In a technician-oriented routine ultrasound program, the value of screening an entire obstetric population for predicting gestational age based on a single measurement was evaluated over selective scans performed on the basis of uncertain menstrual history. Consecutive pregnancies of 4527 women were scanned, and the results were analyzed. Gestational ages were determined by both menstrual history and ultrasonic crown-rump length or biparietal diameter (BPD) measurements. The estimated date of confinement based on ultrasound measurements was compared with menstrual history in its ability to predict the actual onset of spontaneous labor. Of patients with optimal menstrual history, 84.7% delivered within ± two weeks of the date predicted. Only 69.7% delivered within ± two weeks of the estimate date of confinement based on suspect menstrual history. Crown-rump length measurements were as predictive (84.6%) as optimal menstrual history. Biparietal diameter measurements done between 12 and 18 weeks' gestation were significantly more accurate in gestational predictions (89.4%) than those based on menstrual history (P< .001). It is concluded that ultrasound cephalometry before 18 weeks is the single best dating parameter.