Growth percentiles require periodic revision because of changes in the ethnic mix of a population and socioeconomic factors. Anthropometric measurements were derived from singleton livebirths, without lethal malformations, from 22 completed weeks' gestation, at the Mercy Hospital for Women, Melbourne, from 1980 to 1989 (49,429 infants). Infants were included if reasonable assessment of gestation was available. Birth-weight percentiles were derived for the study group, male and female infants separately and for infants whose mothers were born in South-East Asia (Vietnam, Laos and Cambodia). Percentile charts for length, head circumference and ponderal index were also prepared. When compared with the intrauterine birth-weight growth curves reported by Kitchen et al (4) there was generally an elevation of all percentiles. Male infants were larger than female infants. Infants delivered by mothers born in South-East Asia were smaller than the study group as a whole, although the 10th percentiles for birth-weight were similar to the study group especially from 36-39 weeks' gestation. Periodic review of local standards is required to correctly categorize newborn infants' growth characteristics; factors such as sex of the infant and ethnic origin of the mother should be considered.