Birth-weight-specific mortality is lower for black prematures than white prematures of similar low birth weight. The reason for this well-recognized phenomenon is unknown. We investigated the extent to which black and white infants differ in their gestational maturity and incidence of potentially lethal risk factors, and the effect these factors might have on differential mortality risk. The population studied comprised babies bom alive in a tertiary care hospital with birth weights from 700 to 1800 g over a 41/2-year period. Univariate analysis showed no important difference between races for the incidence of lung disease, Apgar scores, birth weight, or gestational age. Infants with a birth weight below the 10th percentile (small for gestational age) were more likely to be black, and infants with a birth weight above the 90th percentile (large for gestational age) who had a survival disadvantage were usually white. The crude odds ratio for the white race was 1.79 (1.18–2.73). When the demographic, morbidity, and growth variables were put into a logistic model, the odds ratio changed only slightly, to 1.52 (1. 14–2.03). We conclude that the majority of the black-white difference in birth-weight-specific survival is not due to a higher incidence of black small-for-gestational-age infants, nor is it due to differences in major morbidities associated with death.