Causes and consequences of natural hybridisation among coral reef butterflyfishes (Chaetodon: Chaetodontidae)

Natural hybridisation is the successful interbreeding of individuals from different populations, distinguishable through one or more heritable characters, and is a widespread phenomenon in the plant and animal kingdoms. The development of hybridisation theories has largely been based on studies in terrestrial and freshwater ecosystems. Hybridisation was traditionally considered rare and unimportant in marine systems and therefore received little attention. Recently however, there has been a surge of reported marine hybrids, particularly in corals and reef fishes. The ecological promoters and evolutionary and adaptive consequences of reef fish hybridisation are yet to be thoroughly evaluated. Butterflyfishes (f. Chaetodontidae) form a disproportionate number of hybrids and therefore represent an appropriate model group to investigate hybridisation in reef fishes. This thesis examines the causes and consequences of hybridisation in reef fishes and focuses on butterflyfishes (genus Chaetodon) at Christmas Island (Indian Ocean), a global hotspot for reef fish hybridisation. The aims of this thesis were to i) review the incidence and ecological/behavioural precursors of hybridisation in reef fishes, while providing a tentative framework for conducting studies within hybrid zones; ii) develop a microsatellite toolkit for species of the Chaetodon genus; iii) compare the ecology, behaviour and population genetics of hybridising sister species of butterflyfishes in order to, not only provide a snapshot of the evolutionary consequences of hybridisation in this group, but also determine which processes are likely to promote it; iv) use a comparative life history approach to determine the fitness of butterflyfish hybrids relative to their parental species. Chapter 1 reviews the current knowledge of hybridisation with a focus on marine fishes. Hybridisation was found to be highly prevalent in marine fish, despite previous assertions of rarity, and showed a taxonomic as well as latitudinal bias. Further, the current marine fish hybridisation literature was found to be largely lacking ecological and behavioural data, in contrast with freshwater counterparts, therefore highlighting the need for a framework to fill the data gap in order to better accompany the wealth of genetic data produced in the assessment of hybridisation. The development of a molecular toolkit, necessary for the rest of our investigations, is presented in Chapter 2. Twenty microsatellite loci were developed using 454 sequencing, to apply to the population genetic analysis of the Chaetodon guttatissimus × C. punctatofasciatus complex. This was done to facilitate direct comparison of the genetic underpinnings of hybridisation in this group to those of another previously studied group (C. trifasciatus and C. lunulatus), for which species-specific microsatellite loci had been designed and used. Chapter 3 uses the molecular toolkit and experimental framework outlined in the previous chapters to examine hybridisation between two butterflyfish sister species, Chaetodon guttatissimus and C. punctatofasciatus. The largely overlapping spatial and dietary ecologies of these species favour heterospecific encounters. Lack of assortative mating and local rarity of C. punctatofasciatus promote the formation of heterospecific breeding pairs. Analyses of mtDNA and microsatellite DNA were consistent with the hybrid status of the intermediately coloured hybrids. Maternal contribution to hybridisation in this complex was bidirectional, and introgression by C. punctatofasciatus mtDNA was detected in C. guttatissimus individuals within and beyond the hybrid zone (almost 1000 km to the west), potentially indicating a Pacific invasion of an Indian Ocean species genome. The comparisons drawn with previous work on hybrdising Chaetodon trifasciatus and C. lunulatus showed that, despite being driven by similar factors, hybridisation in reef fishes can have varying evolutionary consequences, possibly due to the magnitude of the genetic distance between hybridising species. Chapter 4 evaluates hybrid fitness in both Chaetodon hybridising groups presented in the previous study. Histology confirmed the reproductive viability of hybrids, and liver lipid analyses showed that hybrid condition was not different from parental species. Further, otolith data highlighted no difference in growth rate and maximum length between hybrids and parents. According to the fitness-related traits measured here, Chaetodon hybrids are as fit as their parents, and unlikely to promote the formation of a hybrid swarm. However, sufficient fitness and rapid transfer of genetic material between species allow persistence of hybrids within the suture zone, where they positively contribute to genetic diversity. The cases of hybridisation studied here appear to be initiated by similar ecological and behavioural settings, albeit showing different genetic consequences. Determining this was possible through the use of a comprehensive approach, which combined molecular analyses and extensive field observations. Further, the apparent lack of differences in fitness between hybrids and parental species points at the persistence of hybrid individuals within the Indo- Pacific suture zone, where they may continue to contribute positively to genetic diversity. The role of hybridisation in evolution and adaptability had been appreciated in terrestrial and freshwater systems, and this thesis shows that hybridisation can have a role in maintaining reef fish diversity. The studies presented here constitute a comprehensive overview of the relevance of hybridisation for reef fishes and may be a stepping stone toward ascertaining its role in the evolution and adaptation of new species in such a diverse group