testing communities. An opportunity, because with an effective treatment available, the early detection afforded by a genetic screening program will make it possible to reduce the morbidity and mortality associated with the disease. It is a challenge, as well, because the relationship between genotype and clinical phenotype is unclear at present; e.g., not everyone inheriting two mutant copies of the gene responsible for hereditary hemochromato sis (HH) will develop significant disease, and there are concerns about how best to accomplish the screening, what the screening should entail, and social and ethical issues surrounding the screening. With regard to the challenge posed by hemochromato sis screening, there are similarities to issues previously raised with screening for the breast cancer predisposition genes, BRCA1 and BRCA2. For example, not everyone inheriting a BRCA1/BRCA2 mutation will develop breast or any cancer in her or his lifetime. And concerns have been expressed about when and whom to screen, who gets the results, and what might happen with regard to insurability and employability if the results become known to others besides the subject and his/her doctor. However, there are also significant differences between HFE and BRCA1/BRCA2 testing: cancer predisposition, following inheritance of BRCA1/BRCA2 mutations, is a dominant trait, with cancers seen in successive generations; no one really disputes the appropriateness of testing women with a strong family history of early-onset breast and/or ovarian cancer, only whether one extends the screening beyond this, for example, to all Ashkenazi Jewish women. HH, on the other hand, is a recessively inherited disorder, which means that the family history, except for sibs, is likely to be negative for other affecteds, so one cannot limit any testing just to those with a family history of the disease. The mutations in HFE, the gene responsible for HH, are more frequent than are mutations for BRCA1/BRCA2 among caucasians (estimates of carrier frequencies for mutations in HFE and BRCA1/BRCA2 among caucasians are about 1:10 and 1:200, respectively). However, clinically significant disease is less frequent in HH than in heritable breast cancer, a reflection of reduced penetrance in both disorders: for HH, estimates of clinically significant iron overload disease incidence of between 1 in 500 and 1 in 2,000, in different subpopulations have been reported, whereas between 50% and 85% of BRCA1/BRCA2 mutation carriers will develop breast cancer in their lifetimes. In some respects, HH is like another common recessive disease, cystic fibrosis (CF). In HH, homozygosity for one mutation, C282Y, accounts for over 80% of HH among northern Europeans. In CF, one common mutation, D F508, accounts for 75% or more of CF alleles among carriers and affecteds of northern European origin. In CF, the frequency of D F508 is different in different European subpopulations, accounting for only about 50% of all CF mutations in southern Europeans (Italians, Spaniards, Greeks, etc.) and about 30% of total CF in Ashkenazi Jews. In HH, different frequencies for C282Y, the common mutation, have been reported in specific European subpopulations (northern Europeans vs. Italians) and in other parts of the world (see Merryweather- Clarke et al., this issue). With regard to the major questions surrounding HH population screening: