Stroke can be viewed as a paradigm for late-onset, complex polygenic diseases. There are two main clinical phenotypes for stroke: ischemic stroke, responsible for 80-90% of stroke events, and hemorrhagic stroke, responsible for the remaining 10-20%. Stroke may either be the outcome of a number of monogenic disorders or, more commonly, a polygenic multifactorial disease. Cerebral autosomal dominant arteriopathy with subcortical infarcts and leucoencephalopathy (CADASIL), due to mutations in the Notch 3 gene, is the best example of monogenic pathology leading to stroke. The identification of individual causative mutations for polygenic stroke is problematic due to the complexity of such condition. The two main methods of genetic investigation are linkage analyses and association studies, each with advantages and limitations. Associations with polymorphisms in a variety of candidate genes have been investigated, including hemostatic genes, genes controlling homocystein metabolism, the angiotensin-converting enzyme gene, and the endothelial nitric oxide synthase gene. The combination of linkage and association approaches has led to the identification of the first putative gene associated with common polygenic stroke, PDE4D, mapped to chromosome 5q21. The biological revolution of the past years, spurred by the Human Genome Project, promises the advent of novel technologies supported by bioinformatics, which will transform the study of polygenic disorders such as stroke. Understanding the causes of stroke and its effect will allow definition of high-risk populations and make possible specific programs of primary and secondary prevention as well as new therapeutic approaches where prevention has failed.