The evolution of sexual dimorphism is constrained by a shared genome between males and females. This constraint can lead to 'sexual antagonism' where segregating alleles at given genetic loci have opposing fitness effects in each sex. Despite its wide taxonomic incidence, little is known about the identity, genomic location and evolutionary dynamics of sexually antagonistic polymorphisms. This is a major knowledge gap, since a better understanding of antagonistic polymorphisms can shed light on two fundamental questions: (i) how does the genome evolve to accommodate divergent and often contradictory selective pressures, and (ii) what evolutionary forces maintain genetic variation for fitness? In this thesis, I describe the genetics and evolutionary dynamics of sexually antagonistic polymorphisms. I first highlight the limitations of previous genetic studies of sexual antagonism (Chapter 2). Specifically, I re-analyse a prominent study of antagonistic gene expression and show that inferences of antagonistic selection were driven by non-random population structure in the sample of genomes considered, rendering previous conclusions unreliable. I then present the first genome-wide association study of sex-specific fitness and sexual antagonism in a laboratory-adapted population of D. melanogaster (Chapter 3). I show that antagonistic variation disproportionately accumulates in coding regions but not on the X chromosome. I proceed to test whether sexually antagonistic selection maintains population genetic variation (Chapter 4), as has long been proposed but never tested. Consistent with this hypothesis, I find multiple signatures of balancing selection associated with antagonistic loci across populations of D. melanogaster separated over 10,000 years, and possibly across species boundaries. Finally, I present experimental work testing whether a specific candidate gene-fruitless-is under antagonistic selection (Chapter 5). The results presented are consistent with balancing but not antagonistic selection. Overall, this thesis underscores the fundamental difficulty of evolving genetic mechanisms that accommodate the divergent evolutionary interests of each sex.