We report the results of a five year survey of FRAXA and FRAXE mutations among boys aged 5 to 18 with special educational needs (SEN) related to learning disability. We tested their mothers using the X chromosome not transmitted to the son as a control chromosome, and the X chromosome inherited by the son to provide information on stability of transmission. We tested 3738 boys and 2968 mothers and found 20 FRAXA and one FRAXE full mutations among the boys and none among the mothers. This gives an estimated prevalence of full mutations in males of 1 in 5530 for FRAXA and 1 in 23 423 for FRAXE. We found an excess of intermediate and premutation alleles for both FRAXA and FRAXE. For FRAXA this was significant at the 0.001 level but the excess for FRAXE was significant only at the 0.03 level. We conclude that the excess of intermediate and premutation sized alleles for FRAXA may well be a contributing factor to the boys' mental impairment, while that for FRAXE may be a chance finding. We studied approximately 3000 transmissions from mother to son and found five instabilities of FRAXA in the common or intermediate range and three instabilities of FRAXE in the intermediate range. Thus instabilities in trinucleotide repeat size for FRAXA and FRAXE are rare, especially among alleles in the common size range.