Accidental recombination between the differential segments of the X and Y chromosomes in man occasionally allows transfer of Y-linked sequences to the X chromosome leading to testis differentiation in so-called XX males. Loss of the same sequences by X-Y interchange allows female differentiation in a small proportion of individuals with XY gonadal dysgenesis. A candidate gene responsible for primary sex determination has recently been cloned from within this part of the Y chromosome by Page and his colleagues. The observation that a homologue of this gene is present on the short arm of the X chromosome and is subject to X-inactivation, raises the intriguing possibility that sex determination in man is a quantitative trait. Males have two active doses of the gonad determining gene, and females have one dose. This hypothesis has been tested in a series of XX males, XY females and XX true hermaphrodites by using a genomic probe, CMPXY1, obtained by probing a Y-specific DNA library with synthetic oligonucleotides based on the predicted amino-acid sequence of the sex-determining protein. The findings in most cases are consistent with the hypothesis of homologous gonad-determining genes, GDX and GDY, carried by the X and Y chromosomes respectively. It is postulated that in sporadic or familial XX true hermaphrodites one of the GDX loci escapes X-inactivation because of mutation or chromosomal rearrangement, resulting in mosaicism for testis and ovary-determining cell lines in somatic cells. Y-negative XX males belong to the same clinical spectrum as XX true hermaphrodites, and gonadal dysgenesis in some XY females may be due to sporadic or familial mutations of GDX.