order Pedunculata is ofpolyphyletic origin. Since poor taxon coverage or the use of single molecular marker may lead to misleading conclusions, in the later phylogenetic studies, there is an increase in both the numbers of taxa and molecular markers used (Perez-Losada et al, 2004; Perez- . Losada et al, 2008). In the latest and as well the most comprehensive phylogenetic study of Thoracica, 76 thoracican barnacle taxa, three genetic markers and 44

9 morphological characters were incorporated in the study. This comprehensive study gives a clearer picture on the evolutionary relationships between different barnacle suborders, and also answers many fundamental questions in barnacle evolution, such as the time of radiation in thoracican barnacles, and the implication of shell plates in barnacle evolution (Perez-Losada et al., 2008). Despite the number of phylogenetic studies on Cirripedia is growing in recent decades, a majority of these studies aim at elucidating phylogenetic relationship between groups of high taxonomic ranks, between orders, or suborders, for instance (Perez-Losada et al, 2008). These studies seldom included many closely related taxa as representatives, and the resolution of the genetic markers used may not be appropriate in studying phylogenetic relationship at lower taxonomic levels, such as generic level. Meanwhile, there are very little researches that primarily focus on barnacle phylogeny at lower taxonomic levels. Most of our knowledge of the phylogenetic relationships within different barnacle families or genera is largely based on previous taxonomic studies. However, the high morphological plasticity exhibited by various barnacle groups has imposed uncertainty to morphological systematics (Simon- Blecher et al., 2007). So far, only two published works attempt to elucidate phylogenetic relationship of barnacle at the low taxonomic levels. One of the studies aimed at inferring the intrafamilial relationship of coral inhabiting barnacles in family Pyrgomatidae (Simon-Blecher et al, 2007). Using the mitochondrial 12S and 16S markers, which are informative at low taxonomic level studies, together with nuclear 18S segment, the study suggested the paraphyletic origin of this family of • barnacles. Another phylogenetic study that focuses on genus Chthamalus has shown that the outgroups Microeuraphia spp.，which was assumed to be the sister taxon of

10 Chthamalus, clustered within Chthamalus spp (Wares et al, 2009). Both the studies have pointed out one thing in common, that traditional classification based on morphology can be problematic, and detailed reexamination of different groups with molecular analysis are highly desirable. To conclude, the phylogenetic study of barnacles is growing rapidly in the recent decade. The multi-character approaches in both morphological and molecular studies, and in addition, the combination of morphological and molecular phylogenetics helps answer many fundamental questions in barnacle evolution, and address many hypotheses that cannot be tested previously. However, most of these phylogenetic studies focused on solving problems that are of general interests, and little attention was paid to phylogenetic relationship of lower taxonomic groups. Thus, more comprehensive phylogenetic studies are needed for different groups of barnacles, especially on familial and generic level, in order to enhance our understanding towards this group of ancient crustaceans from a new perspective.

2.2.3 Choices on characters for barnacle phylogenetics: Morphological characters vs. molecular characters Morphological characters have for long been widely used in phylogenetic studies of various taxa (e.g. plants: Doyle and Donoghue, 1986, fish: Lauder and Liem, 1983，birds: Cracraft，1988). They also play an important role in studying barnacle phylogeny, especially in earlier ages (Ruedemann, 1918). Several characters have been incorporated into the phylogenetic studies of barnacles. Features such as scutum and tergum morphology, and shapes of madibular “ structures are commonly used in inferring barnacle phylogeny (Perez-Losada et al., 2004). Other features such as sperm morphology are also applied in phylogenetic

11 studies of barnacles (Healy and Anderson, 1990). A major advantage of incorporating morphological characters in cirriped phylogenetic studies is that it allows comparisons between both extant and extinct barnacle species. This advantage is useful as the extant barnacle species only represent a small portion of the overall cirriped species that were once present on earth, and the fossil record of barnacles, especially thoracican barnacles is rich because of their mineralized shells (Buckeridge and Newman, 1992). Morphological phylogeny is undoubtedly crucial in building up our understanding towards the evolutionary relationship between different barnacle species. However, the utilization of morphology solely in inferring barnacle phylogeny has been questioned. It is because morphological characters such as shell plate morphology often show various degree of plasticity, probably in response to different external factors (Lively, 1986; Jarrett, 2009). Besides, even though some of the morphological characters are distinct in nature, such as the presence or absence of a certain character, many of the morphological characters are of continuous variations, such as size of barnacle (Chan and Williams, 2004). The high variation between individuals in morphological characters may make phylogenetic analyses complicated by imposing too many variables in the studies. And these can only be analyzed though complicated statistical models, strong computational devices and large sample size, and greatly increase the effort needed. Moreover, the choice of morphological characters by barnacle taxonomists in phylogenetic studies can be subjective. Take Chthamalidae barnacles as an example, it was suggested by Zullo (1963) that the number of teeth on mandible could serve as • a diagnostic feature in subfamilial level, with the members in subfamily Euraphiinae possessing three teeth while barnacles in subfamily Chthamalinae containing four

12 teeth. However, Pope (1965) criticized the validity of this character based on the observation that some of the individuals that should possess four teeth bear only three, and vice versa (Pope, 1965). Last but not least, there are limited morphological characters available for reconstructing phylogeny (Avise, 2004). Taking the studies on barnacle phylogeny as examples, the hallmark phylogenetic study on cirriped phylogeny employs 44 morphological characters (Perez-Losada et al., 2008) and another extensive study that focuses on the thoracican phylogeny utilizes 37 morphological characters together with 53 multistate larval morphological characters (Perez-Losada et al” 2004). Although these two comprehensive studies involve quite a number of morphological characters, just a single molecular marker, Histone 3 for instance, can provide more than 300 base pairs for analysis, and each base pair can simply be regarded as a single character for phylogenetic study. This also makes reconstructing phylogeny using morphological characters alone a less favorable choice. In contrast to morphological phylogenetics, molecular phylogenetics is growing more popular in these two decades. Various kinds of molecular data, such as DNA sequences of numerous genes have been incorporated in reconstructing barnacle phylogeny (Spear et al, 1994; Harris et al, 2000; Perez-Losada et al, 2004), and molecular information is actually dominating the field in recent years. There are a number of advantages for using molecular data over morphological characters in phylogenetic studies. The largest advantage goes to the availability of characters. Not like morphology, which has limited number of external characters for analysis, a genome can theoretically provide enormous molecular characters for studying phylogeny (Hillis, 1987), as every base pair in a gene simply represents a / character that can be used for phylogenetic comparisons.

13 The role of molecular data becomes particularly important in cross comparisons between evolutionarily very divergent lineages. This is because the earlier the taxa diverged, the less similarities they are likely to share. Thus, it is difficult in finding morphological characters in common for phylogenetic studies in ancient taxa. Take parasitic barnacles (Rhizocephala) and acom barnacles (Thoracica) as examples, even though both the superorders belong to the class Cirripedia, members in these two superorders are very diverse in both morphology and biology. The acom barnacles are free living and possess calcified shell plates, while members in Rhizocephala are parasitic in nature and most of the body structures, including the shell plates are greatly reduced. The two barnacles share very little similarities and can barely be compared in terms of morphology. Even though the two kinds of barnacles may not possess homologous structures, they bear one thing in common, that is, a genome composed of DNAs. And within the genome, part of it is believed to be ancient and conserved, and can be observed in most of the life forms. Thus, molecular data are precious source in providing homologous genetic characters in phylogenetic studies of high taxonomic level, between phyla for instance (Field et al, 1988). Lastly, the popularization of molecular phylogenetics, together with the advance in sequencing technologies in recent years have led to an accumulation of enormous amount of sequence data, and these help build a comprehensive dataset for reconstructing barnacle phylogeny (e.g. Harris et al, 2000; Perez-Losada et al., 2004).

2.2.4 Choices of molecular markers in phylogenetic and population genetic studies on barnacles

14 Mitochondrial markers cells possess two sets of genomes, a nuclear genome that is located i inside the nucleus of the cell, and, a mitochondrial (mt) genome that is found within i mitochondrion, an organelle that is responsible for cellular respiration and energy � 3 > conversion. Rather distinct from the nuclear genome, the mitochondrial genome is in 1 circular form, comprises more than 10,000 base pairs and usually encoded for 22 tRNA genes, 13 protein-coding genes and 2 rRNA genes, 37 genes in total. In j crustaceans, the mt genome is usually greater than 14,000 bps in length. For | instance, the mt genome of the copepod Tigriopus japonicus is 14,628 bp (Machida et al., 2002), while the mt genome size of the stalked barnacle Pollicipes mitella is 14,915 bp (Lim and Hwang, 2006). i Various regions in the mt genome are often employed to infer the phylogeny of j different taxa. For instance, the mt 12S and 16S ribosomal RNA genes and the cytochrome c oxidase subunit I region (COI) are the most popular mitochondrial markers in elucidating phylogeny (Lavery et al, 2004; Hou et al, 2007; Harrison, 2004). Similarly, these genes are commonly used in inferring the phylogenetic relationship in barnacles, particularly at lower taxonomic levels, intrageneric level for instance (12S and 16S: Am et al, 2004; Simon-Blecher et al., 2007; COI: Fisher et al, 2004), The mitochondrial genes are commonly used in phylogenetic studies of barnacles based on a few reasons. Firstly, the mitochondrial genomes are transmitted to the next generation maternally in most of organisms (including barnacles), and without the involvement of paternal side. This feature helps limit the possibility for “ recombination of mtDNA and also the problem of multiple alleles. Therefore, the analyses on mtDNA remain relatively simple, comparing to that of nuclear DNA.

15 , Besides, mitochondria are organelles related to ATP generation, the free radicals produced during cellular respiration process, together with relaxed DNA repairing mechanism allow the mt genome to acuminate mutations in high rates (Brown et al., 1979; Wilson et al., 1985). Thus, mtDNA is an appropriate choice to resolve phylogenetic relationship at low taxonomic levels (Avise, 2000). Another feature that allows mt genes to be appropriate genetic markers in phylogenetic study is the large quantity of mtDNA. Although different from species and cell types, usually more than 80 mitochondria per cell is recorded in different eukaryotes, and an average of 200 to 1，700 mtDNA molecules was observed per cell .. 1 in mammals (Robin and Wong, 1988). The presence of numerous copies of mitochondria in different tissue types makes the extraction of mtDNA and • amplification of mt markers easier to achieve. More importantly, the complete mitochondrial genomes of barnacle have been sequenced (Lim and Hwang, 2006). Thus, less effort is needed in designing primers for genetic analysis when using the mt genome in phylogenetic studies. Although mitochondrial regions seem to be powerful markers in phylogenetic studies, they are not without limitations. First of all, the high mutation rate together with small genome size in mitochondrial genomes is likely to make the issues such as multiple hits obvious. These issues are relatively insignificant in inferring low- level barnacle phylogeny, such as the phylogenetic relationship between and within subgroups in the genus Chthamalus (Wares et al, 2009). However, when mitochondrial markers are used in elucidating phylogenetic relationship of higher taxonomic levels or genetically diverged groups, problems such as saturation of the “ third positions become evident (Wares, 2001; Fisher et al., 2004) and the ability for mitochondrial markers in resolving high level barnacle phylogeny is greatly

16 , depleted. As mentioned previously, mt genomes are maternally inherited. Thus, genetic analysis on mt genomes only gives information on evolutionary history of female lineage, whereas the phylogenies of paternal lineages, or events such as hybridization or introgression remain unidentified. The complete evolutionary history of the target organisms cannot be fully elucidated if only mt markers are used in phylogenetic studies. Moreover, even though the mt genome comprises more than 10,000 base pairs and 37 genes in total, some workers suggest that the whole mt genome can only serve as a single genetic marker in phylogenetic analyses (Harrsion, 1989). It is because very little recombination takes place in mitochondrial genome, and all the mt genes are linked and inherited as a whole, making the evolutionary histories of different mt genes more or less similar.

Nuclear markers Although phylogenetic studies in the past two decades largely rely on mitochondrial markers (Hurst and Jiggins, 2005), the roles of nuclear markers in this field are growing more and more important nowadays. Various nuclear regions were used in the studies of barnacle phylogeny. Among those, 18S RNA gene and histone 3 region were the most commonly used markers in revealing barnacle phylogeny at high taxonomic levels (Spears et al., 1994; Mizrahi et al., 1998; Harris et al., 2000; Perez-Losada et al., 2004; Perez-Losada et al, 2008). Other gene regions, such as elongation factor I a region (Wares et al, 2009) and 28S RNA gene (Perez-Losada et al., 2004) were also employed in phylogenetic studies in barnacles. An obvious advantage of obtaining molecular markers from nuclear genome is the enormous size of nuclear genome. Take human genome as an example, the

17 ,

V ‘ mitochondrial genome is around 15,000 base pairs in size but the nuclear genome comprises more than 3 billion base pairs (Venter et al, 2001)，which is 200,000 times larger than that of mitochondrial genome. With such an enormous size, the nuclear genome theoretically can provide plenty of markers with different degree of variability, for which are suitable for phylogenetic studies at all taxonomic levels. The nuclear genome harbors most of the protein coding genes, and many of the protein coding genes have strong functional constraint, and so, are in general highly invariable across different taxa. Apart from protein coding regions, the majority of the nuclear genome actually comprises non-coding DNAs such as introns, and repetitive sequences. These regions have little or no function in protein production, thus can have a higher genetics variability and evolutionary rate. In addition, these non-coding regions are believed to be selectively neutral, since they usually have little functional constraints. These features allow the non-coding nuclear region to be potential markers in elucidating the phylogenetic relationships in lower taxonomic levels. However, the flexibility of these non-coding markers on the other hand imposes difficulty in data interpretation. Lacking of functional constraints makes changes such as insertion and deletion common in non-coding region and causes sequence ambiguity during analyses. Apart from complicated sequence alignment, the role of indels in reconstructing phylogeny remains controversial. It is because most of the statistical analysis involved in phylogenetic studies is designed for nucleotide sequence comparison, but do not support the use of other genomic changes as characters. . Although some workers suggested indels as the fifth character besides the four types of nucleotides in pHylogenetic studies (Simmons and Ochoterena, 2000)，others have

18 pointed out that the generation of indels is of different mechanism to that of base substitutions. The evolutionary meaning behind indels and their relative weighing in phylogeny reconstruction so far have no definite consensus (Zhang and Hewitt, 2003). And statistical analysis for these genomic changes cannot be designed if the forces that lead to these changes are not made known (Rokas and Holland, 2000). The diploid nature of nuclear genome also brings advantages in phylogenetic studies. Since the nuclear genetic materials are inherited from both parents, this feature allows the studies on both the paternal and maternal lineages, and also helps find out if hybridization or historical introgression has occurred. These phenomena cannot be achieved by haploid markers such as mt gene, and sex chromosome in human. However, the diploid nuclear loci can frequently be heterozygous. The presence of two haplotypes within an individual will complicate the studies and extra effort is needed to isolate the two alleles from heterozygous individuals prior to sequence analysis. Moreover, recombination events (e.g. crossing over in meiosis, hybridization) frequently occur in diploid nuclear genome. These events may mask the true phylogeny between taxa and hence produces misleading result in phylogenetic studies (Posada and Crandall，2001). In contrast to the mitochondrial genome that is commonly regarded as a single inherited unit, different loci in the nuclear genome represent independent evolutionary lineages. This is because nuclear loci are scattered throughout the whole genome on different chromosomes. These loci are usually subjected to different degree of selection pressures and evolutionary rates, and may not be • necessarily inherited to the next generation with other nuclear markers. Thus, the incorporation of miilti-loci with independent evolutionary histories in phylogenetic

19

v" ‘ studies is believed to reduce bias and infer the true phylogeny oftaxa studied. Although nuclear markers appears to promising and are highly recommended for phylogenetic studies in various taxa (e.g., turtle: Barley et al.，2010; ray finned fish: Li et al., 2007), the utilization of nuclear markers in barnacle phylogeny have just started (with 18S as an exception). The largest problem faced by molecular phylogenetics of barnacles goes to the lack of information of the nuclear genome in this group of organisms. Even though there are theoretically numerous markers available for phylogenetic studies, choices on appropriate markers require background knowledge on the nuclear genome. So far, no complete nuclear genomes of barnacle have been sequenced, and information on barnacle nuclear" genome is very limited. The design of nuclear markers for barnacle phylogenetics, ranging from traditionally used 18S RNA marker, to those newly developed markers such as NAKAS and LTRS (Wares et al., 2009), were usually modified from the published nuclear regions of crustaceans (Spears et al., 1994; Wares et al, 2009). This largely diminished the amount and the types of nuclear markers that can be used in phylogenetic studies of barnacles. Take the 18S RNA marker as an example, this nuclear gene was chosen as it is conserved throughout a wide range of taxa, at least in crustaceans (Kim and Abele, 1990). However, since the 18S RNA marker was originally designed to resolve infraorder relationship in crustaceans, the resolution may not be high enough to reconstruct the phylogeny of Cirripedia (Harris et al, 2000). Moreover, as these nuclear markers are often designed based on nuclear region of other crustaceans that are distantly related to barnacles, more efforts are needed on optimization and modification of the primers, before the target regions ., can be successfully amplified.

