Birth size and shape are commonly used as indicators of fetal growth. Epidemiologic studies have suggested a relationship between birth size and the risk of developing cardiovascular disease in later life. Certain "growth phenotypes" have been linked to the development of certain components of cardiovascular disease, particularly babies who display disproportional growth in utero. These observations are based on retrospective analysis of historical data sets. If the "Fetal Origins of Adult Disease" hypothesis is to be generalisable to the present day, then it is essential to establish whether these "growth phenotypes" exist within the normal distribution of birth size. The UCL Fetal Growth Study is a prospective study of antenatal fetal growth assessed by ultrasound at 20 and 30 wk gestation in 1650 low risk, singleton, white pregnancies. Measures of birth size were obtained and analyzed by principal components to explain shape at birth. Birth measures were also related to antenatal growth measurements to determine the strength of ultrasound evaluation in determining subsequent growth. There was significant sexual dimorphism in all measures at birth, with males heavier, longer, and leaner than females. From 20 wk of gestation onwards, males had a significantly larger head size than females. Parity, maternal height, and body mass index were important determinants of birth weight (p < 0.001). Cigarette smoking influenced birth weight, length, and head circumference (p < 0.001) but had no effect on placental size. Principal component analysis revealed that proportionality was the predominant size/shape at birth (55% of variance explained). A further 18% of variance was explained by a contrast between weight, head circumference, and length versus three skinfolds. Anthropometric measures as assessed by ultrasound at 20 and 30 wk gestation were poor predictors of birth length, weight, and head circumference (adjusted R(2) 18, 40, and 28% at 30 wk gestation scan, respectively). These predictions were not improved by including growth patterns between 20 and 30 wk. There is sexual dimorphism in a number of anthropometric measures at birth and in utero. These sex differences are important determinants of body size and shape. In a low risk population delivering at term, body shape was largely determined by proportionality between anthropometric measures. The low correlations between antenatal measures and birth size suggest that it is unwise to ascribe birth shape phenotypes to adverse events at any particular stage of gestation. The weak relationship also suggests that routine antenatal scans around 30 wk of gestation to predict growth problems are unlikely to be of benefit in the majority of cases.