The relation between sonographically measured fetal femur length and abdominal circumference (expressed as femur length/abdominal circumference x 100) was studied in 361 normal pregnancies between 15 and 42 weeks. This ratio was found to be independent of menstrual age, with a normal range after 21 weeks (mean ± 2 SD) of 22 ± 2. This ratio also was evaluated as a predictor of intrauterine growth retardation in 30 cases using the 90th percentile (23.5) as the upper limit of normal, which resulted in identification of 63% of the growth-retarded fetuses. Because it is independent of menstrual age, this ratio should prove most useful in evaluating high-risk patients who present in the third trimester of pregnancy with no dates. It is also of value in detecting growth retardation in patients with good menstrual history, since it was shown to become abnormal in asymmetrically growth-retarded fetuses before the estimated weight fell below the 10th percentile. Fetal intrauterine growth retardation, which is typically defined as a weight below the 10th percentile for age, is associated with increased perinatal morbidity and mortality. Proper obstetric management depends on recognition of this problem in utero, but its recognition is difficult prenatally on clinical findings alone. Efforts have been made to detect fetal growth retardation using a number of sonographic measurements such as biparietal diameter, serial biparietal di ameters, head to body circumference ratios, head to body area ratios, crownrump length times trunk area, total intrauterine volume, and fetal femur length [1, 2]. Proper use of all these measurements, however, requires precise knowledge of the true menstrual age, which often is not known in the high-risk indigent patient. It would be very helpful to have a screening tool for fetal growth retardation, which is age-independent.In our sonographic studies of fetal age using multiple fetal growth factors (biparietal diameter, head circumference, abdominal circumference, femur length) [3], we have noticed that after 21 weeks in a normal pregnancy the femur length and abdominal circumference seem to have a relatively constant relation to one another regardless of menstrual age. We also have observed growthretarded fetuses in which the fetal abdominal circumference was disproportionally reduced compared with the fetal femur length, a finding that is analogous to the postnatal observation in growth-retarded babies of disproportionate de creases in fetal weight compared with fetal crown-heel length (i.e., an abnormal ponderal index) [4]. In view of these observations, our study was undertaken with the following objectives: (1) to quantitate normal data for the femur length/ abdominal circumference ratio throughout gestation and to determine if this relation changes with time, and (2) to determine the usefulness of the femur length/abdominal circumference ratio in detecting intrauterine growth retarda tion.