A sample of 149 twin pairs was ascertained by extracting all pairs of candidates with the same surname, birth date and address from the files of a public examination taken by about two thirds of the 15-year-old cohort in South Australia. Of these there were 39 opposite sex pairs and 100 same-sex pairs whose zygosity was determined by postal questionnaire. This diagnosis was checked and found to be accurate by typing 47 pairs for ten independent genetic markers. Several analyses suggest that the ascertained twin samples is fairly representative of the fraction of the cohort who took the examination so there is no evidence to suggest that between families components of variance will be underestimated for this reason.