Twenty physicians experienced in fetal sonographic evaluation obtained fetal nuchal skinfold thickness measurements in each of five pregnant women between 16 and 18 weeks' gestation. A random effects analysis of variance model was used to explore the sources of variation in the set of measurements, to determine the interobserver variability of sonographically measured second-trimester nuchal skinfold thickness. It was possible to obtain 96 measurements. Four measurements (all in the same patient) were deemed unobtainable due to fetal position. All recorded measurements were between 1.7 and 4.5 mm. The means (and ranges) for the five patients were 2.7 (1.7-3.6), 2.9 (2.0-4.5), 2.7 (2.0-4.0), 3.3 (2.2-4.0), and 2.7 (1.8-4.0) mm. The standard deviation for interobserver variability, caused by the combined effect of physician and machine imprecision, was 0.56 mm and the overall coefficient of variation was 19.8%. Interpatient differences were statistically significant (p = 0.004). Interphysician differences were not (p = 0.11). We conclude that experienced physician sonographers using high-resolution ultrasound equipment are able to obtain second-trimester nuchal skinfold thickness measurements within 1.1 mm of the estimated true value with 95% probability.