In summary, autism genetics has moved from a time of identification of heritability and determination of risk of "lesser variants" or the "broader phenotype" in relatives to a phase where some cases of autism have a definite basis such as maternally inherited duplications of 15q11-q13, identification of mutations causing AS, Rett syndrome, and FRAXA. The first phase of genome-wide screens has not revealed definitive linkage, but as samples are enlarged and meta-analyses performed, the strongest linkage findings are likely to yield susceptibility variants once fine mapping proceeds. Recent statistical and molecular genetic analysis methods make the additional work feasible. However, frustrating it may be to be in this phase of the research, it is an essential part of the process of moving from identification of heritability in autism to understanding of the disorder in a way that may permit improved treatment in the future. If there is an advantage to autism being a complex rather than monogenic disorder, it is that the nature of multiplicative or interacting genetic risk is that prevention or treatment directed to any of the identified genetic risks may be sufficient to break a chain of pathophysiology. More genes increase the chance that one or more will have implications for treatment development sooner.