Graves' disease and Hashimoto's thyroiditis are organ-specific autoimmune disorders of multifactorial aetiology with a polygenic mode of inheritance. Familial clustering and twin studies provide evidence for a genetic predisposition. Three main approaches have been used in the search for susceptibility loci: population-based case-control studies, classical linkage analysis, and intrafamilial linkage disequilibrium. Case-control studies are a sensitive method of gene detection and the collection of subjects is resource-efficient. However, they require prior knowledge of a candidate gene and are prone to inconsistent results due to false positives that may arise from population mismatch. Linkage analysis is a powerful tool for detecting 'major' genes that does not require a candidate gene and is, therefore, a means of genome screening. This method, however, has limited power to detect genes of 'modest' effect, and the collection of sibpairs and multiple family members may be difficult. Intrafamilial linkage disequilibrium analysis is more sensitive than classical linkage analysis, requires only one affected offspring, and eliminates population mismatch. This approach has confirmed linkage disequilibrium of the HLA region with Graves' disease, previously not detected by linkage analysis. Knowledge of a candidate locus is required, however, and this method cannot, therefore, at present be used for genome screening. It is likely that a combination of all three approaches will be required to identify susceptibility loci for autoimmune thyroid disease.