We evaluated the predictive accuracy of short forms of the Wechsler intelligence scales for individuals with high functioning autism. Several short forms were derived from participants who had received the full procedure. Stepwise multiple regression analyses were performed to determine the strength of association between the subtests included in the short form and IQ scores based upon the full test. These analyses were performed for all participants, and also for autism participants with atypical subtest profiles. In all analyses the percentages of explained variance were typically in the 0.8-0.9 range. It was concluded that short forms may be used with good predictive accuracy in individuals with high functioning autism, even when the subtest profile is atypical.