The relationship between the 1st-trimester exposure to the antiemetic Bendectin and several major categories of birth defects among human infants was ascertained through the Metropolitan Atlanta [Georgia, USA] Congenital Defects Program from 1968 to 1978. No associations were found between any of these defect categories and Bendectin exposure. An association was found for the amniotic bands cases, a subgroup of limb reductions. Data were stratified to reflect the 1976 change in formulation when 1 of the 3 ingredients was removed. An association was found between esophageal atresia and the 3-ingredient product and another between encephalocele and the 2-ingredient product. These statistical associations do not establish a causal relationship between Bendectin and the birth defects. If the associations are causal, the implied risk is extremely low, .apprx. 1 in 1000 births.