Understanding the causes of variation in reproductive expenditure in female St. Kilda Soay sheep: a focus on the role of space use and genetic differences
Regan, Charlotte Emily
MetadataShow full item record
Reproductive patterns in the natural world are highly variable both between- and within-species. For example, mammalian females within a single population often vary substantially in their expenditure into offspring, both before and after birth. Understanding the causes of variation in reproductive behaviour remains a key aim within evolutionary ecology, due to its implications for understanding many aspects of biology, from population dynamics to life-history evolution. Nonetheless, little is known about how genetic differences between individuals and variation in individual space use influence investment into reproduction in the wild, particularly in terms of the parental care provided by females during lactation. In this thesis, I use the detailed phenotypic, relatedness, and census data available for the feral population of Soay sheep on the island of Hirta (St. Kilda, Scotland) to understand how differences between individuals driven by variation in genetics and access to resources influence a female's investment before and after birth. First, I present a field behavioural study to understand how a female's home range quality influenced maternal and lamb behaviour over the maternal care period. Despite previous work indicating that females with greater access to high quality vegetation had higher lifetime reproductive success, I found no evidence to suggest that home range quality influenced a mother's investment into care. Second, I used existing data to understand whether an individual's access to high quality vegetation influenced the costs incurred as a result of reproduction. Although I found evidence for significant survival costs, and identified reproductive costs for old animals in poor years, home range quality had no effect on the costs experienced by females. Nevertheless, I did find evidence that females with high quality home ranges were more likely to bear twins and this may explain the positive relationship between home range quality and lifetime reproductive success. Third, I explored methods for accounting for space sharing by related females within quantitative genetic analyses of three early-life (birth weight, birth date, and August weight) and two adult traits (jaw length and metacarpal length). This enabled me to understand the importance of spatial sources of phenotypic similarity in these traits, and to derive estimates of genetic variance components that were not biased by failing to incorporate the space sharing of related females. Females that used space more similarly tended to be more phenotypically similar and had lambs that had more similar trait values. However, contrary to my expectation accounting for home range overlap had little influence on heritability estimates. Finally, I used the models described above to select females that varied in their genetic merit for lamb growth. These females were then included in a second field study to understand whether genetically mediated differences in lamb growth were caused by variation in maternal care behaviours that were, to some degree, genetically determined. Females whose genetic merit was associated with faster lamb growth were less likely to reject sucking attempts and had lambs that spent more time resting, suggesting that maternal care in this population may have a genetic basis. I finish by discussing the implications of these results, drawing attention to limitations and areas worthy of future research.