this report concerns the development and stability of recall memory at 12, 24, and 36 months of age in preterm (<1750g birthweight) and term children from a prospective, longitudinal study. there were 56 preterm children (55.4% male; mean birthweight 1119.5g [sd 279.2], range 551 to 1742g; mean gestational age 29.7 weeks [sd 2.9], range 25 to 36wks) and 126 term controls (51.6% male; mean birthweight 3453.8g [sd 438.2], range 2500 to 4564g; mean gestational age 40.2 weeks [sd 1.1], range 38 to 42wks). recall memory was assessed using an elicited imitation task in which children attempted to reproduce a sequence of actions 15 minutes after they were modeled by the examiner. relative to term children, the preterm group performed poorly at all three ages: they reproduced fewer actions overall and fewer in the correct order. lower birthweight and lengthier postnatal hospitalization were associated with poorer performance in preterm children at 24 and 36 months. although performance improved considerably for both groups with age, there was no evidence of preterm ‘catch-up’: the deficits detected at 12 months did not diminish with age. individual differences in recall were modestly stable from one year to the next, with cross-age correlations highest from 24 to 36 months (r=0.43 to 0.49). these results suggest that preterm children have deficits in recall memory that emerge by 12 months and persist into early childhood.