The objective of htis study is to compare the accuracy of early ultrasonograhic biometry versus reliable menstual dates in predicting the date of spontaneous confinements in pregnant women at the Obafemi Awolowo University Teaching hospital Complex. Ile-Ife .Nigeria. This prospective study was done over a period of twenty four months. The study population included singleton pregnant women recruited at a gestational age of 22 weeks or less who have an unequivocal,reliable last menstrual period with hitherto regular cycles, and corresponding uterine size. They were subjected to ultrasound scanning for pregnancy dating after due counselling,and obtaining informed consent. Every effort was made to keep the sinologist ‘blind’ to their menstrual date as applicable. The estimated gestational age and expected date of delivery were ascertained by fetal biometric variables using the linear regression models. The patient were subsequently followed up in the antenatal clinic according to departmental protocol. At delivery in the hospital,the actual date of spontaneous delivery was noted. A neonatologist assessed the true neonatal age using the Dubowitz score. All patients who were delivered for obstetric emergencies, elective vaginal or abdominal deliveries or complicated pregnancies were excluded