20 2.3 The use of barnacles as model organism in population genetic studies Population genetics is the study on genetic variations, allele frequencies and the genetic structure of populations. The studies on population genetics are of growingly importance in the field of biology. These studies have enhanced our knowledge on the roles of intrinsic factors, such as life history (Palumbi, 1994)，niche preference (Reid et al, 2006), and extrinsic factors like the presence of geographical barriers (Perdices et al., 2002) in shaping the population history of organisms. A number of genetic studies have been conducted in various terrestrial and freshwater animal and plant populations (Tzeng et al” 2006; Ouborg et. al., 1999), and many of them have pointed to a fact that population structuring are common in terrestrial organisms and freshwater fauna. This is likely because these populations can be easily fragmented by the presence of mountain ranges, rivers and oceans (Perdices et al., 2002; Funk et al., 2005). When compared to terrestrial and freshwater taxa, however, the amount of population genetic studies regarding marine organisms still lag behind, and the possible causes for population structuring in marine habitats are not well studied. One the other hand, high connectivity between different marine habitats, together with very little prominent barriers such as mountains that block gene flow between populations may also give an impression that some marine organisms may have little population structuring (Knutsen et al. 2003). Yet with the advance in molecular techniques, a growing number of researches have shown that population structuring is actually not rare in marine environment. . Among these studies, barnacle is often chosen as the model organism in population genetics in marine systems (Wares and Cunningham, 2001; York et al, 2008). It is

21 because many barnacle species are key occupiers in different marine ecosystems, such as rocky intertidal and mangrove (Crisp et al., 1981; Guerry et al, 2009; Perry, 1988). The high prevalence and ecological importance allow them to be representatives of the communities. Besides, barnacles exhibit pelagic larval stage that passively transported by ocean currents and tides (Sotka et al, 2004)，so that this feature allows the studies on the effect of ocean circulation patterns on genetic structure (Quinteiro et al, 2007; York et al., 2008). In the next sections, with barnacles as examples, I will discuss different forces in shaping population structure in marine habitats.

2.3.1 Pleistocene glaciations The Pleistocene glaciations are probably the most well known event that causes population diversification and speciation in the recent time scale (McManus, 1985). It is remarkable for the numerous glacial events and dramatic decline in global temperature (Richmond et al, 1986). Despite the well-known fact that the change in climate and onset of glaciations affected the population structure of terrestrial fauna and flora (Hewitt, 1996), they also greatly altered the population structure of organisms in marine habitats. A recent study on the population genetics of the stalked barnacle Pollicipes pollicipes throughout the European coastal region detects a sign of rapid demographic expansion from a small historical population, and the event is dated back to late Pleistocene. The adverse climate together with the glacial cycles during the Pleistocene is believed to contribute to not only the genetic pattern, but also the contemporary distribution range of this species (Campo et al., .,

2010). Apart from drastic climatic change, during Pleistocene glaciations, the ice caps

22 in the two polar regions enlarged dramatically. Vast amount of water in the marine system was locked in the massive ice caps, leading to the lowering of global sea ‘ level. The sea level dropped to 120-140 m below contemporary sea level during the last glacial maxima (LGM) (Voris, 2000). These, in turn, caused the exposure of landmass in different parts of the world, such as the Sunda Shelf in Indo-Malay region (Voris, 2000)，and the formation of a land bridge that connected Tasmania and Southern Australia (York et al., 2008). Analyses on the mitochondrial cytochrome c oxidase subunit I (COI) and control region on Australian surf barnacle Catomerus Polymerus has clearly shown that the closure of the Bass strait between mainland Australia and Tasmania during LGM is responsible for the genetic differentiation between eastern and western Australian C. polymerus populations (York et al, 2008). This geographical barrier probably caused population fragmentation in the ancestral population of C, polymerus. The lack of gene flow between populations, together with small population size made effects like genetic drift and inbreeding significant, and these finally induce population differentiation in the surf barnacle.

2.3.2 Oceanographic pattern and habitat availability In addition to historical events such as glaciations, ocean current patterns can also generate or maintain population structure in marine organisms. For example, in P. pollicipes, the northward flowing Iberian Poleward Current on the East Atlantic \ coast helps homogenize population structure between European populations, and the eastward flowing Canary Current on the other hand facilitates gene flow between the two African populations (Quinteiro et al, 2007). The opposite direction of the two .. current systems together imposes restriction on gene flow between the European and African populations; Analysis on microsatellites in the Australian C. polymerus

23 populations also shows the role of ocean currents in inducing population structuring (York et al., 2008). The East Australian Current in maintaining gene flow between eastern coast populations, while the Zeehan Current is responsible for transport barnacle larvae from mainland Australia to Tasmania. The presence of unsuitable habitats within the distribution range can also prohibit gene flow between populations. For instance, in Southern Australia, the presence of sandy Ninety Mile beach is unsuitable for barnacle settlement, and contributes to the mere structuring between central Victoria and eastern Australian C. polymerus populations (York et al., 2008). Despite the ability of population genetic study in elucidating the environmental forces affecting the evolutionary histories of organisms (Cavalli- Sforza, 1998), study on population structure of barnacles can also provide us with information in other aspects. For instance, by studying the population structure of the Chthamalus proteus, an alien barnacle species in the Pacific region, Zardus and Hadfield (2005) are able to identify that the invasion of this barnacle species is not a single event. Instead, the Chthamalus barnacle is likely to be introduced from Brazil and Panama to the Pacific for several times, probably through the shipping between Carribean and Pacific waters (Zardus and Hadfield, 2005). On the other hand, study on the population structure of the turtle barnacle Chelonibia testudinaria on logger-head sea turtle provides evidence on host- passenger interaction. It was shown that the migration pattern, such as route of migration of host turtles, was fundamental in mediating the population structure of this group of barnacles (Rawson et al, 2003). . To conclude, investigating the population genetics of barnacles enable us to leam more about'how populations of marine organisms respond to different

24 historical events and environmental factors. It also helps reveal the population history and trace back the origins of a population, or allows the study of interactions between different marine taxa.

25 Chapter 3

Phylogenetic relationship of barnacles in family Chthamalidae

3.1 Introduction Barnacles in the family Chthamalidae are widely distributed in shallow waters of the temperate and tropical regions (Newman and Ross, 1976; Southward and Newman, 2003). A majority of the members are key occupiers on intertidal rocky shores and mangroves, and are of high regional abundance (Navarrete and Castilla, 1990; Lopez et al, 2007). The high ecological value and abundance make them suitable candidates for testing such ecological hypotheses as interspecific competition (Connell, 1961). These barnacles can also serve as model organisms in environmental science, to demonstrate the effects on worldwide climatic change (Southward, 1991; Barry et al, 1995), for instance. From the evolutionary point of view, the phylogeny of this group of barnacles is of great interest. They are believed to be the most primitive acorn barnacles and could provide valuable information to reveal the mode and pattern of barnacle evolution during the transition from stalked barnacles to symmetric acom barnacles (Darwin, 1854; Newman and Ross, 1976). Even though the phylogeny of barnacles is important in different aspects, it has remained obscure and unsettled for more than 150 years since Charles Darwin presented the first systematic investigation of the group (Darwin, 1854). The family Chthamalidae currently comprises 15 genera in three subfamilies with more than 40 nominal species. Yet at the time Darwin first established the family Chthamalic^e under the superfamily (Darwin, 1854), this

26 family only comprised one subfamily, the Chthamalinae (Fig. 3.1). Many revisions have been made to the taxonomy of this family in recent decades. Utinomi (1968) added a new subfamily Pachylasminae into the family Chthamalidae, which harbored a group of Chthamalidae that inhabits the deep sea and possesses eight shell plates. Newman and Ross (1976) placed the genera Euraphia and Octomeris, which were originally under the subfamily Chthamalinae, in a novel subfamily Euraphiinae as they both possess calcareous basis and tridentoid mandible. Soon after this, in a major revision made in 1978, the subfamily Pachylasminae was removed from the Chthamalidae and raised to a full familial status (Foster, 1978). Therefore, the Chthamalidae only comprised two subfamilies of which all members are from shallow waters. The last major revision was made in 1987，in which the new subfamily Notochthamalinae was erected in the Chthamalidae (Foster and Newman, 1987), with some rearrangements in generic level that followed subsequently (Fisher et al., 2004). In this currently adopted classification scheme, the Chthamalidae are divided into three subfamilies, Chthamalinae, Euraphiinae and Notochthamalinae. The subdivision of Chthamalidae is largely based on the mandibular structure and shell plate morphology. These features are believed to reflect the trend of specialization during the evolution of this family. Furthermore, Ross (1971) and Newman (1987) proposed that there is a trend of gradual loss of shell plates with time in the Chthamalidae based on fossil records. Accordingly, the most primitive members in the Chthamalidae bear the greatest number of shell plates while the most advanced forms bear the least, as suggested by plate morphology and the fossil record of Chthamalidae (Newman, 1987). This phylogenetic hypothesis, however, has not been tested-using a cladistic approach and the utility of those morphological

27 features in phylogenetic inference has been challenged (Pope, 1965; Dando and Southward 1980). Pilsbry (1916) made the first attempt to infer the phylogenetic relationship of this group using the number of teeth in the mandible. This character has been used in subsequent studies as a diagnostic feature within the Chthamalidae (Nilsson-Cantell, 1921; Zullo, 1963). On the other hand, Pope (1965) noted that intraspecific variation exists in mandible morphology and the number of teeth in the mandible could also vary among individuals of the same species. Other generic diagnostic characters, such as shape and position of adductor muscle scar on opercular plates were also reported to be variable within a species (Dando and Southward, 1980). These variations pose questions on the validity of using these characters in phylogenetic inference. The high plasticity of shell morphology exhibited by the Chthamalidae barnacles further complicates the taxonomic problem. Darwin (1854) pointed out that the shell morphology of different chthamalids is highly plastic and they share high similarities with each others. Extrinsic factors play an important role in determining Chthamalidae shell morphology. For example, predation by gastropods can exert pressure to the survival of barnacles and induce shell polymorphism (Lively, 1986). Moreover, different degrees of wave impact can cause variations on cirral morphology (Chan and Hung, 2005). These challenge the taxonomy in this group of barnacles, as shell and cirral morphologies are the principal characters for early barnacle taxonomy. Most of the past systematic studies on barnacles were mainly based on morphological and anatomical analyses. Recently, molecular techniques have been .. incorporated into phylogenetic studies of the Chthamalidae (e.g. Fisher et al., 2004; Wares, 2001; Wares et al” 2009). For instance, a molecular phylogenetic study on

28 the Chthamalidae suggested the transfer of Octomeris from the subfamily Euraphiinae to the Notochthamalinae (Fisher et al., 2004). This study also rejects the hypothesis of stepwise plate reduction in Chthamalidae evolution. On the other hand, a molecular phylogenetic study of the most species rich genus Chthamalus (subfamily Chthamalinae) using four molecular markers reveals that it is not monophyletic, as Microeuraphia (subfamily Euraphiinae) is nested within Chthamalus (Wares et al, 2009). This suggests that further investigations and taxonomic revisions are needed in this family (Wares et al, 2009). Yet the phylogenetic relationships within the Chthamalidae have not been fully investigated in these molecular studies. In this study, I attempted to investigate the phylogenetic relationship among subfamilies and genera of the family Chthamalidae from a new perspective. With the incorporation of multiple molecular markers, the mitochondrial 12S ribosomal RNA (12S) gene, the nuclear histone 3 (H3) gene and the nuclear Elongation Factor 1 alpha (EFla) region, the result would be more informative and convincing than single-maker analysis. The loci chosen are responsible for structural functions, such as protein and rRNA coding, and are assumed to be independent from the morphological traits. These loci are highly conserved and suitable for phylogenetic study of higher taxa, such as on the intergeneric level. This would enable us to answer questions such as (1) if shell plate number polymorphism follows evolutionary trend, and (2) if there is a trend in the reduction in the number of shell plates, for how many times this occurred during evolution of the family. Results from this study could lay the foundation for further taxonomic revisions, which aim • at a natural classification of the Chthamalidae.

29 . TN,

Newmsn and Rosi 197fi Fiwiter _ Newman 1,S7 Family Chthamalidae Family Chtbamalidae sub&mily Chtbamaiinae subf^ily Chthamalmae siibf^ly Pachylasmiiiae subfamily Euraphiinae � subfimiily EnrapiiilBae sabfamity Notoeiithaiiialinae

18^4 等 “ 1968 1976 1978 1987

Darwin 1854 Utinonii 1968 Fi»ter 1978 Family Chthamalidae Family Chthamaiidae Family Pachylasmiciae Family Chthamalidae subfemily Chthamalinae subfemily Chthamalinae subCEunlly Pachylasmliuie subfamily Chthamalinae sub&mily Euraphiinae

Fig. 3.1 Changes in the classification scheme of thefemUy Chdiamalida e overtime. Bol d charactersrepresent majo rrevisions mad e to the family Chthamalidae.

30

• 3.2 Materials and methods 3.2.1 Sample collection, DNA extraction and amplification Barnacle samples were either directly collected in the field, or provided by collaborator Dr Benny Chan. Out of the 15 genera in the family Chthamalidae, 13 were included in this study, with one to three representatives from each genus, except that five species were included for the most species rich genus Chthamalus. As the samples from the sister group of the family Chthamalidae, Catomerus polymerus were not available, another barnacle from Balanomorpha, Amphibalanus amphitrite (family Balanidae) was used as the outgroup (Table 3.1). The samples were preserved in ethanol (70% or higher) until analysis. The adductor muscle or soft tissue from the abdomen were dissected and subjected to DNA extraction using QIAamp Tissue Kit (QIAGEN). All specimens were subjected to morphological identification prior to molecular analysis. Most of the samples could be identified to the species level, except the seven specimens that could only be identified to genus level due to morphological obscurity. To determine the species identity of the seven specimens, partial sequences of mitochondrial COI were amplified using universal primers (Folmer et al., 1994) and uploaded to the Basic Local Alignment Search Tool (BLAST) provided by the National Centre for Biotechnology Information (NCBI) for DNA barcoding purpose. These COI sequences were compared with existing sequences with known identity on the database. For sequences shared similarity higher than 98%, they would be classified as belonging to the same species as reported in database, while for those with similarity lower than 98%, the sequences were designed as unidentified species of . the genus.

31

» Table 3.1. Classification, sampling localities and molecular markers used in the present study. Rows shown in grey represent samples which have all

the three gene segments successfully amplified and were included in the phylogenetic analyses.

