Assessment of cervical status is an important component of the management of patients at risk for preterm delivery. Although digital examination is the most common method of cervical assessment, there has been recent interest in sonographic cervical examination. To compare the accuracy of digital examination and ultrasound, 20 nongravid women undergoing total hysterectomy for gynecologic indications had measurements of cervical length performed digitally and with both transabdominal and transvaginal ultrasound before surgery. These measurements were then compared with measurements made with a ruler immediately after hysterectomy. Separate examiners performed the digital, ultrasound, and ruler measurements, and each was blinded to the results of the others. Digital examination underestimated cervical length by an average of 13.6 mm and was significantly shorter than ruler measurement (P = .0001). Neither ultrasound method differed significantly from ruler measurement (P greater than .9 for each), and measurements were similar between the sonographic techniques (P greater than .9). These results validate the accuracy of sonographic estimation of cervical length. In addition, they suggest that sonographic measurement is more accurate than digital examination in predicting true cervical length. Finally, in the nonpregnant women, neither ultrasound technique seems superior to the other.