While sexual size dimorphism (SSD) is abundant in nature, there is huge variation in both the intensity and direction of SSD. SSD results from a combination of sexual selection for large male size, fecundity selection for large female size and ecological selection for either. In most vertebrates, it is variation in the intensity of male–male competition that primarily underlies variation in SSD. In this study, we test four hypotheses regarding the adaptive value of SSD in sharks—considering the potential for each of fecundity, sexual, ecological selection and reproductive mode as the primary driver of variation in SSD between species. We also estimate past macroevolutionary shifts in SSD direction/intensity through shark phylogeny. We were unable to find evidence of significant SSD in early sharks and hypothesise that SSD is a derived state in this clade, that has evolved independently of SSD observed in other vertebrates. Moreover, there is no significant relationship between SSD and fecundity, testes mass or oceanic depth in sharks. However, there is evidence to support previous speculation that reproductive mode is an important determinant of interspecific variation in SSD in sharks. This is significant as in most vertebrates sexual selection is thought to be the primary driver of SSD trends, with evidence for the role of fecundity selection in other clades being inconsistent at best. While the phylogenetic distribution of SSD among sharks is superficially similar to that observed in other vertebrate clades, the relative importance of selective pressures underlying its evolution appears to differ.