Subfamily Species Sampling locality COI H3 EFla Chthamalinae Chinochthamalus sciiteliformis Hong Kong, China .…… / f5Shani:linae》運丄'challen^;i Japan..…,- �、厂

Chthamalinae Chthamalus malayensis Mersing Island, Malaysia / / Chthamalinae Chthamalus maIc^'ensis (TW) Hualien, Taiwan / / / _ ^ _ __ fcbMrna^. f ^^g^v^^^icr^ir^：^•！：二ISS：怎 Chthamalinae Chthamalus moro Hualien, Taiwan / / / Chthamalinae Chthamalus neglectus Hong Kong, Chma … / / I Chthamalinae Chthamalus sp. 』 Sidney, Australia / . / / / \ r Chthamalinae Jehlius cirratus - Chile - '' / / ^ f Chthamalinae ‘ Hainan Island, China P / / / / ‘ I Tetrachthamalus‘ sinensis- • - , ；Notochthamalinae -Chamaesipho sp. . New Zealand , / / / / { Notochthamalinae HexechamaesipKo pilsbryi “： Hualien, Taiwan / , / ^ / ^ / ； [No^o^fflaij^lpAe- ^ Nesochthamalus inteftextus ^^�Oj^aw^' Japan^J^HL^^ Notochthamalinae ^ scabrosus ^^^C'hile : ‘ • - ‘ ^^^ \ Notochthamalinae Octomeris sp.^ “ ^^ Phuket^ Thailand , - 、 / / “ V . ' / Z 广j ‘ I Notochthinialinae Pseudooctomeris sulcala . ;•滅…..Taiwan. — 乂 • . 1 也4£S_」 Notochthamalinae Rehederella belyaevi Easter Island, Chile 了 riuraphiinae ' Gaudomfauhila caudata^�^^懸备i身ipi叩oi^e ——一 丄 ,Euraphiinae— ^Igg^g^印…)广）“ ——————— L.Eura'phima& ^— J^jcmeui^gphia yyiihersi l:i.L.Hong Kong, .Chin^. 二“-ji-jjJ^yLsi J

32 Four out of seven samples were successfully identified to the species level. They were ’ C. malayensis (Indo-Malay clade), C malayensis (South China Sea clade) and C. malayensis (Taiwan clade). The clade designation was based on a recent study on C, malayensis which showed that it is a species complex with three distinct lineages (Tsang et al, 2008). Since the GenBank database cannot provide sequences with similarities of higher than 98% for the remaining three specimens, they were identified only to the genus level {Chthamalus sp., Chamaesipho sp. and Octomeris sp.) according to morphological examination. Partial sequences of mitochondrial 12S gene, nuclear H3 and nuclear EFla region were amplified using the corresponding primer sets, i.e. 12S-FB and 12S-R2 for mitochondrial 12S gene (Tsang et al., 2009); H3-AF and H3-AR for nuclear H3 gene (Colgar et al., 1998)，and EFl-for (5，GATTTCATCAAGAACATGATCAC-3') and EFl-rev (5' AGCGGGGGGAAGTCGGTGAA-3') for the nuclear EFl region (designed in this study). The PGR amplifications of different markers were performed using the following protocol: IX PGR reaction buffer, 2 mM MgCl2, 200 nM of each primer, 200 |xM dNTPs, 1.5 units of Taq polymerase (Takara), 1 |xL of template DNA and ddHiO. The volume of each reaction mixture was 50 |xL. The PGR profile was as follows: 3 minutes of initial denaturation at 94°C, followed by 35 cycles of denaturation at 94°C for 30 seconds, annealing for 30 seconds at 48°C to 58°C (varied according to taxa and primers), elongation at 72°C for 30 seconds to 1 minute (depending on primers) and finishing with an extension at 72°C for 3 minutes. The products from PGR were subsequently purified using the QIAquick gel purification kit (QIAGEN). Sequences were generated using the same sets of primers and an Applied Biosystems (ABI) 3730A;/ automated sequencer. The ABI Big-dye Ready-Reaction mix kit was used in sequencing, following the standard cycle

33 sequencing protocol.

3.2.2 Phylogenetic analyses Alignment of sequences was performed using CLUSTAL W (Thompson et al., 1994). The DNA sequences were translated into amino acid sequences, as to confirm the alignments of H3 and EFla. The sequences of the three markers were combined for phylogenetic analyses using two different tree-building algorithms, maximum likelihood (ML) and Bayesian inference (BI). ML analysis was performed on PAUP * v4.0 blO (Swofford, 2002)，using the best-fit model of molecular evolution for the dataset as determined by Modeltest 3.7 (Posada and Crandall, 1998). The ML topologies were generated using heuristic search and tree-bisection-recoimection with 1,000 random addition sequence replicates. The support for each node on the ML tree was estimated from 1,000 bootstrap (BP) replicates, and each of them had one random addition sequence replicate. For BI analysis, it was performed using MrBayes v.3.12 (Ronquist and Huelsenback, 2003). In the analysis, two independent runs were preformed with four differentially heated Metropolis coupled Monte Carlo Markov Chains (MCMC Chain) for 5,000,000 generations that begins from a random tree. The model parameters for the analysis were calculated and the frequency of chain sampling equaled to 500 generations. Convergence of the analyses was confirmed by the standard deviation of split frequencies and monitoring the likelihood values over time graphically by incorporating Tracer vl.4 (Rambaut and Drummond, 2007). The trees prior to the achievement of stationarity of the log likelihood values (i.e. 2,000 trees) were discarded as bum-in. A 50% majority-rule . consensus tree was constructed from the remaining trees to estimate posterior probabilities (PP):

34 3.3 Results The aligned 12S, H3 and EFla region contained 502, 296 and 878 base pairs respectively. No insertion/deletion or stop codon was observed in the protein-coding H3 and EFla loci. The numbers of variable and parsimony informative sites are shown in Table 3.2. The concatenated dataset for phylogenetic analysis consisted of 1,676 characters from 16 taxa (including outgroup). Chinochthamalus scuteliformis, Rehderella belyaevi and Microeuraphia depressa were not included in subsequent phylogenetic analyses due to unsuccessfiil amplification of some or all the gene markers. In total, 13 out of 15 genera of Chthamalidae were collected and 11 genera were successfully amplified for molecular analyses. In the ML analysis (Fig. 3.2), BP values equal to or larger than 70% are regarded as a strong support, while PP values higher than 95% are regarded as statistically confident in the BI analysis (Fig. 3.3). The overall tree topologies from ML and BI analyses are similar. In both trees, Pseudoctomeris sulcata (subfamily Notochthamalinae) is in the basal position and does not cluster with the other Notochthamalinae, making this subfamily polyphyletic. The Chthamalidae taxa other than R silcata are divided into two major lineages and this is supported by both the BP and PR The first lineage comprises a majority of the remaining members from the Notochthamalinae, and also a member from subfamily Euraphiinae, Caudoeuraphia caudata. Yet inconsistency of tree topology between the two analyses, together with low statistical support within this lineage making the phylogenetic relationship within the group inconclusive. The other lineage consists of all members examined in the subfamily . Chthamalinae as well as Microeuraphia withersi from the subfamily Euraphiinae. As the latter is nested within the Chthamalinae, Chthamalinae is a paraphyletic

35 assemblage in its current definition. Euraphiinae is also revealed to be polyphyletic as the two representatives from the genera Caudoeuraphia and Microeuraphia are separately grouped into two different lineages in the phylogenetic trees. The relationships between members in the second lineage are better resolved than those in the first one, as indicated by the strong statistical support and the consistent tree topology from the two analyses. Instead of forming a clade, members of Chthamalus intermingle with the other genera in Chthamalinae. M withersi, Tetrachthamalus sinensis, C. malayensis and Chthamalus sp. are consistently clustered together in both the ML and BI trees with strong support, with the other two Chthamalus species and Julius cirratus in basal positions in this clade. J. cirratus and C. challengeri are shown to be sister taxa in both analyses, although this grouping only gains strong support in the ML analysis. Therefore, the genus Chthamalus is polyphyletic. The number of plates of the different species is shown in Fig. 3.2. Most members examined in this study have six shell plates, except Pseudoctomehs and Octomeris, which have eight shell plates, and Tetrachthamalus and Chamaesipho, which have four plates. Barnacles in each of the two major lineages do not have the same number of plates. The first major lineage (mainly comprising the Notochthamalinae) contains individuals with four, six or eight plates. Moreover, the eight-plated Octomeris is relatively derived and does not occupy a basal position as if one would deduce from the plate reduction hypothesis. On the other hand, the second lineage mainly comprises individuals with six shell plates, with the exception of the four-plated Tetrachthamalus. It is clear that members with different numbers of shell plates intermix with each other in the phylogenetic trees. ..

36 Table 3^2 Number of character, number of variable and number of parsimony informative sites of the diree maimers

— Number of character (aligned)Number of variable sites Number of parsimony informative sites

‘� Mitochondrial 12S region m m Nuclear H3 gene 2% 82 59 Nuclear EFla region ^ ^

V

37 • Z Amphibalamis amphitrite Pseudoctomeris sulcata | Notochthamalinae R——— H exechamaesip bo pllsbryi Nesochthamalus intertextus �了 丄丄, , � � Notochthamalinae 51 Octotneris sp. 67 401 Chamaesipho tasntanica F I Caudoeuraphia caudata 画 Euraphiinae Notochthamalus scabrosiis | Notochthamalinae 781 M icroeurap hi a withersi | Euraphiinae 84 78 I Tetrachthamalus sinensis ^ Chthamalus malayensis 仲 Chthamalus s^. �73 1 JehUtis drraUts Chthamalinae 100 I Chthamalus chnlfengeri L—— Chthamalus fragilis i 002 1 Fig 3.2 Maximum likelihood tree from the combined 12S, H3 and EFla analysis under the best-fitting model CTR+I+G. Bootstrap values are denoted on the branches. The subfamily of the species following Foster and Newman (1987) is shown to the right.

38

. • - I————— Amphibalanm amphitrite . Pseudoctomeris sulcata 8 壓 Notochthamalinae Hexechamaesipho pilsbryi 6 og Nesochthamalm intertextm 6 Notochthamalinae " Octomeris sp. 8 y Caudoeuraphia caudata 石讓 Euraphiinae LZZI Notochthamalusscabrosus Notochthamalinae I Chamaesipho tasmamca ^ | Microeuraphia withersi Euraphiinae Teirachthamalus sinensis “ 1 ---- Chthamalus malayensis 6 • "III Chthamalus sp. 6 p ——— Chthamalus montagui ^ Chthamalinae 1 I ••‘ Jehlius cirratus ^ o^ssL——— Chthamalus challengeri 6 Chthamalusfragilis 6

w Fig. 3.3 Bayesian inference tree from combined 12S, H3 and EFla analysis. Posterior probability (PP) of each group is denoted on branches. The number of partite plates of each species is shown after the species name. The subfamily of the species following Foster and Newman (1987) is shown to the right

39 3.4 Discussion Results from the present molecular phylogenetic analysis show a strong disconcordance to the current morphology-based systematics of the Chthamalidae. None of the three subfamilies in the Chthamalidae are shown to be monophyletic in the gene trees.

3,4.1 Subfamily Notochthamalinae The Notochthamalinae is apparently polyphyletic as the basal Pseudoctomeris is not grouped with all the other members of this subfamily examined in this study. On the other hand, Caudoeuraphia caudata, which is classified as an Euraphiinae, is found to be closely related to members in the subfamily Notochthamalinae. It is not surprising that the Notochthamalinae is of a polyphyletic origin, as this subfamily was erected to include barnacles with characters that are intermediate between those in Chthamalinae and Euraphiinae, such that they cannot be grouped into either one of these two subfamilies (Newman and Foster, 1987). Thus the diagnostic characters of the subfamily are mostly likely not synapomorphic (see below). In fact, members in the Notochthamalinae are very diverse in terms of morphology. For instance, species of the Notochthamalinae possess different numbers of shell plates, ranging from four to eight, and the shell plates could either be fused or solitary. The morphological plasticity imposed difficulties in choosing appropriate diagnostic characters that can unify the group. The deep interlocking of the elongated scutum with the narrow tergum was suggested to be a major diagnostic feature of this subfamily (Newman and Foster, 1987). However, the suctum and tergum were found to be fused instead of interlocked in soyne Notochthamalinae such as Pseudoctomeris, and individuals with

40 one pair of scutum and tergum fused while the other pair interlocked were reported occasionally (Newman and Foster, 1987). The presence of multicuspidate setae (card setae) on cirri I-III, which is also an important diagnostic feature of the subfamily, is not confined to the Notochthamalinae but can also be observed in certain members of the Chthamalinae (e.g. C. malayensis; Southward and Newman, 2003). The lack of distinctive diagnostic features makes the Notochthamalinae an assemblage of very diverse species. In this study, Pseudoctomeris is basal to all the other Chathamalidae investigated. Actually, its morphology is rather distinct from other Chthamalidae as its scutum and tergum are fused. It is also noteworthy that it can grow to a much larger size (i.e.,� 3mm0 ) when compared to other Chthamalidae, which are generally small high shore inhabitants (i.e., -10 mm). The clustering of C. caudata with the rest of the Notochthamalinae in the gene trees is collaborated by a similarity shared between C. caudata and Rehederella belyaevi, a species from the Notochthamalinae. Both of them are the only two Chthamalidae species with a caudal appendage. This character, often found in the Lepadomorpha (stalked barnacles), is believed to be an ancestral character in barnacles (Darwin, 1851; Newman and Ross, 1976). Thus, taxonomists consider it as a plesiomorphic feature that is not informative for phylogenetic inference in the Chthamalidae. The close relationship between C. caudata and the Notochthamalinae, as indicated by the present molecular data, suggests the potential importance of the caudal appendage in the systematics of Notochthamalinae. However, since R. belyaevi was not included in the present study, the relationship between C. caudata and R. belyaevi’ together with the origin of the caudal appendage (whether it is asympomorphy o/ synapomorphy) is to be investigated in future studies.

41 3.4,2 Subfamily Chthamalinae When compared to the Notochthamalinae, the clustering of representatives in the Chthamalinae is more compelling. Yet this subfamily is paraphyletic, with Microeuraphia withersi from the Euraphiinae nested within. This finding is consistent with results from previous molecular analyses which suggest that Microeuraphia (as Euraphia) groups with Chthamalus (Fisher et al, 2004; Wares et al., 2009), and provides further solid support to the paraphyly of Chthamalinae. Yet in the present study that has incorporated more genera, we are convinced that, instead of paraphyletic, as shown in previous studies (Fisher et al, 2004; Wares et al., 2009)，the genus Chthamalus is actually a polyphyletic assemblage, since it is grouped with three other genera. The topology also suggests that the presence of six plates may not be an appropriate diagnostic character in classifying Chthamalus, as Tetrachthamalus with four plates clusters with six-plated Chthamalus with robust statistical support. The four-plated configuration of Tetrachthamalus is likely to be a derived character from a six-plated ancestor.

3.4.3 Subfamily Euraphiinae The gene tree shows that the subfamily Euraphiinae is not a natural grouping. The systematics of the subfamily Euraphiinae has remained controversial over the last few decades. Newman and Ross (1976) transferred the genera Euraphia (originally under the Chthamalinae) and Octomeris (now characterized as Notochthamalinae) to a novel subfamily, Euraphiinae, based on the presence of tridentoid mandible and calcareous basis. However, the reliability of these diagnostic features has beep criticized. For instance, Octomeris possesses a membranous basis

42 instead of a calcareous basis. Moreover, Pope (1965) noticed that the mandibular characters often show intraspecific variations; some individuals of the typically four-toothed Chthamalus species have only three teeth, and some members of three-toothed Euraphia were found to have four teeth. These observations certainly have raised questions on the validity of the Euraphiinae. In fact, the distinction of the Euraphiinae from Chthamalinae is unclear. Apart from the aforementioned diagnostic characters, members of the Euraphiinae actually highly resemble Chthamalus. Prior to the establishment of the subfamily Euraphiinae, Pope (1965) even suggested Euraphia as just a form of Chthamalus. Our present molecular study, together with previous observations on Euraphiinae morphology, clearly shows that the subfamily Euraphiinae is an artificial assemblage of distantly related barnacle species. Yet the phylogenetic relationship among other Euraphiinae barnacles, Euraphia and Pseudoeuraphia for instance, remains unsolved owing to the limited taxon coverage of the present study. We are also convinced that the morphology of mandible in this polyphyletic group is probably a homoplasy, or represents convergent evolution in other words, which is therefore, not an appropriate diagnostic character in reconstructing Chthamalidae phylogeny. This also brings out the significance of the molecular approach in inferring barnacle phylogeny, particularly in those groups with high morphological plasticity, such as the Euraphiinae and Chthamalidae.

3.4.4 Phylogenetic relationship in the family Chthamalidae The separation of the subfamily Notochthamalinae and Chthamalinae generally matches the current molecular phylogeny, if the placement of Pseudoctomeris, Jehlius and Tetrqchthamalus is reconsidered. On the contrary, the Euraphiinae is a

43 polyphyletic assemblage and the two exemplars, Caudoeuraphia cmdata and Microeuraphia withersi included in the present study are found to nest deep inside the Notochthamalinae and Chthamalinae respectively. Hence, a taxonomic revision is prompted. However, only two of the three genera of Euraphiinae were included in this study, and most importantly, the type genus Euraphia was not examined owning to the lack of samples. The validity of the Euraphiinae can only be confirmed when this genus is included in further studies. The family Chthamalidae is believed to be an ancient group of acom barnacles. Darwin (1854) was the first to suggest that Catomerus polymerus, a primitive acom barnacle closely related to the Chthamalidae, was evolved from multi-plated stalked barnacles. Newman (1987) further expanded this idea and proposed that the family Chthamalidae experienced a trend of reduction in shell plate number during its evolutionary history, from the basal eight-plated Octomeris to the six-plated Chthamalus, and finally to the most advanced four-plated Tetrachthamalus. Although our molecular study cannot completely reject this hypothesis owning to the moderate statistical support, the topology of the phylogenetic tree does not support a trend of reduction in plate number with time. Firstly, one of the eight-plated species, Octomeris does not occupy the basal position in the phylogenetic tree. And the four-plated Chamaesipho and Tetrachthamalus, which should be the most advanced Chthamalidae according to the plate reduction hypothesis, are not the most derived as shown in the tree. Secondly, the clustering of barnacles in the trees did not show obvious correlation to the three parietal plate configurations. The phylogenetic analysis shows that even if there is a trend of shell plate number reduction, this event at least happened twice in the family independently. This is because four and six-plated men;bers were found in both subfamilies Notochthamalinae and

44 Chthamalinae. This result is similar to the finding of Fisher et al. (2004). This may also support the hypothesis of Ross (1971) that the four-plated Chthamalidae were attained through different modes of parietal plate reduction, i.e. Tetrachthamalus attained its four plated stage by the fusion of the rostrum with the rostrolateral plate, while the four plates present in Chamaesipho is a result of the merging of rostrolateral plate and lateral plate.

