The value of the Rossavik growth model [P=c(t)k + s(t)] was evaluated in 39 patients with singleton pregnancy who had neonatal weight outcome above the 90th percentile of our birth weight distribution for gestational age. Individual fetal growth curve standards for head and abdominal circumferences, femur diaphysis length, and weight were determined from the data of two scans obtained before 26.1 weeks' gestation and separated by an interval of at least 5 weeks. Projected crown-heel lengths were calculated from projected femur diaphysis length values. Comparisons between actual and predicted birth characteristics were expressed by the Growth Potential Realization Index (GPRI) and Neonatal Growth Assessment Score (NGAS). Excessive growth at birth was seen in almost all cases as indicated by high GPRI for weight and abdominal circumference and abnormal NGAS values. In eight of the 33 patients who delivered after 38 weeks, excessive growth was detected only by comparing birth characteristics to their predicted values at 38 weeks' gestation. Our data suggest that individual growth curve standards may identify several patterns of excessive fetal growth that could represent different pathophysiologic mechanisms, ie, failure to terminate growth after 38 weeks versus a defect in a still unknown growth regulator. The individual fetal growth curve standards method gives additional information and discriminates well between normal and excessive fetal growth.