BACKGROUND: The homologous 4q and 10q subtelomeric regions include two distinctive polymorphic arrays of 3.3 kb repeats, named D4Z4. An additional BlnI restriction site on the 10q-type sequence allows to distinguish the chromosomal origin of the repeats. Reduction in the number of D4Z4 repeats below a threshold of 10 at the 4q locus is tightly linked to Facioscapulohumeral Muscular Dystrophy (FSHD), while similar contractions at 10q locus, are not pathogenic. Sequence variations due to the presence of BlnI-sensitive repeats (10q-type) on chromosome 4 or viceversa of BlnI-resistant repeats (4q-type) on chromosome 10 are observed in both alleles. RESULTS: We analysed DNA samples from 116 healthy subiects and 114 FSHD patients and determined the size distributions of polymorphic 4q and 10q alleles, the frequency and the D4Z4 repeat assortment of variant alleles, and finally the telomeric sequences both in standard and variant alleles. We observed the same frequency and types of variant alleles in FSHD patients and controls, but we found marked differences between the repeat arrays of the 4q and 10q chromosomes. In particular we detected 10q alleles completely replaced by the 4q subtelomeric region, consisting in the whole set of 4q-type repeats and the distal telomeric markers. However the reciprocal event, 10q-type subtelomeric region on chromosome 4, was never observed. At 4q locus we always identified hybrid alleles containing a mixture of 4q and 10q-type repeats. CONCLUSIONS: The different size distribution and different structure of 10q variant alleles as compared with 4q suggests that these loci evolved in a different manner, since the 4q locus is linked to FSHD, while no inheritable disease is associated with mutations in 10qter genomic region. Hybrid alleles on chromosome 4 always retain a minimum number of 4q type repeats, as they are probably essential for maintaining the structural and functional properties of this subtelomeric region. In addition we found: i) several instances of variant alleles that could be misinterpreted and interfere with a correct diagnosis of FSHD; ii) the presence of borderline alleles in the range of 30-40 kb that carried a qA type telomere and were not associated with the disease.