3.4.5 Suggestions on taxonomy of Chthamalidae Based on the findings from the present molecular study, the following suggestions are made on the taxonomy of the family Chthamalidae.

I. Subfamily Notochthamalinae and subfamily Chthamalinae should be retained in the family Chthamalidae. However, in order to confirm the validity of the subfamily Euraphiinae and its inclusion in this family, further investigations are needed probably by including all the species, especially the type genus Euraphia in this subfamily. II. Pseudoctomeris (subfamily Notochthamalinae) is obviously distinctly related to the rest of the Chthamalidae barnacles, in terms of both genetic data and morphological characters. A new subfamily may be erected to hold this genus. III. Caudoeuraphia caudata is nested within the subfamily Notochthamalinae in the present study with moderate bootstrap support. Transferring this genus from subfamily Euraphiinae to subfamily Notochthamalinae should be considered.

45 IV. Microeuraphia withersi groups with Chthamalinae barnacles with strong support. Transferring this taxon from subfamily Euraphiinae to subfamily Notochthamalinae is prompted.

Although these suggestions are largely based on molecular data, with little incorporation of morphological characters, it can serve as a valuable reference for future taxonomic revision in this group of intertidal barnacles.

46 Chapter 4

Cryptic Diversity and Genetic Structure of the Acorn Barnacle Chthamalus mow in the Northwest Pacific

4.1 Introduction Recently, a growing number of genetic studies have shown that population structuring is not rare in various marine taxa, and the structuring in many of them is believed to have resulted from Pleistocene glaciations (Benzie et al., 2002; Rohfritsch and Borsa, 2005), or from ocean circulation patterns (Knutsen et al., 2003; York et al., 2008). However, our knowledge on the population structure of many species inhabiting the Northwest Pacific is lacking and the interactions between regional geography, ocean currents and the evolutionary history on the population structuring of these marine species remain unclear. The NW Pacific region was greatly affected by glaciations during the Pleistocene (Voris, 2000). A few hypotheses have been proposed on the geography of the region during the Pleistocene, especially those related to the Ryukyu Island Chain (Kizaki and Oshiro, 1977; Ujii^, 1990). It has been suggested the Ryukyu Islands and Taiwan were connected by a land bridge, making the East China Sea an inland sea almost isolated from the nearby seas during the Pleistocene glaciations (Fig. 4.1; Ujiie et al., 1991). The exposure of the Sunda Shelf and the linkages between different islands in the Philippines caused the South China Sea to become another semi-enclosed gulf in the region. These enclosed seas may serve as potential refugia . for various marine taxa in facing habitat shortage and adverse climate during the Pleistocene. ‘‘

47 , ‘ ： :: .J董n

Ryukyu / E^^P^ \ Islands

South Jjt； Ml ？ina t Philippines _ Sea

Fig. 4.1 Map showing geographical differences in the Northwest (NW) Pacific region between the present day and the Last Glacial Maximum during the Pleistocene. Redrawn based on Ujii6 et al. (1991) and Ahagon et al. (1993). Regions shown in J green represents the present day geography of NW Pacific, while regions colored in deep blue indicates the land exposed during LGM.

‘ ..

48 Although only very few comparative phylogeographic studies have been performed to identify the roles of the marginal seas in shaping the genetic structure of the marine species in the NW Pacific (Liu et al., 2007). In addition, Pleistocene glaciations also play a role in leading to the generation of genetic structuring and diversification in marine populations, by the exposure of landmasses and land bridges between refugia. One well-known example is the exposure of the Sunda shelf in Southeast Asia, which has prohibited gene flow between populations in different refugia in some marine taxa, such as the barnacle Chthamalus malayensis (Tsang et al, 2008) and sea horse Hippocampus spp. (Lourie et al., 2005). The land bridges, such as the Ryukyu island chain, and the narrowed straits between the marginal seas were likely barriers, which prohibited gene flow between populations. Apart from Pleistocene glaciations, the ocean current system in the NW Pacific may also be responsible in shaping the population structure of marine organisms. The Kuroshio Current (KC) is the most prevalent current affecting this region. The main path of KC starts near northern Philippines, flows along the eastern side of Taiwan and enters the East China Sea through the Ilan ridge. The KC then splits into two main branches near northern Ryukyu Islands. While one of the branches flows into the Sea of Japan, the other one passes through the Tokara strait in northern Ryukyu Islands and flows along the eastern side of mainland Japan (Figs. 4.2 & 4.3; Ujiie and Ujiie, 1993). It is reported that a countercurrent of KC is generated in

49 ：：：」

瞧！麵 一，

^^^^ Xjy Kuroshio .Current (KG)

^ South m 囊？： . ^^ * North Equatorial Current- 紀一

Fig. 4.2 Map of Northwest (NW) Pacific region. The main path of the North Equatorial Current (NEC) and Kuroshio Current (KC) are shown. The two dash arrows represent side branch of KC while dotted arrow refers to the countercurrent of KC.

50 .：LuZQTll 攀 魁八 Philippine m_ South Chin a > i j< 圖 Sea P^es "loo kT ^ : Fig. 4.3 Map showing the main straits in Northwest Pacific region. The main path of Kuroshio Current (KC) is shown. The asterisks (*) indicate the positions of the three straits, the Luzon Strait, Ilan Strait and Tokara Strait respectively and concentric circle shows the location of Kerama Gap. The two dash arrows represent side branches of KC while dotted double lined arrow refers to the countercurrent of KC.

51 the East China Sea by the northward flowing movement of the main branch of the KC (Qiu and Imasato，1990). This KC countercurrent flows in a direction opposite to the main branch and its path is close to the western coastal region of the Ryukyu Islands. Two KC side branches are also observed in the region (Ujiie and Ujiie, 1993). One of them flows into the South China Sea (SCS) through the Luzon Strait between Taiwan and the Philippines, while the other diverges from the main branch of the KC before entering the Ilan ridge, and flows along the eastern side of Ryukyu Island Chain (Figs. 4.2 & 4.3). However, the role of the Kuroshio Current in shaping population structure of marine organisms remains largely unstudied. The Kerama gap located between Taiwan and Ryukyu, which is a deep strait (i.e., ~600 m) with complicated hydrobiology, is also believed to be a barrier that affects the population structure of some marine taxa, such as snails and stony corals (Fig. 4.3; Kojima et al., 2006; Nishikawa, 2008). Intertidal barnacles in the genus Chthamalus have recently become models in population genetics studies of the Western Pacific. Previous population studies on the tropical C. malayensis and temperate C. challengeri have covered most of the continental intertidal region in the western Pacific, and revealed very different population structure patterns. While three distinct lineages are recovered in C. malayensis in the Indo-West Pacific (Tsang et al, 2008), no strong population structuring is observed in C. challengeri (Cheang et al., submitted). These two Chthamalus species, however, are continental species and their distribution ranges seldom cover the intertidal habitats on islands. Given that the Western Pacific region comprises many islands, and the interaction between oceanic species with the . different factors in this region is largely unstudied, we investigated the population structure of the ‘barnacle Chthamalus mow in the NW Pacific region in this study.

52 C. mow is a common intertidal space occupier of the mid tide zone on rocky shores of the NW Pacific and South Pacific region. Its distribution covers many islands, including the Ryukyu Islands, Ogasawara Islands, Pacific coast of Taiwan, Paracel Islands, Philippines, Indonesia, Mariana Islands, Caroline Islands, Fiji and Samoa (Southward and Newman, 2003). With its wide distribution range, high ecological value in the intertidal community and typical life cycle of many intertidal organisms, C. mow is an ideal representative of intertidal ecosystem in NW Pacific, and the population genetics study on C. mow can serve as a model in demonstrating how is the population structure of island fauna and flora attributed to the different environmental forces in the region.

4.2 Materials and Methods 4.2.1 Sample collection, DNA extraction and amplification Samples of C. mow were collected from 13 sites in Northwest Pacific by Dr. Benny KK Chan (Fig. 4.4). Two museum samples from Fiji were also included in this study for comparison. Preliminary study of this project indicated that there are more than one genetically distinct lineage of C. mow in Okinawa, and thus sampling were carried out in five sampling sites on the island to determine if there were variations in the relative abundance of the different lineages. The samples were preserved in ethanol (70% or higher), and the adductor muscle or soft tissue from the abdomen was dissected and subjected to DNA extraction using QIAamp Tissue Kit (QIAGEN). ‘ .

53 c- \ , Okinawa ri . . South . . , :: Island .China • ...: :一.'： .…...... —一 fc/ Se�..Phme s Fj

‘:。:'：.麵:,�t�^^^^^^^^^

Fig. 4.4 Chthamalus mow. Sampling localities are shown in capital letters. See Table 4.1 for the abbreviations of localities.

I 奄，

54 Partial sequences of mitochondrial cytochrome c oxidase subunit I (COI) gene were amplified using the universal primer set LC01490 and HC02198 (Folmer et al., 1994). The PGR amplifications of COI were conducted using IX PGR reaction buffer, 2 mM MgCh, 200 nM of each primer, 200 ^iM dNTPs, 1.5 units of Taq polymerase (Takara), 1 \iL of template DNA and ddH20 in 50 |il total reaction volume, and the PGR profile was as follows: initial denaturation at 94°C for 3 min., followed by 35 cycles of denaturation at 94°C for 30 s, annealing at 49°C for 30 s, elongation at 72°C for 30 s and a final extension at 72°C for 3 min. The PGR products were then purified using the QIAquick gel purification kit (QIAGEN). Sequences were generated using the same set of primers and Applied Biosystems (ABI) 371 Ox/ automated sequencer using the ABI Big-dye Ready-Reaction mix kit, following the standard cycle sequencing protocol.

4.2.2 Phylogenetic analysis Sequences were aligned using CLUSTAL W (Thompson et al., 1994). The alignments were translated from DNA sequences into amino acid sequences, to ensure no stop codons were present. The mtCOI sequences were subjected to phylogenetic analysis using neighbor-joining method (NJ). Kimura 2-parameter (K2P) distance analysis on MEGA 4.0 (Tamura et al., 2007) was used. Sequences of two congeneric barnacle species, C. malayensis, C. challengeri and a closely related species, Hexechamaesipho pilsbryi were also determined and included as outgroups in the analysis. The bootstrap support for each internal node was estimated by using 1000 pseudoreplicates. ‘ •

4.2.3 Population genetic analysis

55 Basic population genetic indices such as haplotype diversity (h) and nucleotide diversity (兀）and their standard deviations were analyzed at both population and lineage (as defined by the above phylogenetic analysis) level using ARLEQUIN 3.0 (Excoffier et al., 2005). Average nucleotide divergences between the lineages were calculated based on K2P distance using MEGA 4.0. To detect the signal of population structuring and estimate the degree of gene flow among populations within each lineage, pairwise OST was also calculated. Tajima's D test (Tajima, 1989) was applied to test for the departure from equilibrium between genetic drift and mutation for a selectively neutral marker. Negative D value implies an excess of low frequency or rare haplotypes, which in turn, could be the signal of sudden demographic expansion. Fu's Fs test (Fu, 1997) also serves a similar purpose as Tajima's D test, for which negative Fs value suggests the populations might have experienced rapid demographic expansion. Mismatch distribution analysis (Rogers and Harpending, 1992) was used to reveal demographic history exhibited by different lineages. Unimodal distribution model observed in lineages represents rapid population expansion. All the pairwise Ost, Tajima's D test, Fu's Fs test and mismatch distribution analysis were computed using ARLEQUIN 3.0 package (Excoffier et al, 2005). Bayesian Skyline plot were produced for the three lineages respectively using a strict molecular clock. The program was run using default priors for Bayesian skyline analysis for 50 million generations, with one sampling for every 1000 generations. Mutational units from the raw result were converted to years for the skyline plot curves, under the assumption that the COI divergence rate of C. mow is 1.5 % per million years, which is employed from the estimate for the . congeneric trans-isthmian Chthamalus species (Wares 2001). Analysis of molecular variance (AMOVA) and corrected pairwise genetic differences between

56 geographical regions were also conducted by using ARLEQUIN 3.0. Three groups were assigned in accordance to the geographical barriers in NW Pacific, excluding Fiji in the South Pacific where only two individuals were available. The groups are (1) Ogasawara Islands, which are isolated from the rest of the localities by the northern Philippine Sea, (2) Central Ryukyu Islands, which are separated from the areas to the south by the Kerama Gap, and (3) West Pacific, including Taiwan, Paracel Islands and the Philippines. In order to investigate if the co-distributed lineages (as defined by phylogenetic analysis) exhibit stratification in shore level, C. mow collected from three sampling sites in Okinawa Island (Seragaki, Cape Tenyi and Cape Hedo) were chosen for analysis. The C. mow in the three sites were differentiated according to the shore level they inhabit (i.e. high, mid and low shore). The data at the same shore level from the three sites were combined, and the relative distribution of respective lineages on the three shore levels was determined.

4.3 Result 4.3.1 Phylogenetic analyses Partial sequences of mtCOI were obtained from 248 individuals from 14 localities, yielding a total of 225 unique haplotypes. Out of 531 bp, there are 87 polymorphic sites. No indels were observed in any of the partial COI sequences. Fig. 4.5 shows the NJ tree reconstructed from these unique haplotypes and the outgroups. The topology reveals a prominent genetic structure in C. mow, exhibiting three distinct lineages each with very strong (>90%) bootstrap support. . The first lineage, referred to as Ryukyu lineage, is mainly found in Ryukyu Islands, though some of its members can also be found in Ogasawara Island and Taiwan.

57 I F

厂 邑 Ryukyu ^^ I Lineage

&• •

C" f- 85 |[

力I

91 象

！^ Ogasawara •§kc= ‘ Lineage r-^ t

I . i y V Fig. 4.5 Chthamalus moro. 99 Taiwan Inferred phylogenetic tree I ^ Lineage of mitochondrial COI ^^ haplotypes using F Neighbor-joining (NJ) ^ method. Percentages of bootstrap replicates are [ shown on the . 三 corresponding branches “1 I.,I.r •"_�•”�,“7;;»^.»»*iioiS! S for values greater than 75. , ~"..一一 Chthamalus challengeri, z C. malayensis and Hexechamaesipho pilsbry are used as outgroups.

58 A majority of the second lineage, called Ogasawara lineage, is distributed on Ogasawara Island and Ryukyu Islands, and one individual from this lineage is observed in Turtle Island, Taiwan. The last lineage, referred to as Taiwan lineage, constitutes individuals from Taiwan, Paracel Islands, the Philippines and Fiji. Few individuals in this lineage are found in western coast of Okinawa Island. The relative abundance of the lineages in each locality is shown in Fig. 4.6. Compared to the distribution of Taiwan lineage, the Ryukyu lineage and Ogasawara lineage share a relatively northern distribution, and are highly sympatric in their range. They co-occur in all localities north of Taiwan but in Hualien and Kenting of Taiwan, only members from the Ryukyu lineage are found. The Taiwan lineage has the widest distribution range among the three lineages; it can be found in the Philippines Sea, South China Sea and the South Pacific. This lineage is allopatrically distributed with Ryukyu lineage and Ogasawara lineage, with overlapping range with the two lineages at different magnitudes. The overlapping zone of the Ryukyu and Taiwan lineages is larger than that between Ogasawara and Taiwan lineage, as the former two lineages are found to be co-distributed in the whole Taiwan and the southwestern side of Okinawa Island (Manza and Seragaki). The number of individuals involved in mixing is also greater compared to the case in Taiwan and Ogasawara lineage. In contrast, the Ogasawara and Taiwan lineages only overlapped in Turtle Island of Taiwan, and the southwestern part of Okinawa Island (Manza and Seragaki) at a very small scale, with only one Ogasawara lineage individual recorded in Turtle Island (Fig. 4.6). ‘ .

