<p>Identifying sources of variation in individual reproductive success is crucial to our understanding of population dynamics and evolutionary ecology. In many systems, the determinants of success are not well known. Where species have parental care, for example, determinants of success can be particularly challenging to partition between parents and offspring. In this thesis I investigate drivers and consequences of variable life histories, for a small reef fish that exhibits male parental care (the common triplefin Forsterygion lapillum). I examined the influence of individual life history, phenotype and behaviour on (1) the performance of recently settled juveniles, and (2) the reproductive success adult males. I made field-based observations of adult males during the breeding season, measured their phenotypic traits (body size and condition) and used their otoliths to reconstruct life history characteristics (hatch dates and mean growth rates). My life history trait reconstructions suggested two alternate pathways to ’success’ for adult males. Successful males hatched earlier and therefore had a developmental ’head start’ over less successful males (i.e., males with eggs > male territory holders without eggs > floaters). Alternatively, males can apparently achieve success by growing faster: for males born in the same month, those with eggs grew faster than those with territories and no eggs, and both groups grew faster than floaters. These results suggest that accelerated growth rate may mediate the effects of a later hatch date, and that both hatch dates and growth rates influence the success of adult males, likely through proximate effects on individual phenotypes. Identifying sources of variation in individual reproductive success is crucial to our understanding of population dynamics and evolutionary ecology. In many systems, the determinants of success are not well known. Where species have parental care, for example, determinants of success can be particularly challenging to partition between parents and offspring. Male parental care is common among fishes, where resources such as high quality territories and mates often may be limiting. In such systems, individual success of offspring may result from distinct life history pathways that are influenced by both parental effects (e.g., timing of reproduction) and by the offspring themselves (e.g., ’personalities’). These pathways, in turn, can induce phenotypic variation and affect success later in life. The drivers and consequences of variable life histories are not well understood in the context of reproductive success. In this thesis I investigate drivers and consequences of variable life histories, for a small reef fish that exhibits male parental care (the common triplefin Forsterygion lapillum). I examined the influence of individual life history, phenotype and behaviour on (1) the performance of recently settled juveniles, and (2) the reproductive success adult males. I made field-based observations of adult males during the breeding season, measured their phenotypic traits (body size and condition) and used their otoliths to reconstruct life history characteristics (hatch dates and mean growth rates). Some males showed no evidence of territorial defence and were defined as ’floaters’; others defended territories, and a subset of these also had nests with eggs present. Adult male body size was significantly higher for males that defended breeding territories, and body condition was significantly higher for the males that had eggs (i.e., had successfully courted females). My otolith-based reconstructions of life history traits suggested two alternate pathways to ’success’ for adult males. Successful males hatched earlier and therefore had a developmental ’head start’ over less successful males (i.e., males with eggs > male territory holders without eggs > floaters). Alternatively, males can apparently achieve success by growing faster: for males born in the same month, those with eggs grew faster than those with territories and no eggs, and both groups grew faster than floaters. These results suggest that accelerated growth rate may mediate the effects of a later hatch date, and that both hatch dates and growth rates influence the success of adult males, likely through proximate effects on individual phenotypes. I evaluated the effects of variable life history in a complimentary lab-based study. Specifically, I manipulated the developmental environments (feeding regime and temperature) for young fish and evaluated the direct effects on life history traits and phenotypes. Then, I conducted an assay to quantify the indirect effects of developmental environment, life history traits, and phenotypes on aggression and performance of young fish. These developmental environments did not have a clear, overall effect on juvenile phenotype or performance (i.e. behavioural aggression and the ability to dominate a resource). Instead, individuals (irrespective of developmental environment) that grew faster and/or longer pelagic larval durations had increased odds of dominating a limited resource. I attributed the non-significant direct effect of developmental environment to within-treatment mortality and variation among individuals in terms of their realised access to food (i.e., dominance hierarchies were apparent in rearing chambers, suggesting a non-uniform access to food). Fish that were more likely to dominate a resource were also more aggressive (i.e., more likely to engage in chasing behaviours). Fish that were larger and more aggressive established territories that were deemed to be of higher ’quality’ (inferred from percent cover of cobble resources). Overall, this study suggests a complex interplay between social systems, phenotype and life history. Developmental environments may influence phenotypes, although behavioural differences among individuals may moderate that effect, contributing to additional variation in phenotypes and life history traits which, in turn, shape the success of individuals. Collectively, my thesis emphasises the consequences of life history variability on success at multiple life stages. These results may be relevant to other species that exhibit male parental care or undergo intense competition for space during early life stages. In addition, my results highlight interactions between life history, phenotype and behaviour that can have important implications for population dynamics and evolutionary ecology.</p>
The effect of pollen source on seed traits and dispersability in the heterocarpic annual Crepis sancta
