Stroke represents an enormous health problem worldwide. It describes a clinical syndrome which can be caused by a number of different pathologies, rather than a single disease. Over 80% of strokes are ischaemic, as opposed to haemorrhagic. This review covers advances in the genetics of both monogenic and multifactorial ischaemic stroke. Like many other complex diseases, progress in identifying genes for multifactorial stroke has been disappointing. However, genome-wide association study (GWAS) technology is starting to have a major impact on our understanding of the genetics of stroke. Early studies have shown that genetic associations identified with other diseases known to be associated with stroke, such as coronary heart disease and atrial fibrillation, are themselves genetic risk factors for stroke. A number of stroke GWASs are nearing completion; these have identified novel associations with ischaemic stroke. Most associations reported to date are with specific stroke subtypes. This parallels findings from monogenic causes of stroke where individual mutations usually predispose to specific stroke subtypes. This has implications for the understanding of the pathogenesis of stroke, and emphasizes the importance of careful stroke subtyping in genetic epidemiology studies. So far, studies have looked for genetic risk factors for stroke acting independently of environmental factors. However, we know that conventional environmental risk factors are important in stroke pathogenesis, and considerable evidence suggests that gene-environment interactions will be important. Identifying these is likely to require much larger sample sizes.