Epidemiological studies have implicated an interplay between genetic and environmental factors in the aetiology of multiple sclerosis (MS). There is a familial recurrence rate of approximately 15%. Meta-analysis of the recurrence risk shows that the rate is highest overall for siblings, then parents and children, with lower rates in second- and third-degree relatives. Recurrence is highest for monozygotic twins. Conversely, the frequency in adoptees is similar to the population lifetime risk. The age-adjusted risk for half siblings is also less than for full siblings. Recurrence is higher in the children of conjugal pairs with MS than the offspring of single affected. These classical genetic observations suggest that MS is a complex trait in which susceptibility is determined by several genes acting independently or epistatically. Comparisons between co-affected sibling pairs provide no evidence for correlation with age or year at onset and mode of presentation or disability. Thus far, the identification of susceptibility genes has proved elusive but genetic strategies are now in place which should illuminate the problem. The main dividend will be an improved understanding of the pathogenesis. To date, population studies have demonstrated an association between the class II major histocompatibility complex (MHC) alleles DR15 and DQ6 and their corresponding genotypes. An association with DR4, with or without the primary DR15 link, is seen in some Mediterranean populations. Candidate gene approaches have otherwise proved unrewarding. Four groups of investigators have undertaken a systematic search of the genome. In common with most other complex traits, no major susceptibility gene has been identified but regions of interest have been provisionally identified. These genetic analyses are predicated on the assumption that MS is one disease. Genotypic and phenotypic analyses are beginning to question this assumption. A major part of future studies in the genetics of MS will be to resolve the question of disease heterogeneity.