Multiple sclerosis (MS) is a common and severe CNS disorder that is characterized by myelin loss, chronic inflammation, axonal and oligodendrocyte pathology, and progressive neurological dysfunction. Extensive epidemiological data confirm that genetic variation is an important determinant of susceptibility to MS, and suggest that such variation also influences the timing of symptom onset, the course of the disease, and the treatment response. Multicenter international collaborations have allowed large and well-characterized sample collections to be assembled that, when coupled with high-powered laboratory technologies, afford the opportunity to analyze the genome with increasing resolution and detail. The seven MS genome-wide association screens that have been completed in the past 3 years have substantially lengthened the list of MS genetic risk associations. Nevertheless, our knowledge of MS genetics remains incomplete, with many risk alleles still to be revealed, although progress is likely to be rapid in the near future. The ensuing challenge will be to design effective functional studies that convincingly link genetic variation to the underlying pathophysiology of MS. Establishment of such connections might translate into clinically useful genetic biomarkers and reveal novel targets for therapy. This Review briefly summarizes well-established concepts of MS epidemiology and susceptibility, and discusses new knowledge emerging from genome-wide association studies.