Fetal alcohol effects in 359 infants born to disadvantaged women identified as having a history of alcohol abuse or as controls and who provided reports of alcohol use in pregnancy are being studied in a prospective design. Alcohol abuse was assessed with the Michigan Alcoholism Screening Test (MAST). Alcohol use (AA/day) was based on short-term recall covering 2-week periods prior to each antenatal visit. A tally of anomalies associated with fetal alcohol syndrome was obtained in a blinded examination of each infant. This tally was significantly related to the MAST classification and, for the MAST+ subjects, the tally was related to first trimester AA/day. Birth weight, length, and head circumference were negatively correlated with AA/day (entire pregnancy); however, the effect was attenuated and not statistically significant in models with covariate control. It is possible that these measures were near the threshold of effect. Scale scores of the Brazelton Neonatal Behavioral Assessment Scale and three scale scores of the Graham/Rosenblith Behavioral Examination of the Neonate were unrelated to the MAST classification and to AA/day.