To assess the combined risks of being born preterm and to a teenage mother, and to evaluate the effects of an early intervention, preterm infants born to lower-class, black, teenage mothers were provided a home-based, parent-training intervention, and their development was then compared with that of nonintervention controls, of term infants of teenage mothers, and of term and preterm infants of adult mothers. Despite equivalence on prenatal care, factors which placed the preterm infant of the teenage mother at greater risk at birth were the small-for-date size of the infant and the less realistic developmental milestones and child rearing attitudes expressed by the mother. The preterm infants of teenage mothers who received intervention showed more optimal growth, Denver scores, and face-to-face interactions at 4 months. Their mothers rated their infants' temperaments more optimally, expressed more realistic developmental milestones and child-rearing attitudes, and received higher ratings on face-to-face interactions. At 8 months, the intervention group received superior Bayley mental, Caldwell, and infant temperament scores.