Evolutionary quantitative genetics is the study of how complex traits evolve over time. While this field builds on traditional concepts from quantitative genetics widely used by applied breeders and human geneticists (in particular, the inheritance of complex traits), its unique feature is in examining the role of natural selection in changing the population distribution of a complex trait over time. Our review focuses on this role of selection, starting with response under the standard infinitesimal model, in which trait variation is determined by a very large number of loci, each of small effect. We then turn to issues of measuring fitness (and hence natural selection) on both univariate and multivariate traits. We conclude by examining models that treat fitness itself as a complex trait.