To test whether the presence of thyroid antibodies in a parent is a risk factor for meiotic nondisjunction, we measured the levels of thyroid antibodies in serum samples drawn during early pregnancy from 101 gravidas who delivered a child with a trisomy, from 11 gravidas who had had a trisomic child in a previous pregnancy, and from 44 of their husbands. For each case mother, three controls were randomly selected from the same population and matched for age, race, sex of the child, and hospital of birth. Cases and controls came from two longitudinal populations, the Child Health and Development Studies (CHDS) and the national Collaborative Perinatal Project (CPP), together comprising more than 70,000 live births. All cases with both a definite diagnosis of trisomy-Down syndrome (DS) or other-and available serum were included. Overall, there was no association between the presence of thyroid antibodies in a mother and a trisomy in her offspring (odds ratio [OR] = .98, confidence interval [CI] = .54-1.85). The lack of association was seen in all three subgroups (DS only, other trisomies, and DS in a previous pregnancy), in all ethnic groups, and in the age groups of white mothers either less than 30 years of age (OR = .80, CI = .40-1.6) or greater than or equal to 30 years of age (OR = 1.26, CI = .82-1.9). In the CHDS population, case fathers, as compared with control fathers, did not have a higher prevalence of thyroid antibodies.(ABSTRACT TRUNCATED AT 250 WORDS)