We read with interest the recent report by Castlenovoet al of the first reported case of Friedreich's ataxia presenting with a pure spastic paraparesis.1 Since the identification of the frataxin gene in 1996 the phenotypic range of Friedreich's ataxia has been greatly expanded. After this report we therefore analysed the GAA repeat length in the first intron of the frataxin gene by polymerase chain reaction, using techniques previously described,2 in affected members from eight families with a spastic paraparesis and evidence of autosomal recessive inheritance. In each case the presenting feature was of a slowly progressive spastic paraparesis. At least one affected member of each family had undergone a full series of investigations based on those proposed by the Hereditary Spastic Paraplegia Working Group to exclude other causes of a spastic paraparesis.3 The age of onset ranged from 5 to 50 years. Additional neurological features such as peripheral neuropathy, mild ataxia, and intellectual impairment developed later in the course of the disease in affected members from four of the families. The GAA repeat lengths in all patients tested fell within the normal range. We therefore conclude that the presentation of Friedreich's ataxia as an autosomal recessive spastic paraparesis is likely to be rare.