59 • Pacific I

A, 八1多• •

Nsft^^—f 少.,., FJ

• Taiwan clade ^ 卿叫 fji ^： 〇 Ryukyu clade & ^oookm \ 一 ® Ogasawara clade 、、 、： Fig. 4.6 Chthamalus mow.. Relative abundance of the three lineages in each population is represented as pie charts. See Table 4.1 for the abbreviations of localities.

/

60 The COI sequence divergences between the three lineages are high. The Ryukyu and Ogasawara lineages are sister lineages in the inferred tree and differentiated by -4%, The Taiwan lineage is more distantly related to the other two lineages. The greatest sequence divergence (-9%) is observed between Taiwan and Ogasawara lineage, and the sequence divergence between Taiwan and Ryukyu lineage is �7% Th.e inter-lineage sequence divergences are much higher than the divergence within each lineage (1-2%). Two sublineages are observed in the Taiwan lineage in the inferred tree, with less than 0.8% genetic divergence. No geographical structuring is observed for the two sublineages throughout the range of this lineage (Fig. 4.5).

4.3.2 Population genetics analyses, demographic history and neutrality Since C. mow investigated in this study was found to be a complex consisting of three genetically very distinct lineages with sympatric distribution, the population genetic analyses were carried out at three levels: (1) among lineages, (2) among populations, and (3) among populations in each lineage. The overall genetic diversity of C mow is high (0.993-0.998) and most of the haplotypes are sampled from a single individual only. The Ogasawara lineage is found to harbor the highest haplotype (0.9984) and nucleotide diversity (0.01616) (Table 4.1). The lowest haplotype diversity (0.993) is found in the Taiwan lineage, while the lowest nucleotide diversity (0.008934) is observed in the Ryukyu lineage.

t

if

61 Table 4.1. Chthamalus moro. Sampling localities, co-ordinates (exact co-ordinate: TY, SE, MY，CH), abbreviation and genetic diversity indices of each lineages, localities and subpopulations respectively. Sample size (n), number of haplotypes, haplotype diversity (h 土SD)， nucleotide diversity (7c±SD)，Tajima's D and Fu's Fs values are shown. * = p< 0.05，**= p < 0.01 and ***= p < 0.005.

•� Coordinates No. of Lineage Abbreviation n haplotype h±SD JC±SD Tajima's D Fu's Fs Ogasawara lineage Og 72 69 0.9984 ± 0.0028 0.016160 ± 0.008374 -2.03250*** -24.74487*** Ryukyu lineage Ru 89 82 0.9974 ± 0.0024 0.008934 ± 0.004919 -2.44688*** -25.69471*** Taiwan lineage Tw 87 74 0.9930 士 0.0040 0.015211 士 0.007927 -1.74005* -24.83782*** Population Hahaiima Island, Ogasawara 26''37'N/142°9'E I I ^ OG 35 34 0.9983 ± 0.0074 0.019039 ± 0.009907 -1.72353* -24.65269*** 0„asawara lineage OG O 31 30 0.9978 士 0.0089 0.013592 士 0.007282 -2.13843** -24.98849*** ^ 一 0.007533 士 Ryukyu lineage OG R 4 4 1.0000 ± 0.1768 0.005661 -0.82407 -0.82472 Amami Island, Ryukyu 28°23'N/129°27'E Islands AM 6 6 1.0000 ±0.0962 0.029506 ± 0.017780 0.10365 -0.35952 Ogasawara lineage AM O 2 2 1.0000 ± 0.5000 0.013183 士 0.014093 0 1.94591 Ryukyu lineage AM—R 4 4 1.0000 士 0.1768 0.013183 士 0.009371 -0.84532 -0.09451 Kume Island Rvukvu Islands 26°22'N/126°47'E kU 28 28 1.0000 ± 0.0095 0.027404 ± 0.014085 -0.70047 -17.53653*** Kumelsland==l= 二 13 1.0000 ± 0.0302 0.016756 士 0.009288 -1.78562* -6.2632P- g二二 KU~R 15 15 1.0000 ±0.0243 0.008607 ± 0.005049 -1.97596* -13.06175*** 26。30’25 7"N/ Seragaki, Okinawa Island 127°52’44.0，，E SE 25 24 0.9967 土 0.0125 0.025123 ± 0.013028 -1.39241 -1^0065*- Taiwan lineaee SE T 1 1 1.0000 士 0.0000 0.000000 士 0.000000 0 NA Ogasawara lineaee SE"0 5 5 1.0000 士 0.1265 0.021092 ± 0.013527 -1.01279 -0.18978 RX^Sc SE"R 19 18 0.9942 土 0.0193 0.008045 土 0.004715 -2.01720** -16.78107***

62 Coordinates No. of Lineage Abbreviation n haplotype h±SD Jt±SJ) Tajima's D Fu's Fs 26。32’45.8"N/ Cape Tenyia, Okinawa Island 128°06’10.0，，E TY 24 23 0.9964 士 0.0133 0.023221 士 0.012114 -1.14885 -11-88160*- Oaa<;awaralineaee TY O 6 6 1.0000 ± 0.0962 0.020716 ± 0.012698 -1.20333 -0.81298 "Ryukyu lineage TYR 18 17 0.9935 士 0.0210 0.008739 士 0.005027 -1.95243* -13.55109*** 26°47'16.8"N/ Cape Hedo, Okinawa Island 128。13’14.0，’E CH 15 15 1.0000 士 0.0243 0.029147 士 0.015471 -0.44008 -5-46581* igasawara lineage CH O 6 6 1.0000 ± 0.0962 0.020716 士 0.012698 -0.65102 -0.81298 R^u lineage CH~R 9 9 1.0000 士 0.0524 0.010149 士 0.006126 -1.65253* -4.62107** Manza Okinawa Island 26°12’N/127°40’E MZ 7 7 1.0000 ± 0.0764 0.044628 ± 0.025673 -0.19658 -0.30169 ’ Taiwan lineaee MZ T 1 1 1.0000 ± 0.0000 0.000000 ± 0.000000 0 NA Ogasawaralineaie MZ"o 3 3 1.0000 ± 0.2722 0.017577 ± 0.013918 0 1.06599 rXu lineage MZ"r 3 3 1.0000 士 0.2722 0.007533:^0.006402 0 0.13353_ Mijmke Beach, Okinawa If^l^^：}^^^ MY 14 14 1.0000 士 0.0270 0.026895 土 0.014568 0.30913 -6.05642- Oaasawara lineage ‘ MY O 5 5 1.0000 ± 0.1265 0.007707 ± 0.005404 0.29358 -1.71605 R^^lineS MY"R 9 9 1.0000 士 0.0524 0.010360 ± 0.006393 -1.18793 -5.31090*** Tci.nri 24°50'N/121°56'E TI 29 29 1.0000 ± 0.0091 0.040295 士 0.020580 -0.4822 -17.52826*** Turtle Island, Taiwan 丁I 丁 23 1.0000 士 0.0128 0.017542 土 0.009529 -1.40282 -19.76436-* OaasLaralineaL Tfo 1 1 1.0000 ± 0.0000 0.000000 ± 0.000000 0 NA RX^Se TfR 5 5 1.0000 ± 0.1265 0.007345 ± 0.005167 -0.6909 „ „ 23。29，N/121。30’E ^ 12 11 0.9848 ± 0.04030.025171 ±0.013752 -1.12599 -2.26355 Hualien，laiwa^ HL T 11 10 0.9818 士 0.0463 0.014388 ± 0.008233 -0.31633 -3.16243 R^k^linLge HL'R 1 1 1.0000 ± 0.0000 0.000000 士 0.000000 0 NA V T。二 21°58’N/120M2’E ^ 24 ^ 1.0000 ± 0.0120 0.025643 ± 0.013347 -0.80652 -14.72001*** Ken Ding, Taiwan 22 22 1.0000 ± 0.0137 0.014686 士 0.007968 -0.61746 -17.90165*** R^^ lineage KT"r 2 2 1.0000 士 0.5000 0.007968 ± 0.008909 0 LM^_ Paracellslands/China 16°21’N/111°57，E PS 3 3 l._0土 0.2722 0.008878 ± 0.007412 _0 。麗3 Puerto Galera Philippines 13°29’N/120°57，E PG 24 22 09891 ±0.0173 0.015859 士 0.008478 -0.87927——-12.93623… Fiji Fiji Island17°42’S/178。3’E ^ 2 2 1.0000 ± 0.5000 0003781 ± 0.004628 0 0.69315

63 Table 4.2. Chthamalus mow. Pairwise Ost values for the three lineages based on COI region. * = p< 0.05，**= p < 0.01 and ***= p < 0.005.

Taiwan lineage Ryukyu lineage Ryukyu lineage 0.83538*** Ogasawara lineage 0.80928*** 0.69467***

64 Table 4.3. Chthamalus mow. Pairwise Ost values for populations of North West Pacific region based on COI region. * = p < 0.05，**= p < 0.01 and ***= p < 0.005. The abbreviations of localities are shown in Table 4.1.

— OG AM KU SE TY CH MZ MY TI HL KT PS PG AM 0-261** / KU 0.190*** -0.052 / SE 0.381*** -0.033 0.041 / TY 0.367*** -0.029 0.028 -0.011 / CH 0.255*** -0.066 -0.026 0.001 -0.010 / MZ 0.176** -0.087 -0.021 0.040 0.053 -0.032 / MY 0.255*** -0.063 -0.017 0.021 -0.002 -0.034 -0.016 / TI 0.625*** 0.477*** 0.530*** 0.508*** 0.543*** 0.502*** 0.393*** 0.520*** / HL 0.743*** 0.647*** 0.657*** 0.649*** 0.685*** 0.643*** 0.538*** 0.669*** 0.003 / KT 0.734*** 0.652*** 0.660*** 0.650*** 0.682*** 0.651*** 0.566*** 0.670*** 0.004 -0.021 / DS 0.800*** 0.730* 0.714*** 0.715*** 0.750*** 0.702*** 0.582* 0.738*** 0.073 0.219 0.109 / PG 0.797*** 0.770*** 0.739*** 0.738*** 0.765*** 0.745*** 0.690*** 0.765*** 0.078** 0.017 -0.004 0.225 / FJ 0.787*** 0.699* 0.695** 0.698** 0.735** 0.677*** 0.526* 0.715* 0.085 0.176 0.140 0.703 0.318**

�

65 The pairwise OST values among the three lineages are large (0.69467-0.83538), with robust statistical support (p < 0.005) (Table 4.2). At the population level, the Ogasawara population shows high pairwise OST values when compared with all other populations (Table 4.3). The remaining populations can be roughly divided into two big groups, one consisting of populations in the Ryukyu Islands (AM, KU, CH, TY, SE，MZ and MY), while the other group comprising populations from Taiwan (TI, I HL，KT), Paracel Islands, the Philippines and Fiji. Almost no population structuring is observed within the two groups respectively, as indicated by low, negative pairwise ST values with no significant statistical support. The only significant and positive values within groups are observed in TI-PG pair and PG-FJ pairs respectively. However, high pairwise OST values are recorded for all the cross-group comparisons (Table 4.3)，suggesting significant population structuring among lineages. The result of pairwise O ST analysis carried out among populations within the same lineages is similar to that of population level analysis (Table 4.4a，b and c). All pairwise Ost values are either negative or insignificant among populations of the Ogasawara lineage (Table 4.4a). For the Ryukyu and Taiwan lineages (Table 4.4b and c), most of the pairwise O ST values are not significantly deviated from zero, and the only significant positive values are found in six population pairs in the Ryukyu lineage that includes either MY or KT individuals (MY_R-KU_R, MY_R-SE_R, MY_R-TY_R; KT_R-KU_R, KT_R-TY_R and KT_R-CH_R) (see Table 4.1 for the abbreviations), and four population pairs in the Taiwan lineage (TI—T-DS一T， TI_T-FJ_T, HL-T-DS_T and HL_T-FJ_T). .

66 Table 4.4a. Chthamalus mom. Pairwise �S valueT s for populations of Ogasawara lineage based on COI region. * = p < 0.05, **= p < 0.01 and ***= p < 0.005. The abbreviations of localities are shown in Table 4.1.

OG Q AM O KU O S^ TY O CH Q ML 0 MY O ILQ 0G_0 / AM_0 -0.04497 / KU_0 -0.01785 -0.05174 / SEJO -0.00525 -0.12589 -0.05382 / TY_0 0.0293 -0.10103 0.00999 -0.01145 / CHIO 0.03585 -0.06706 -0.02365 -0.09151 -0.00775 / MZ:0 -0.00408 -0.1413 -0.00884 -0.06118 -0.02498 -0.01087 / MYIO 0.03354 0.07563 0.0422 0.02062 -0.03448 0.07109 0.12075 / Tl O 0.60894 0.66667 0.50893 0.30864 OJ 0.30526 0.55932 0-78495 /_

67 Table 4.4b. Chthamalus mow. Pairwise

� OG R AM R KU R SE R TY R CH R MZ R MY R TI R HLR KT R OG_R / AM_R -0.07317 / KU_R -0.05999 0.00624 / SEJl -0.03428 0.01845 -0.00516 / TY:R -0.00929 0.0358 -0.00763 0.00951 / CH:R -0.02727 -0.01258 -0.00676 0.01502 -0.01481 / MZIR -0.04348 -0.05775 0.01015 0.00109 0.01035 0.03271 / MYIR 0.0059 0.02255 0.0415* 0.05532** 0.04474* 0.03144 0.00169 / TI R .0 03877 -0.06334 -0.03578 -0.01477 -0.01434 -0.03881 -0.00227 0.03568 / HL"R 0 -0 27273 0.07445 0.07241 0.08394 0.03 -0.2 -0.07075 0.15217 / KT—R 0.2381 0.00498 0.18997* 0.21622 0.21359* 0.16854* 0.22581 0.13009 0.19804—— !__

68

• - Table 4.4c. Chthamalus moro. Pairwise �S valueT s for populations of Taiwan lineage based on COI region. * = p < 0.05，**= p < 0.01 and ***= p < 0.005. The abbreviations of localities are shown in Table 4.1.

� SE T MZ T TI T HL T KT T DS T PG T FJ T SE_T / MZ_T 1 / TI_T 0.19251 0.20687 / HL_T 0.04286 0.08636 0.00573 / KT:T 0.18497 0.19708 -0.02802 0.00179 / DSIT 0.72093 0.72093 0.18051* 0.35374** 0.22965 / PGT 0.18598 0.14055 -0.02263 -0.00538 -0.02634 0.22351 / FJ'T 0.81818 0.83333 0.32913* 0.35188** 0.32008*** 0.7 0-31918** L

69 • z Table 4.5. Chthamalus mow. Analysis of Molecular Variance (AMOVA) for genetic structure between different localities (based on COI). p = probability of having a more extreme variance component and O statistics than values observed by random event.

Source of variation Variance components ^^arilrior^ O statistics P-value Among groups 10.60291 60.95 0.60954 0.00098

Among populations 0.13059 0.75 0.01923 0.14761 within groups .

Within populations 6.66141 38.3 0.61705 ^

70 All the positive pairwise O ST values observed in population pairs within the Taiwan lineage involve comparisons with the Paracel Island or Fiji populations. The sample size of these two populations are too low (two individuals each) and likely lead to 'false positive' values when compared with populations with large sample size. This explanation can also be applied on population pairs that involve KT population (two individuals) (KT_R-KU_R, KT_R-TY—R and KT_R-CH_R) in the Ryukyu lineage. In AMOVA, a majority of molecular variations is observed at two levels, between groups (Ogasawara, Ryukyu and West Pacific) and within populations (Table 4.5). Whereas around 60% of the observed molecular variations are confined to between group level, around 38% of variation goes to intra-population level, and the level 'among populations in a lineage' only accounts for less than 1% of total variation. The Tajima's D and Fu's Fs values of the three lineages are negative and statistically significant (Table 4.1). The highest D and the lowest Fs values are observed in the Ryukyu lineage. The haplotype mismatch distribution of the Ryukyu and Ogasawara lineages (Fig. 4.7a，b) are both unimodal and generally matches the expected distribution under rapid demographic expansion (the small peak appeared in the mismatch distribution curve of the Ogasawara lineage is to be explained in the next section). The haplotype mismatch distribution of the Taiwan lineage appears to be bimodal and is greatly deviated from the expected curve (Fig. 4.7c). The results from the Bayesian skyline plot also indicate that the estimated population size of the three lineages were much smaller in the Middle Pleistocene

71 (200,000 years before present) than those in present day (Fig. 4.8). This analysis also shows that all the three lineages have experienced a rapid demographic expansion in the past, yet at different intervals. The Ogasawara lineage expands the earliest, as it has experienced a demographic expansion at around 450,000 years before present, and the demographic expansion of Taiwan and Ryukyu lineages took place at 300.000 and 250,000 years ago. All the demographic expansions took place before the Last Glacial Maximum (-20,000 years ago). For the analysis on shore level stratification of Ryukyu and Ogasawara lineages on Okinawa Island, the great difference between datasets from the three sites led to the high standard deviations observed in the combined dataset, and thus, this analysis fails to reveal any difference in shore level distribution between the two lineages in Okinawa Island (Fig. 4.9).

