The frequencies of various genetically defined spinocerebellar ataxias (SCAs) vary in different populations presumably due to founder effects. No data have been published on the Australian population. Although predominantly of Anglo-Celtic extraction, Australia has also received considerable influx from southeastern Europe and more recently eastern and southeastern Asia. We examined the frequency of mutations for SCA types 1, 2, 3, 6, and 7 in southeastern Australia. Of 88 pedigrees with multiple-affected members, SCA type 1 (SCA1) accounted for 16%, SCA2 for 6%, SCA3 for 12%, SCA6 for 17%, SCA7 for 2%, and 47% (41 pedigrees) were negative for each of SCA1, 2, 3, and 6. Twenty of the 41 negative pedigrees were also negative for dentatorubralpallidoluysian atrophy, and indeed dentatorubralpallidoluysian atrophy has not been reported in Australia. In addition, no pedigree information was available on a further four patients with SCA1, three patients with SCA2, three patients with SCA3, and three patients with SCA6. One SCA1 and two SCA2 patients had no other known affected family members. In total, of 63 pedigrees or individuals with positive tests, 30% were those with SCA1, 15% with SCA2, 22% with SCA3, 30% with SCA6, and 3% with SCA7. Judging by pedigree names, four of the nine SCA2 positive individuals/pedigrees were of Italian extraction, and four of the 14 SCA3 positive individuals/pedigrees were of Chinese descent, whereas only 1 of the 20 SCA1 positive individuals/pedigrees were non-Anglo-Celtic. These results are in accordance with the known ethnic composition of the Australian population and with gene frequencies in these constituent ethnic groups reported by others. The frequency of large-normal alleles for SCA1 and SCA3 in the population reflects the prevalence of these two diseases, supporting the hypothesis that disease alleles arise by expansion of large-normal alleles.