The Australian Paediatric Surveillance Unit and the Rett Syndrome Association of Australia (a parent support group) were the main sources of ascertainment of cases of Rett syndrome (RS) aged 18 years or less in Australia. Further information was obtained from questionnaires to paediatricians and families. Cases were categorised as classical, atypical, potential or non-verifiable, using internationally recognised criteria. Capture recapture methods were used to adjust prevalence measures for missing cases, and Kaplan Meier survival analysis to estimate cumulative incidence. A total of 134 cases of RS was identified, and of those over 5 years of age, 68% were classical and 32% were atypical. The adjusted prevalence was 0.72 per 10,000 females, and the cumulative incidence was 0.96 per 10,000 females to age 12 years. The prevalence was similar to that in Sweden and Scotland, but higher than in Texas. Comparing successive birth cohorts, there was evidence to suggest that RS is being diagnosed earlier and that more cases are being diagnosed as atypical.