Affective (mood) disorders are common. There are several methodological impediments to genetic studies of affective disorders, including uncertainties about the best definition of disease phenotype, difficulties in the assessment of lifetime diagnosis and variable age of onset of illness. Despite these difficulties, family, twin and adoption studies provide compelling evidence for the existence of important genetic factors in determining susceptibility to affective disorders. However, the mode of inheritance is unknown. Simple mendelian inheritance may occur in some families but cannot explain the majority of cases. With the advent of polymorphic DNA markers, linkage and association studies have become more useful methods for the genetic analysis of complex disorders such as affective illness. No consistent finding has yet emerged, although chromosomal region 11p15 (and to a lesser extent Xq28) are of continuing interest. In addition to further study of these regions it will also be necessary to look for susceptibility loci in other parts of the genome. Large samples will almost certainly be required. If susceptibility loci of major effect exist then linkage approaches will find them. However, if there are only loci of small effect, then association approaches will be necessary. At present, it seems prudent to pursue both linkage and association approaches together.