Immune activation is generally acknowledged to be costly. These costs are frequently assumed to result from trade-offs arising due to the reallocation of resources from other life-history traits to be invested in immunity. Here, I investigated the energetic basis of the costs associated with immune activation in Drosophila melanogaster. I found that immune activation significantly reduced fly fecundity (45%) and also caused a decline in metabolic rate (6%) but had no effect on body weight. To understand the factors behind reduced fecundity and metabolic rate I measured feeding and found that food intake was reduced by almost 31% in immune-challenged D. melanogaster. These findings suggest that fecundity costs of immune activation result not from the commonly accepted resource reallocation hypothesis but probably because resource acquisition is impaired during immune responses. The individuals of any animal population generally vary greatly in their ability to resist infectious disease. This variation arises due to both environmental heterogeneity and genetic diversity. Genetic variation in disease susceptibility has generally been considered to lie in the nuclear genome. Here, for the first time, I explored the influence of mitochondrial genetic (mtDNA) variation on disease susceptibility. I crossed 22 mitochondrial haplotypes onto a single nuclear genome and also studied epistasis interactions between mitochondrial and nuclear genomes (mitonuclear epistasis) by crossing five haplotypes onto five different genetic backgrounds. I found that fly susceptibility to Serratia marcescens was influenced significantly by mtDNA allelic variation. Furthermore, the effect of mitonuclear epistasis on fly susceptibility to S. marcescens was twice as great as the individual effects of either mitochondrial or nuclear genome. However, susceptibility to Beauveria bassiana was not affected by mtDNA allelic variation. These findings suggest the mitochondrial genome may play an important role in host-parasite coevolution. The Mother’s Curse hypothesis suggests that sex-specific selection due to maternal mitochondrial inheritance means that mitochondria are poorly adapted to function in males, resulting in impaired male fitness. Mother’s Curse effects have previously only been studied for two phenotypic traits (sperm-infertility and ageing) and their generality for broader life-history has not been explored. I investigated the impact of mtDNA allelic variation on 10 phenotypic traits and tested whether the patterns of phenotypic variation in males and females conformed to the expectations of the Mother’s Curse hypothesis. I found that seven of the 10 traits were significantly influenced by mtDNA allelic variation. However, there was no evidence that the effects of this variation differed between males and females. I therefore concluded that Mother’s Curse is unlikely to be a general phenomenon, nor to provide a general explanation for sexual dimorphism in life-history traits. Overall, this thesis explored the impacts of immunity costs, mitochondrial genetic variation, mitonuclear epistasis and sex-specific mitochondrial selection on D. melanogaster life-history.