The molecular diagnosis of a genetic disease can be made by demonstrating linkage of suitable markers to the disease allele. However, it is generally agreed that it is better to define the mutation responsible. There is a plethora of techniques available for the detection of mutations within genes. No one method is predominant, although single-stranded conformational polymorphism (SSCP) may be the single most widely used technique. The choice in any given situation is a complex function of a particular method's efficiency (i.e., the proportion of all mutations in a given set that are detected), reproducibrhty, ease, speed, and cost, coupled with local considerations, such as the equipment, expertise, and budget available. Factors specific to the disease and gene(s) concerned also play an important part: genes can vary greatly in size; mutations may be clustered in regions or occur in "hot spots"; some specific mutations may occur at a high frequency in a particular population; mutations may be mostly either mis-sense or non-sense; some diseases may have a high new mutation rate. In addition, the nature of the material available for diagnosis (e.g., stored DNA or lymphoblastoid cell line, formalin-fixed, or frozen tissue) may be a deciding factor. The priorities of providing a clinical service may be somewhat different from those of a research laboratory, but any laboratory's decision on the methods it employs will be governed by the costtobenefit ratio. All these points are illustrated when considering the molecular diagnosis of familial adenomatous polyposis (FAP). For service laboratories, however, there is the special consideration regarding whether the expense of mutation detection can be justified by the clinical benefit, but a discussion on the assessment of this is beyond the scope of this chapter. Climcally, FAP is characterized by the development of multiple gastromtestmal (GI) adenomas, usually hundreds to thousands, one or more of which if left untreated inevitably progress to carcinomas.