4.4 Discussion 4.4.1 Origin and the systematic status of the three lineages Significant population structuring is observed in C. mow in the Northwest Pacific Ocean. Three genetically distinct lineages are identified from this study, with mean COI divergence ranging from 4% to 9% among the three lineages. Large and highly significant Ost values observed between the three lineages further confirm the strong genetic structuring in C. moro. The lower inter-lineage sequence divergence between the Ryukyu and Ogasawara lineages indicates that the two lineages diverged

72 1200 -

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Pairwise difference Fig. 4.7a Chthamalus moro. Observed and expected distribution of pairwise sequence divergences under the sudden expansion model (Rogers 1995) of COI region of Ogasawara lineage. Observed (Bars) and expected (curve) distributions. Harpending's raggedness index: r = 0.00977693 (p = 0.55).

73 1000 - 900 - 800 -

丨;IJIiil^” 1 2 3 4 5 6 7 8 9 10 11 12 13 .14 15 Pairwise difference Fig. 4.7b Chthamalus mow. Observed and expected distribution of pairwise sequence divergences under the sudden expansion model (Rogers, 1995) of COI region of Ryukyu lineage. (Bars) Observed and (curve) expected distributions. Harpending's raggedness index: r = 0.02543284 (p = 0.05).

74 1200 -

1000 - 八 I . \ I • I \ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Pairwise difference Fig. 4.7c Chthamalus mow. Observed and expected distribution of pairwise sequence divergences under the sudden expansion model (Rogers, 1995) of COI region of Taiwan lineage. (Bars) Observed and (curve) expected distributions. Harpending's raggedness index: r = 0.00654859 (p = 1).

75 ~~^ 1.E1

Ogasawara clade —

Ryukyu clade “―• m

Taiwan clade — (y^. . a r^i

— • ., 1.E.3 800,000 700,000 600,000 500,000 400.000 300,000 200,000 100,000 0 Years before present

Fig. 4.8 Chthamalus mow. Bayesian skyline plots of Ogasawara lineage (black),

Ryukyu lineage (green) and Taiwan lineage (red) based on COI dataset inferred with

BEAST.

76 100 1

80

§ H« «Ryukyu lineage 8 40 n T « ‘ "'|B "Ogasawara lineage y aiil ^ High shore Mid shore Low shore

Fig. 4.9 Chthamalus mow. Relative distribution (in %) of Ryukyu and Ogasawara lineages in different shore level respectively in Okinawa Island. Dataset from three sites (SE: Seragaki, TY: Cape Tenyi and CH: Cape Hedo) in Okinawa Island were selected for combined analysis.

. /

if

I

77

•:/ - ••‘..

-. .V relatively recently, while the Taiwan lineage diverged earlier from the remaining two lineages. The strong population divergence observed is unexpected, since no morphological or ecological differentiations have ever been reported in this species. The observed phylogenetic breaks between the Ryukyu, Ogasawara and Taiwan lineages may be brought about by Pleistocene glaciations. Using the molecular clock of closely related barnacles in the genus Euraphia, which is around -3.1% sequence divergence per million years (Wares, 2001)，the time of divergence of the three lineages roughly dated back to 1.3-2.9 million years ago, i.e., within the Pleistocene epoch (12,000 y to 3 Mya). The Pleistocene is characterized by numerous glacial cycles in various magnitudes (Richmond and Fullerton, 19&6)，which led to severe change in climate and lowering of sea level. In the Northwest Pacific region, marginal seas such as East China Sea and South China Sea were formed owing to the lowering of sea level (Voris, 2000). The East China Sea shrank into a trough and was almost isolated from other waters by the Taiwan-Ryukyu land bridge. The exposure of Sunda Shelf and linkages between Philippines Islands on the other hand made the SCS a semi-enclosed region. However, the islands in northern part of the Philippines Sea, such as Ogasawara Islands, remain less affected by Pleistocene glaciations. These marginal seas, together with the northern part of the Philippine Sea located between Ogasawara Islands and Ryukyu Islands, may be the origin of the three lineages. This suggestion is possible because the contemporary distribution of the three cryptic lineages recovered in this study more or less coincide with the localities of the three isolated seas in the Pleistocene; high frequency of Ryukyu lineage may indicate the origin in the East China Sea, the Ogasawara lineage maybe the relict lineage from the refUgium in the northern part of Philippine Sea, which is less affected by glaciations, and the wide distribution of Taiwan lineage throughout the

78 southern part of NW Pacific may be originated from South China Sea refugium, or the imglaciated islands in southern part of the Philippines Sea (Fig. 4.1). In fact, genetic studies on spotted sea bass L. maculates (Liu et al, 2006) have suggested that the East China Sea may serve as a refugium during the ice age. Similarly, study on the redlip mullet Chelon haematocheilus in the Northwestern Pacific region (Liu et al, 2007) have shown that the marginal seas such as South China Sea and East China Sea may be important in promoting genetic differentiation in this fish species. These examples, together with the present finding shed lights on the possible role of marginal seas in affecting the evolutionary histories of some marine taxa in the NWP region. The adverse climate in the Northwest Pacific region and the presence of land bridges may also contribute to the genetic divergence observed between the three lineages. It was reported that the ocean current systems altered dramatically in the Pleistocene epoch, owing to the change in seaways in the region. The warm Kuroshio Current did not enter the Ilan Strait but deflected eastward to the Pacific Ocean during the Pleistocene (Ujiie et al., 2003). This event led to the drastic decrease in temperature and resulted in adverse climate in Northwest Pacific region. Together with habitat reduction, these incidents might greatly reduce the demographic size of ancestral C. moro populations, making inbreeding and bottleneck effect significant. On the other hand, the land bridge that connected Taiwan and the Ryukyu Islands, and the land bridges between different islands in the Philippines (Voris, 2000) seemed to be effective barriers in isolating the fragmentized populations and prohibiting gene flow between different refugia. The combination of the aforementioned factors is likely to induce population differentiation if isolation is effective and long enough. In fact, the Taiwan-Ryukyu land bridge is suggested as a

79 possible cause for population differentiation in the neon damselfish Pomacentrus coelestis (Liu et al., 2008). It has been suggested that the presence of contemporary barriers and difference in ecological niche are also responsible to induce allopatric and sympatric diversifications respectively in some marine taxa. For instance, sediment discharged from Amazon river serves as a physical barrier to induce endemism in coastal marine fishes (Rocha, 2003), and ecological speciation in marine snail Littorina saxatilis is suggested to be induced by adaptation to different shore levels (Conde-Padm et al., 2006). Yet we believe that these factors are less important in contributing strong diversification observed in C. mow. The major physical barrier in the NW Pacific region, the Kerama Gap that separates central Ryukyu from southern Ryukyu, is known to contribute to the genetic structuring of some marine taxa such as tideland snail Batillaria flectosiphonata (Kojima et al., 2003). However, the time of divergence for the three lineages in C. mow did not match the age of the formation of Kerama Gap, which is a relatively young geological feature (10,000 yr bp) induced by the northward collision of the Taiwan-Luzon Arc and Gagua Ridge (Ujiie et al., 1997). Thus the observed deep divergence between Taiwan lineage and the other two lineages is not the result of the formation of the Kerama Gap. For the possibility of sympatric differentiation, we have conducted molecular analyses for C. mow according to shore levels in three localities in Okinawa Island, and found that although the Ryukyu and Ogasawara lineages are sympatrically distributed, individuals of the two lineages show no significant difference in stratification in shore level (Fig. 4.9). The lack of evidence in differentiation of niche preference of the two lineages makes sympatric diversification a less likely option. Rather, postglacial demographic expansion and dispersal is a more preferable choice

80 in explaining the contemporary sympatric distribution pattern observed in C. mow. Although the genetic divergence observed between the three lineages (-4-9%) is comparable to the cut-off values of species boundary in some barnacle species, such as Euraphia (Wares, 2001), this evidence alone is too weak to define the three lineages as distinct species. The ecology and morphology of C. mow have not been well studied, and so far no morphological and ecological differentiations have been documented in this taxon. Moreover, whether hybridization occurs between the three lineages is unknown since no analysis of nuclear genes was performed. So far it is clear that the C. mow comprises three evolutionarily distinct lineages, yet the systematic status of the lineages can only be clarified when more morphological, ecological and molecular evidences are available.

4.4.2 Demographic history of the three lineages The Tajima's D values and Fs values are negative and significant in all three lineages, indicating that the lineages have undergone rapid population expansion from a relatively small population. The unimodal mismatch distribution curves of the Ogasawara and Ryukyu lineages further confirmed this conclusion. Similarly, Bayesian skyline plot analysis has revealed rapid demographic growth in all the three lineages, and the expansion of the lineages happened at different time periods. The rapid demographic expansions are probably due to range expansion after the Pleistocene glaciations. The environmental changes during the Pleistocene greatly affected marine fauna and flora by reducing their abundance and distribution range, followed by rapid expansion in population size and distribution range after glaciations (Hewitt, 2000). The severe change in climate, together with the loss of habitats in NW

81 Pacific region probably imposed strong pressure to this tropical barnacle species and led to drastic population decline during glaciations. After glaciations, the populations may expand their range rapidly from refugia and colonized the habitats that are previously not available (Hewitt, 2000), resulting in demographic expansion observed in the three lineages. High haplotype diversity together with relatively low nucleotide diversity observed in the three lineages also supports this suggestion. It was proposed that high haplotype diversity together with low nucleotide diversity is a sign of population expansion after a long period of low effective population size, during which the bottleneck effect followed by rapid population growth helps accumulate a wide variety of mutations separated by a small number of steps (Grant and Bowen, 1998). This phenomenon (high h and low 71) is observed in other taxa originated in the Pleistocene (Grant and Bowen, 1998), such as several butterfly fishes (Fauvelot et al, 2003) and shortfm mako (Heist et al, 1996). The finding in this study further supports the importance of this epoch in shaping genetic differentiation in marine organisms. The Bayesian skyline plots indicate that the three lineages have undergone demographic expansions, yet all took place in Middle Pleistocene, ranging from 450,000 - 250,000 years ago. All the three C. mow lineages have already attained a stable population before LGM (-20,000 years ago), during which the population size was comparable to the present day scale. Similar study has been done on the barnacle Tertraclita species complex (Tsang et al., 2011)，and the results show that Tertraclita kuroshioensis, a species that is more associated to island habitat, has undergone demographic expansion at approximately 250,000 years ago. However, other continental species such as Tetraclita squamosa and T. singaporensis have experienced demographic expansions in a relatively recent time scale (-100,000 and

82 50,000 years before present). From the present study, together with the study by Tsang et al, it seems that barnacle species that inhabit in islands are less affected by Pleistocene glaciations compared to their continental counterparts, as indicated by earlier population expansion time, and rather stable and large effective population size throughout the late Pleistocene including LGM. These findings also tend to to support the hypothesis of island habitats as potential refugia for marine taxa during the Pleistocene glaciations (Maggs et al., 2008). The mismatch distribution analysis of the Taiwan lineage appears to be bimodal and greatly deviated from the simulated curve for rapid demographic expansion. The inclusion of two lineages within Taiwan lineage should account for this. Phylogenetic analysis (Fig. 4.5) also shows that the Taiwan lineage can be further subdivided into two distinct groups. However, as the differentiation between the two groups is minute, they are analyzed as a whole and the second peak in the bimodal mismatch distribution curve is contributed by the pairwise differences between the two groups. Similar to the case in Taiwan linage, the bimodal mismatch distribution curve observed in Ogasawara lineage can be explained by the presence of two genetically divergent groups within the Ogasawara lineage. However, the second peak in the mismatch distribution curve of Ogasawara lineage is much smaller than the one in the mismatch distribution curve of Taiwan lineage. This is because one of the groups within Ogasawara lineage only consists of very few individuals, and so the high pairwise differences (the minor peak) generated from comparison between the two groups only account for very small portion of the mismatch distribution curve of Ogasawara lineage and so the observed dataset is not greatly deviated from the expected one.

83 4.4.3 Contemporary distribution and genetic connectivity of the three lineages While Pleistocene glaciations maybe crucial to the present distribution of the three lineages, the oceanographic patterns and the geography of NW Pacific region may be important in shaping the distribution and genetic connectivity of the lineages in different ways. The Taiwan lineage has the widest distribution range in the NW Pacific region. All individuals sampled from Fiji, Philippines, Paracel Islands and most of those from Taiwan belong to this lineage. However, within such a wide range, almost no population structuring was observed between populations in this lineage (with some exceptions, to be discuss later). The Kuroshio Current (KC) seems to be a crucial feature in maintaining the gene flow among these populations (Figs. 4.2 and 4.3). Originating from the North Equatorial Current (NEC) that flows westward, the warm KC starts off near northern Philippines and flows northward along the eastern coast of Taiwan (Chu, 1972). This flow likely maintains the genetic connectivity between populations in Philippines and the Taiwan. The side branch of KC that enters the South China Sea through the Luzon Strait between Taiwan and the Philippines (Hu et al, 2000) may probably facilitate larval transport between the Paracel Island population and the rest of the Taiwan lineage colonies. The KC further flows into the East China Sea through the Ilan ridge, moves along the Ryukyu Island chain and finally leaves the East China Sea through the Tokara Gap. This may homogenize the genetic structure of both the Ryukyu lineage and Ogasawara lineage between different localities in the Ryukyu Islands. After passing through the Tokara Gap in northern Ryukyu, the KC flows along the coast of mainland Japan, goes towards Ogasawara Islands and allows larval dispersal of Ryukyu lineage to the Ogasawara Islands. The movement of KC along the Ryukyu Islands produces a

84 countercurrent that flows in an opposite direction of KC. This countercurrent may be important in transporting pelagic barnacle larvae (mainly Ryukyu lineage individuals) from Ryukyu Islands southward to Taiwan. Even though the Kuroshio Current affects many localities in the region, this current seems not able to completely homogenize the three lineages, and discrete distribution ranges can still be observed in different lineages. It is found that the relative abundance of the Taiwan lineage declines drastically in the Okinawa Island, where it only accounts for less than 3% of the whole population in Okinawa Island that also appears to be the northern limit of the Taiwan lineage. Similarly, the relative abundance of the Ryukyu and Ogasawara lineages reduces from 67% and 29% on Okinawa Island respectively, to 17% and 4% in Turtle Island of Taiwan. The Kerama Gap, possibly contributes to the break observed by limiting larval dispersal between the populations on the two sides. The Kerama Gap is a deep ridge (>600 m) located between Southern and Central Ryukyu Islands (Choi et al., 2002). Besides the role of this gap in prohibiting gene flow between some terrestrial taxa such as the herb Ophiorrhiza japonica (Nakamura, 2010)，it has been suggested that the complicated hydrography, such as eddy currents in the Kerama Gap can probably act as barrier to larval dispersal and induces population structuring in marine taxa, such as tideland snail Batillaria flectosiphonata and Cerithidea spp. (Kojima et al., 2003, 2006). Results from AMOVA in this study also suggest the role of the Kerama Gap in prohibiting gene flow between the two sides of the Gap, as the separation accounts for most of the genetic variations (>60%). Results of pairwise O ST values on localities (Table 4.3) show that the population composition of the Ogasawara Island is significantly different from those observed in all other localities. The Ogasawara Island is actually the only locality

85 where the Ogasawara lineage individuals make up the majority of the population (-88%). In the other localities where it is found, the Ogasawara lineage only accounts for 20 to 46% of the total population. AMOVA result shows that the Northern Philippine Sea between the Ogasawara Islands and the Ryukyu Islands is imperative in generating population differentiation observed between them. The existence of seasonal eddy currents in the Northern Philippine Sea near the Izu-Bonin Ridge (Ihara et al., 2002) may be responsible for retaining the larval supply of C. moro (mainly Ogasawara lineage individuals) in Ogasawara Islands, and reduces the connectivity between Ogasawara Islands and the other localities. The Turtle Island of Taiwan also shows a difference in population, composition as compared to the Philippines. This observation may be attributed to the inclusion of quite a number of individuals of the Ryukyu and Ogasawara lineages in the Turtle Island, thus, resulting in the significant positive Ost value between the island the Philippines where only the Taiwan lineage is found. To summarize, the present molecular study reveals that C. moro in the Northwest Pacific exhibits prominent genetic structuring, which is probably brought about by glacial cycles during the Pleistocene. However, when comparing with barnacles that are associated to continental habitat, the island inhabiting C. moro are more stable in terms of population size, and recovered earlier under the stress of the Pleistocene glaciations. On the other hand, contemporary ocean current patterns are crucial in shaping the present distribution ranges of the three lineages. The complex hydrography in the Kerama Gap between Taiwan and Ryukyu seems to be an essential factor in limiting homogenization between the southern Taiwan lineage and the two northern lineages, leading to the distribution pattern observed.

