An investigation was made of the behavioral effects of caffeine in a sample of 313 newborns and their mothers. A weighted measure of caffeine based on daily ingestion of coffee, tea, and cola was derived from a maternal interview. The majority of mothers consumed the equivalent of about 1.3 cups of coffee per day. Infant outcome measures included birth size, gestational age, and behavior as measured by the Brazelton Neonatal Behavioral Assessment Scale. To control for potential confounding variables possibly responsible for any observed caffeine effects, a list of 30 potential confounds was compiled. All control variables that correlated with a given substance at an alpha level less than or equal to .10 were considered potential confounds and included in regression analyses. Results showed that maternal caffeine consumption prior to pregnancy predicted lower birthweight, poorer orientation, and greater arousal and irritability. Each of these outcomes, except orientation, was related to exposure in a dose-dependent fashion. Related research found that infants whose mothers consumed higher levels of caffeine while breastfeeding had higher motor development scores than infants of mothers consuming lower levels of caffeine. Additionally, a study of the reliability and validity of maternal report of caffeine use indicated that, while women had the most difficulty recalling their consumption of caffeinated soft drinks during pregnancy, their reports were quite reliable regarding coffee and tea. (RH)