To the Editor.—
I do not expect to be the only person who has written to you about an elementary flaw in the article by Berrettini et al.1 At least three of the pedigrees, arguably four, show more than one source of potential disease genes. In a situation of genetic heterogeneity, as exists here, linkage analysis is uninterpretable if there is more than one source of disease gene in the pedigrees. Therefore, pedigrees 0024, 0441, and 0488 should have been excluded from the analysis. Moreover, if the authors' explanation that the disorder in family 0048 is transmitted to the youngest generation by the marrying-in female (414), thereby ruling out what looks like male-to-male transmission, then this pedigree should have been excluded on the same grounds. However, if there is male-to-male transmission in family 0048, the family has no place in a study of X-linkage. Hence, it should have been excluded either way.