86 Chapter 5

Concluding remarks

The present molecular phylogenetic study is the first comprehensive molecular research that aims at elucidating the intrafamilal relationship of barnacles in family Chthamalidae using a molecular approach. The results from this study have revealed discrepancies from the traditional systematics of Chthamalidae, for which none of the three subfamilies appears to be monophyletic. It is also shown that the most species rich genus in Chthamalidae, Chthamalus, is of polyphyletic origin, as Jehlius and Microeuraphia clustered with different species of Chthamalus. These findings have posted questions on the validity of some key morphological characters that are traditionally used in Chthamalidae classification, and, at the same time, pointed out the needs for major taxonomic revision of this family, including both the subfamilial and genus levels. Moreover, this phylogenetic study has successfully demonstrated the possible role of molecular data in resolving the phylogeny of morphologically plastic lineages. And, this molecular study has provided new insights on barnacle evolution, by falsifying the hypothesis of reduction in parietal plate number as an evolutionary trend in the Chthamalidae. Regarding the population genetics of Chthamalus mow, it is one of the very few phylogeographic studies of marine organisms in the NWP region. Despite recovering unexpected genetic diversity with three lineages of C. mow, this study contributes to the understanding on the role of different environmental factors in affecting the population structure of C. moro in NWP. While Pleistocene glaciations are likely to cause diversification in C. mow, the most prevalent current in the region,

87 the Kuroshio Current are likely to maintain gene flow between populations within the distribution range of the three lineages respectively. The Kerama Gap in Ryukyu Islands may also play a role in the contemporary distribution of the three lineages, probably by preventing large scale mixing between the lineage south of the Ryukyu Islands and the other two lineages in the north. To conclude, as a typical island intertidal inhabitant in NWP, the present phylogeographic study on C. mow provides insights on how the population genetic structure of intertidal organisms on islands are influenced by various factors, such as historical events, oceanographic patterns or physical barriers in the region.

88

V. , References

Alvarez EA, Hines H, Reaka-Kudla ML (1995) The effects of parasitism by the barnacle Loxothylacus panopaie (Gissler) (Cirripedia: Rhizocephala) on growth and survival of the host crab Rhithwpanopeus harrisii (Gould) (Brachyura: Xanthidae). J Exp Mar Biol Ecol 192: 221-232 Anderson DT (1994) Barnacles - Structure, function, development and evolution. Chapman and Hall, London Ara BR, Yamaguchi T, Watabe S (2004) Molecular phylogeny of thoracican barnacles based on the mitochondrial 12S and 16S rRNA genes. Sess Org 21: 47-54 Avise JC (2000) Phylogeography. Harvard University Press, Boston, Avise JC (2004) Molecular markers, natural history and evolution. 2nd edition. Sinauer Associates, Sunderland, Massachusetts Bane GW, Zullo VA (1980) Observations on a stranded goosebeaked whale {Ziphius cavirostris, Cuvier 1823) and its ectocommensal barnacles {Xenobalanus globicipitis). J Elisha Mitchell Sci Soc 96: 1-3 Barley AJ，Spiks PQ, Thomson RC, Shaffer HB (2010) Fourteen nuclear genes provide phylogenetic resolution for difficult nodes in the turtle tree of life. Mol Phylogenet Evol 55: 1189-1194 Barry JP, Baxter CH, Sagarin RH, Oilman SE (1995) Climate-related, long-term faunal changes in a California rocky intertidal community. Science 267: - 672-675 •“ Benzie J AH, Bailment E, Forbes AT, Demetriades NT, SugamaK, Haryanti, Moria S (2002) Mitochondrial DNA variation in Indo-Pacific populations of the giant

89 tiger prawn, Penaeus monodon. Mol Ecol 11: 2553-2569 Bielecki J, Chan BKK, Hoeg JT, Sari A (2009) Antennular sensory organs in cyprids of balanomorphan cirripedes: Standardizing terminology using Megabalanus rosa. Biofouling 25: 203-214 Brown WM，George M，Wilson AC (1979) Rapid evolution of animal mitochondrial DNA. Proc Natl Acad Sci USA 76: 1967-1971 Buckeridge JS (2000) Neolepas osheai sp. no v., a new deep-sea vent barnacle (Cirripedia: Pedunculata) from Brothers caldera, south-west Pacific Ocean. N Z J Mar Freshw Res 34: 409-418 Buckeridge JS, Newman WA (1992) A reexamination of Waikalasma (Cirripedia: Thoracica) and its significance in balanomorph phylogeny. J Paleontol 66: 341-345 Buhl-Mortensen L, Hoeg JT (2006) Reproduction and larval development in three scalpellid barnacles {Scalpellum scalpellum, Ornatoscalpellum stroemii and Arcoscalpellum michelottiamm, Crustacea: Cirripedia: Thoracica): Implications for reproduction and dispersal in the deep-sea. Mar Biol 149: 829-844 Callow ME, Callow JA (2002) Marine biofouling: a sticky problem. Biologist 49: 10-14 Campo D, Molares J, Garcia L, Femandez-Rueda P, Garcia-Gonzalez C, Garcia-Vazquez E (2010) Phylogeography of the European stalked barnacle {Pollicipes pollicipes)'. identification of glacial refugia. Mar Biol 157: 147-156 Cavalli-Sforza LL (1998) The DNA revolution in population genetics. Trends Genet 14: 60-65 Chan BKK, Hung OS (2005) Cirral length of the acom barnacle Tetraclita japonica

90 (Cirripedia: Balanomorpha) in Hong Kong: effect of wave exposure and tidal height. J Crustacean Biol 25: 329-332 Chan BKK, Williams GA (2004) Population dynamics of the acom barnacles, Tetraclita squamosa and Tetraclita japonica (Cirripedia: Balanomorpha) in Hong Kong. Mar Biol 146: 149-160 Cheang CC, Tsang LM, Ng WC, Williams GA, Chu KH, Chan BKK (submitted) Lack of genetic differentiation of the cold-water barnacle, Chthamalus challengeri, in the NW Pacific: persistence through the Last Glacial Maximum. Choi BH, Kim KO, Bum HM (2002) Digital bathymetric and topographic data for neighboring seas of Korea. J Korean Soc Coastal and Ocean Engrs 14: 41-50 (in Korean) Chopin T，Bird CJ, Murphy CA，Osborne AJ, Patwary MU, Floch JY (1996) A molecular investigation of polymorphism in the North Atlantic red alga Chondrus crispus (Gigartinales). Phycol Res 44: 69-80 Chu TY (1972) A study on the water exchange between Pacific Ocean and the South China Sea. Acta Oceanogr Taiwan 2: 11-24 Colgar DJ, McLauchlan A, Wilson GDF, Livingston SP, Edgecombe GD, Macaranas J, Cassis G, Gray MR (1998) Histone H3 and U2 snRNA DNA sequences and molecular evolution. Aust J Zool 46: 419-437 Conde-Padm P, Carvajal-Rodriguez A, Carballo M, Caballero A, Rolan-Alvarez E (2007) Genetic variation for shell traits in a direct-developing marine snail involved in a putative sympatric ecological speciation process. Evol Ecol 21: 635-650 Connell JH (1961) The influence of interspecific competition and other factors on the distribution of the barnacle . Ecology 42: 710-723

91 Cracraft J，Prum RO (1988) Patterns and processes of diversification: speciation and historical congruence in some neotropical birds. Evolution 42: 603-620 Crisp DJ，Southward AJ, Southward EC (1981) On the distribution of the intertidal barnacles Chthamalus stellatus, Chthamalus montagui and Euraphia depressa. J Mar Biol Ass UK 61: 359-380 Dando PR, Southward AJ (1980) A new species of Chthamalus (Crustacea: Cirripedia) characterized by enzyme electrophoresis and shell morphology: with a revision of other species of Chthamalus from the western shores of the Atlantic. J Mar Biol Assoc UK 60: 787-831 Darwin C (1851) A monograph of the fossil Lepadidae, or pedunculated cirripedes of Great Britain. Paleontography Society, London Darwin C (1854) A monograph on the sub-class Cirripedia, with figures of all the species. The Balanidae, (or sessile cirripedes); the Vermcidae, etc. The Ray Society, London Darwin C (1859) On the origin of species. Penguin, London Dean TA, Hurd LE (1980) Development in an estuarine fouling community: the influence of early colonists on later arrivals. Oecologia 46: 295-301 Doyle JA, Donoghue MJ (1986) Seed plant phylogeny and the origin of angiosperms: an experimental cladistic approach. Bot Rev 52: 321-431 Excoffier L, Laval G, Schneider S (2005) Arlequin ver. 3.0: an integrated software package for population genetics data analysis. Evol Bioinform Online 1:47-50 Fauvelot C, Bemardi G, Planes S (2003) Reductions in the mitochondrial DNA diversity of coral reef fish provide evidence of population bottlenecks resulting from Holocene sea-level change. Evolution 57: 1571-1583 Field KG, Olsen GJ，Lane DJ, Giovannoni SJ, Ghiselin MT, Raff EC, Pace NR, Raff

92 RA (1988) Molecular phylogeny of the animal kingdom. Science 239: 748-753

Fisher T，Katcoff DJ，Achituv Y (2004) A phylogenetic study of chthamaloids (Cirripedia; Thoracica; Chthamaloidae) based on 16S rDNA and COI sequence analysis. Biol J Linn Soc 83: 39-45

Foster BA (1978) The marine fauna of New Zealand: barnacles (Cirripedia: Thoracica). Mem NZ Ocean Inst 69: 1-160 Foster BA, Buckeridge JS (1987) Barnacle paleontology. In: Southward AJ (Ed.), Barnacle Biology, Crustacean Issues 5. Balkema, Rotterdam Frick MG，Williams KL, Robinson M (1998) Epibionts associated with nesting loggerhead sea turtles (Caretta caretta) in Georgia USA. Herpetol Rev 29: 211-214 Fu YX (1997) Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics 147: 915-925 Funk WC, Blouin MS, Com PS, Maxell BA, Pilliod DS, Amish S, Allendorf FW (2005) Population structure of Columbia spotted frogs {Rana luteiventris) is strongly affected by the landscape. Mol Ecol 14: 483-496 Glenner H，Grygier MJ, Heeg JT, Jensen PG, Schram FR (1995) Cladistic analysis of the Cirripedia Thoracica (Crustacea: Thecostraca). Zool J Linn Soc 114: 365-404 Glenner H, Hebsgaard MB (2006) Phylogeny and evolution of life history strategies of the parasitic barnacles (Crustacea, Cirripedia, Rhizocephala). Mol Phylogenet Evol 41: 528-538 Goldstein DB, Linares AR，Cavalli-Sforza LL, Feldman MW (1995) Genetic absolute dating based on microsatellites and the origin of modem humans. Proc

93 Natl Acad Sci USA 92: 6723-6727 Gomes-Filho JGF, Hawkins SJ，Aquino-Souza R, Thompson RC (2010) Distribution of barnacles and dominance of the introduced species Elminius modestus along two estuaries in South-West England. Marine Biodiversity Records 3: e58 Grant WS, Bowen BW (1998) Shallow population histories in deep evolutionary lineages of marine fishes: Insights from sardines and anchovies and lessons for conservation. J Hered 89: 415-427 Guerry AD, Menge BA, Dunmore R (2009) Effects of consumers and enrichment on abundance and diversity of benthic algae in a rocky intertidal community. J Exp Mar Biol Ecol 369: 155-164 Haeckel E (1901) The riddle of the universe at the close of the nineteenth century. Watts, London Harris DJ, Maxson L, Braithwaite LF, Crandall KA (2000) Phylogeny of the thoracican barnacles based on 18S rDNA sequences. J Crustacean Biol 20: 393-398 Harrison JS (2004) Evolution, biogeography, and the utility of mitochondrial 16S and COI genes in phylogenetic analysis of the crab genus Austinixa (Decapoda: Pinnotheridae). Mol Phylogenet Evol 30: 743-754 Harrison RG (1989) Animal mitochondrial DNA as a genetic marker in population and evolutionary biology. Trends Ecol Evol 4: 6-11 Healy JM，Anderson DT (1990) Sperm ultrastructure in the Cirripedia and its phylogenetic significance. Rec Aust Mus 42: 1-26 Heist EJ, Musick JA, Graves JE (1996) Genetic population structure of the shortfm mako {Isurus oxyrinchus) inferred from restriction fragment length polymo^hism analysis of mitochondrial DNA. Can J Fish Aquat Sci 53:

94 583-588 Hewitt GM (1996) Some genetic consequences of ice ages, and their role in divergence and speciation. Biol J Linn Soc 58: 247-276 Hewitt GM (2000) The genetic legacy of the Quaternary ice ages. Nature 405: 907-913 Hillis DM (1987) Molecular versus morphological approaches to systematics. Annu Rev Ecol Evol Syst 18: 23-42 Hou Z, Fu J, Li S (2007) A molecular phylogeny of the genus Gammarus (Crustacea: Amphipoda) based on mitochondrial and nuclear gene sequences. Mol Phylogenet Evol 45: 596-611 Hu JY, Kawamura H, Hong HS, Qi YQ (2000) A review on the currents in the South China Sea: seasonal circulation, South China Sea Warm Current and Kuroshio intrusion. J Oceanogr 56: 607-624 Hurst GDD, Jiggins FM (2005) Problems with mitochondrial DNA as a marker in population, phylogeographic and phylogenetic studies: the effects of inherited symbionts. Proc R Soc B 272: 1525-1534 Ihara C, Kagimoto T, Masumoto Y, Yamagata T (2002) Eddy formation near the Izu Ogasawara Ridge and its link with seasonal adjustment of the subtropical gyre in the Pacific. J Korean Soc Oceanogr 37: 134-143 Jarrett JN (2009) Predator-induced defense in the barnacle Chthamalus fissus. J Crustacean Biol 29: 329-333 Kawai T, Tokeshi M (2004) Variable modes of facilitation in the upper intertidal: goose barnacles and mussels. Mar Ecol Prog Ser 272: 203-213 Kim W, Abele LG (1990) Molecular phylogeny of selected decapod crustaceans based o^ 18S rRNA nucleotide sequences. J Crustacean Biol 10: 1-13

95 Kizaki K, Oshiro I (1977) Paleogeography of the Ryukyu Islands. Mar Sci Monthly 9: 542-549 (in Japanese with English abstract) Kizaki K，Oshiro I (1980) The origin of the Ryukyu Islands. In: Kizaki K (ed) Natural History of the Ryukyus. Tsukiji-shokan, Tokyo (in Japanese) Kjer K (2004) Aligned 18S and insect phylogeny. Syst Biol 53: 506-514 Klepal W (1985) Ibla cumingi (Crustacea, Cirripedia) - A gonochoristic species (Anatomy, dwarfing and systematic implications). P.S.Z.N.I. Mar Ecol 6: 47-119 Knutsen H, Jorde PE, Andre C, Stenseth NC (2003) Fine-scaled geographical population structuring in a highly mobile marine species: the Atlantic cod. Mol Ecol 12: 385-394 Kojima S, Kamimura S, lijima A, Kimura T, Kurozumi T and Furota T (2006) Molecular phylogeny and population structure of tideland snails in the genus Cerithidea around Japan. Mar Biol 149: 525-535 Kojima S, Kamimura S，Kimura T, Hayashi I，lijima A, Furota T (2003) Phylogenetic relationships between the tideland snails Batillaria flectosiphonata in the Ryukyu Islands and B. multiformis in the Japaneses Islands. Zool Sci 20: 1423-1433 Lauder GV, Liem KF (1983) The evolution and interrelationships of the actinopterygian fishes. Bull Mus Comp Zool 150: 95-197 Lavery S, Chan TY, Tarn YK, Chu KH (2004) Phylogenetic relationships and evolutionary history of the shrimp genus Penaeus s.l. derived from mitochondrial DNA. Mol Phylogenet Evol 31: 39-49 Lessios HA, Kane J, Robertson DR (2003) Phylogeography of the pantropical sea urchin Tripneustes: contrasting patterns of population structure between oceans.

96 Evolution 57: 2026-2036 Li CH, Orti G，Zhang G, Lu GQ (2007) A practical approach to phylogenomics: the phylogeny of ray-finned fish (Actinopterygii) as a case study. BMC Evol Biol 7:44 Lim JT, Hwang UW (2006) The complete mitochondrial genome of Pollicipes mitella (Crustacea, Maxillopoda, Cirripedia): non-monophylies of Maxillopoda and Crustacea. Mol Cells 22: 314-322 Liu JX，Gao TX, Wu SF, Zhang YP (2007) Pleistocene isolation in the northwestern Pacific marginal seas and limited dispersal in a marine fish, Chelon haematocheilus (Temminck & Schlegel, 1845). Mol Ecol 16: 275-288 Liu JX, Gao TX, Yokogawa K, Zhang YP (2006) Differential population structuring and demographic history of two closely related fish species, Japanese sea bass {Lateolabrax japonicus) and spotted sea bass {Lateolabrax maculatus) in Northwestern Pacific. Mol Phylogenet Evol 39: 799-811 Liu SY，Kokita T, Dai CF (2008) Population genetic structure of the neon damselfish {Pomacentrus coelestis) in the northwestern Pacific Ocean. Mar Biol 154: 745-753 Lively C (1986) Predator-induced shell dimorphism in the acom barnacle . Evolution 40: 232-242 Lopez BA, Guaitro SY, Lopez DA (2007) Density-dependent effects on the cirral structure of the barnacle, Jehlius cirratus (Darwin, 1854) (Cirripedia). Crustaceana 80: 793-801 Lourie SA, Green DM, Vincent ACJ (2005) Dispersal, habitat differences, and comparative phylogeography of Southeast Asian seahorses (Syngnathidae: Hippocqmpus). Mol Ecol 14: 1073-1094

97 Machida RJ, Miya MU, Nishida M, Nishida S (2002) Complete mitochondrial DNA sequence of Tigriopus japonicus (Crustacea: Copepoda). Mar Biotechnol 4: 406-417 Martin JW, Davis GE (2001) An updated classification of the Recent Crustacea. Nat Hist Mus Los Angeles Cy Sci Ser 39: 1-124 McManus JW (1985) Marine speciation, tectonics and sea-level changes in south east Asia. Proc 5th Int Coral Reef Con 4: 133-138 Maggs CA, Castilho R，Foltz D，Henzler C, Jolly MT, Kelly J, Olsen J, Perez KE, Stam W, Vainola R，Viard F, Wares J (2008) Evaluating signatures of glacial refligia for North Atlantic benthic marine taxa. Ecology 89: S108-S122 Mizrahi L，Achituv Y, KatcoffDJ, Perl-Treves R (1998) Phylogenetic position oilbla (Cirripedia: Thoracica) based on 18S rDNA sequence analysis. J Crustacean Biol 18: 363-368 Murosaki T，Noguchi T, Kakugo A, Putra A, Kurokawa T, Furukawa H, Osada Y, Gong JP, Nogata Y, Matsumura K, Yoshimura E，Fusetani N (2009) Antifouling activity of synthetic polymer gels against cyprids of the barnacle {Balanus amphitrite) in vitro. Biofouling 25: 313-320 Murphy WJ，Eizirik E, Johnson WE, Zhang YP, Ryder OA, O'Brien SJ (2001) Resolution of the early placental mammal radiation using Bayesian phylogenetics. Nature 409: 614-618 Nakamura K，Denda T, Kokubugata G, Suwa R, Yang ATY, Peng CI, Yokota M (2010) Phylogeography of Ophiorrhiza japonica (Rubiaceae) in continental islands, the Ryukyu Archipelago, Japan. J Biogeogr 37: 1907-1918 Navarrete S, Castilla JC (1990) Barnacle walls as mediators of intertidal mussel recruitment: effects of patch size on the utilization of space. Mar Ecol Prog Ser

98 68: 113-119 Newman WA (1979) A new scalpellid (Cirripedia); a Mesozoic relic living near an abyssal hydrothermal spring. Trans San Diego Soc Nat Hist 19: 153-167 Newman WA (1987) Evolution of cirripeds and their major groups. In: Southward AJ (Ed.), Barnacle Biology, Crustacean Issues 5. Balkema, Rotterdam Newman WA, Ross A (1976) Revision of the balanomorph barnacles; including a catalog of the species. San Diego Soc Nat Hist Mem 9: 1-108 Nilsson-Cantell CA (1921) Cirripeden Studien Zur Keimtnis der Biologie, Anatomic, und Systematick dieser Gruppe. Zool Bidr Uppsala 7: 75-390 Nishikawa A (2008) Degree and pattern of gene flow in several scleractinian corals in the Ryukyu Archipelago, southern Japan. Pac Sci 62:413-421 Ota H (1998) Geographic patterns of endemism and speciation in amphibians and reptiles of the Ryukyu Archipelago, Japan, with special reference to their paleogeographical implications. Res Popul Ecol 40: 189-204 Ouborg NJ, Piquot Y，van Groenendael JM (1999) Population genetics, molecular markers and the study of dispersal in plants. J Ecol 87: 551-568 Palumbi SR (1994) Genetic divergence, reproductive isolation, and marine speciation. Annu Rev Ecol Evol Syst 25: 547-572 Palumbi SR, Grabowsky G, Duda T, Geyer L, Tachino N (1997) Speciation and population genetic structure in tropical Pacific sea urchins. Evolution 51: 1506-1517 Perdices A, Doadrio I，Economidis PS, Bohlen J, Banarescu P (2003) Pleistocene effects on the European freshwater fish fauna: double origin of the cobitid genus Sabanejewia in the Danube basin (Osteichthyes: Cobitidae). Mol Phylogenet Evol 26: 289-299

99

J Pfe-Losada M，Harp M, Hoeg JT, Achituv Y，Jones D, Watanabe H, Crandall KA (2008) The tempo and mode of barnacle evolution. Mol Phylogenet Evol 46: 328-346 Perez-Losada M, Hoeg JT, Crandall KA (2004) Unraveling the evolutionary radiation of the Thoracican barnacles using molecular and morphological evidence: A comparison of several divergence time estimation approaches. Syst Biol 53: 244-264 Perl-Treves R，Mizrahi L, Katcoff DJ, Achituv Y (2000) Elucidation of the phylogenetic relationships of three thecostracans, Verruca, Paralepas, and Dendrogaster based on 18S rDNA sequences. J Crust Biol 20: 385-392 Perry DM (1988) Effects of associated fauna on growth and productivity in the red mangrove. Ecology 69: 1064-75 Pilsbry HA (1916) The sessile barnacles (Cirripedia) contained in the collections of the U.S. National Museum; including a monograph of the American species. Bull US Natl Mus 93: 1-366 Pope EC (1965) A review of Australian and some Indomalayan Chthamalidae (Crustacea: Cirripedia). Proc Linn Soc New South Wales 90: 10-77 Posada D, Crandall KA (1998) Modeltest: Testing the model of DNA substitution. Bioinformatics 14: 817-818 Posada D, Crandall KA (2001) Selecting the best-fit model of nucleotide substitution. Syst Biol 50: 580-601 Qiu B, Imasato N (1990) A numerical study on the formation of the Kuroshio Counter Current and the Kuroshio Branch Current in the East China Sea. Cont Shelf Res 10: 165-184 Quinteiro J, Rodriguez-Castro J, Rey-Mendez M (2007) Population genetic structure

100 of the stalked barnacle Pollicipes pollicipes (Gmelin, 1789) in the northeastern Atlantic: influence of coastal currents and mesoscale hydrographic structures. Mar Biol 153:47-60 Rainbow PS, White SL (1990) Comparative accumulation of cobalt by three crustaceans: a decapod, an amphipod and a barnacle. Aquat Toxicol 16: 113-26 Rambaut A, Dmmmond AJ (2007) Tracer vl.4. Institute of Evolutionary Biology, University of Edinburgh, Edinburgh, Scotland Rawson PD, Macnamee R, Frick MG, Williams KL (2003) Phylogeography of the coronulid barnacle, Chelonibia testudinaria, from loggerhead sea turtles, Caretta caretta. Mol Ecol 12: 2697-2706 Reid DG, Lai K，Mackenzie-Dodds J, Kaligis F，Littlewood DTJ, Williams ST (2006) Comparative phylogeography and species boundaries in Echinolittorina snails in the central Indo-West Pacific. J Biogeogr 33: 990-1006 Richmond GM (1986) Stratigraphy and chronology of glaciations in Yellowstone National Park. Quat Sci Rev 5: 83-98 Richmond GM, Fullerton DS (1986) Summation of Quaternary glaciations in the United States of America. Quat Sci Rev 5: 183-196 Robin ED, Wong R (1988) Mitochondrial DNA molecules and virtual number of mitochondria per cell in mammalian cells. J Cell Physiol 136: 507-513 Rocha LA (2003) Patterns of distribution and processes of speciation in Brazilian reef fishes. J Biogeogr 30: 1161-1171 Rohfritsch A, Borsa P (2005) Genetic structure of Indian scad mackerel, Decapterus russelli: Pleistocene vicariance and secondary contact in the Central Indo-West .. Pacific Seas. Heredity 95: 315-326 Rokas A, Hojland PW (2000) Rare genomic changes as a tool for phylogenetics.

101 Trends Ecol Evol 15: 454-459 Ronquist F, Huelsenbeck JP (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572-1574 Ross A (1971) A new genus of Chthamalidae (Cirripedia) from the southeastern Pacific island of San Ambrosio. Trans San Diego Soc Nat Hist 16: 265-278 Ross A, Newman WA (2000) Pyrgoma kuri Hoek, 1913: a case study in morphology and systematics of a symbiotic coral barnacle (Cirripedia: Balanomorpha). Contrib Zool 68: 245-260 Ruedemann R (1918) The phylogeny of the acom barnacles. Proc Natl Acad Sci USA 4: 382-384 Sanderson MJ, Donoghue MJ (1989) Patterns of variation in levels of homoplasy. Evolution 43: 1781-1795 Simmons MP, Ochoterena H (2000) Gaps as characters in sequence based phylogenetic analyses. Syst Biol 49: 369-381 Simon-Blecher N, Huchon D，Achituv Y (2007) Phylogeny of coral-inhabiting barnacles (Cirripedia; Thoracica; Pyrgomatidae) based on 12S, 16S and 18S rDNA analysis. Mol Phylogenet Evol 44: 1333-1341 Sotka EE, Wares JP, Barth JA, Grosberg RK, Palumbi SR (2004) Strong genetic dines and geographical variation in gene flow in the rocky intertidal barnacle Balanus glandula. Mol Ecol 13: 2143-56 Southward AJ (1991) Forty years of changes in species composition and population density of barnacles on a rocky shore near Plymouth, England, UK. J Mar Biol Assoc UK 71: 495-514 „ Southward AJ, Newman WA (2003) A review of some common Indo-Malayan and westem^Pacific species of Chthamalus barnacles (Crustacea: Cirripedia). J Mar

102 Biol Assoc UK 83:797-812 Spears T, Abele LG, Applegate MA (1994) Phylogenetic study of cirripedes and selected relatives (Thecostraca) based on 18S rDNA sequence analysis. J Crust Biol 14: 641-656 Swofford D (2002) Phylogenetic Analysis Using Parsimony (PAUP) v. 4.0bl0. Tajima F (1989) Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 123: 585-595 Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 24: 1596-1599 Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 4673-4680 Tsang LM，Chan BKK, Wu TH，Ng WC, Chatterjee T, Williams GA，Chu KH (2008) Population differentiation of the barnacle, Chthamalus malayensis: postglacial colonization and recent connectivity across Pacific and Indian Oceans. Mar Ecol Prog Ser 364: 107-118 Tsang LM, Chan BKK, Shih FL, Chu KH, Chen CA (2009) Host-associated speciation in the coral barnacle Wanella milleporae (Cirripedia: Pyrgomatidae) inhabiting the Millepora coral. Mol Ecol 18: 1463-1475 Tsang LM, Wu TH, Ng WC, Williams GA, Chan BKK, Chu KH (2011) Comparative phylogeography of Indo-West Pacific intertidal barnacles. In: Koenemann S, Schubart CD, Held C (Eds.) Phylogeography and Population Genetics in Crustacea. CRC Press, Baton Raton, FL (in press) Tzeng CS, Lin YS, Lin SM, Wang TY, Wang FY (2006) The phylogeography and

103

• I population demographics of selected freshwater fishes in Taiwan. Zool Stud 45: 285-297 Ujiie H (1990) Geological history of the Ryukyu Island Arc. In: Ujiie H (Ed.) Nature of Okinawa: Geomorphology and geology. Hirugisha, Naha (in Japanese) Ujiie H, Nakamura T, Miyamoto Y (1997) Holocene turbidite cores from the southern Ryukyu Trench slope: Suggestions of periodic earthquakes. J Geol Soc Jpn 103: 590-603 Ujiie H, Ujiie Y (1993) Late Quaternary course changes of the Kuroshio Current in the Ryukyu Arc region, northwestern Pacific Ocean. Mar Micropaleontol 37: 23-40 Ujiie Y, Ujiie H，Taira A, Nakamura T, Oguri K (2003) Spatial and temporal variability of surface water in the Kuroshio source region, Pacific Ocean, over the past 21,000 years: Evidence from planktonic foraminifera. Mar Micropaleontol 49: 335-364 Utinomi H (1968) A revision of the deep-sea barnacles Pachylasma and Hexelasma from Japan, with a proposal of new classification of the Chthamalidae (Cirripedia, Thoracica). Publ Seto Mar Biol Lab, 16: 21-39 Venter JC, Adams MD, Myers EW, Li PW，Mural RJ, Sutton GG, Smith HO, Yandell M，Evans CA, Holt RA, et al. (2001) The sequence of the human genome. Science 291: 1304-1351 Voris HK (2000) Maps of Pleistocene sea levels in Southeast Asia: shorelines, river systems and time durations. J Biogeogr 27: 1153-1167 Wares JP (2001) Patterns of speciation inferred from mitochondrial DNA in North American Chthamalus (Cirripedia: Balanomorpha: Chthamaloidea). Mol Phylogenet Evol 18: 104-116

104 Wares JP, Cunningham CW (2001) Phylogeography and historical ecology of the North Atlantic intertidal. Evolution 12: 2455-2469 Wares JP, Pankey MS, Pitombo F, Daglio LG, Achituv Y (2009) A "shallow phylogeny" of shallow barnacles {Chthamalus). PLos One 4: e5567 Wiegemann M (2005) Adhesion in blue mussels {Mytilus edulis) and barnacles (genus Balanus): mechanisms and technical applications. Aquat Sci 67: 166-176 Wilson AC, Cann RL, Carr SM, George M，Gyllensten UB, Helm-Bychowski KM, Higuchi RG，Palumbi SR, Prager EM, Sage RD, Stoneking M (1985) Mitochondrial DNA and two perspectives on evolutionary genetics. Biol J Linn Soc 26: 375-400 York KL, Blacket MJ, Appleton BR (2008) The Bassian Isthmus and the major ocean currents of southeast Australia influence the phylogeography and population structure of a southern Australian barnacle Catomerus polymerus (Darwin). Mol Ecol 17: 1948-1961 Zabin CJ (2009) Battle of the barnacle newcomers: niche compression in invading species in Kaneohe Bay, Oahu, Hawaii； Mar Ecol Prog Ser 381: 175-182

A Zardus JD, Hadfield MG (2005) Multiple origins and incursions of the Atlantic barnacle Chthamalusproteus in the Pacific. Mol Ecol 14: 3719-3733 Zhang DX, Hewitt GM (2003) Nuclear DNA analyses in genetic studies of populations: practice, problems, and prospects. Mol Ecol 12: 563-584 Zhou XJ，Zhang Z, Xu Y, Jin CL, He HP, Hao XJ, Qian PY (2009) Flavone and isoflavone derivatives of terrestrial plants as larval settlement inhibitor of the barnacle Balanus amphitrite. Biofouling 25: 69-76 Zullo VA (1963) A classification and phylogeny of the Chthamalidae (Cirripedia:

105 Thoracica). XVI International Congress of Zoology 1: 190 Zullo VA (1992) Revision of the balanid barnacle genus Concavus Newman, 1982， with a description of a new subfamily, two new genera, and eight new species. J Paleontol 66: 1-46

• \

106 _ 丨 CUHK Libraries

.• i MMMMM004864717