Researchers have now identified many highly penetrant genetic risk factors for autism spectrum disorder (ASD). Some of these genes encode synaptic proteins, lending support to the hypothesis that ASD is a disorder of synaptic homeostasis. Less attention, however, has been paid to the genetic risk factors that converge on events that precede synaptogenesis, including the proliferation of neural progenitor cells and the migration of neurons to the appropriate layers of the developing neocortex. Here I review this evidence, focusing on studies of mutant mouse phenotypes, human postmortem data, systems biological analyses, and non-genetic risk factors. These findings highlight embryonic neurogenesis as a potentially important locus of pathology in ASD. In some instances, this pathology may be driven by alterations in chromatin biology and canonical Wnt signaling, which in turn affect fundamental cellular processes such as cell-cycle length and cell migration. This view of ASD suggests the need for a better understanding of the relationship between variation in neuron number, laminar composition, and the neural circuitry most relevant to the disorder.