A large number of studies have documented a strong correlation between size at birth and subsequent height, although the reported incidence of catch-up growth and consequently the impact on final height has varied with time and between countries. These variations may be real, but could also be related to a number of methodological problems. The aim of this study was to explore two important aspects related to postnatal growth after disturbed fetal growth: first, the definition of small for gestational age (SGA), including the selection of cut-off points in defining shortness; and, secondly, the importance of the general socio-economic status of the population with regard to the incidence of growth faltering in early life. Data were analysed from two longitudinal population-based studies, one from Sweden and one from Hong Kong. Of the Swedish cohort, 3.8% had a birth length below -2 SD scores; in the Hong Kong population the corresponding value was 11.9% (Swedish reference values were used in both studies). The following conclusions were made. Size at birth is important for postnatal growth, and the difference in length at birth of 9-10 cm between the two extreme birth length subgroups remains, on average, until maturity. This seems to be true for the two study populations with different degrees of socio-economic development. However, the rate of catch-up growth is highly dependent on the definition of SGA, on the rate of catch-up growth in early life and on the incidence of growth faltering between 6 and 18 months of age.