A large and compelling experimental literature has documented the adverse impact of prenatal alcohol exposure on the developing brain of the offspring. This is the first report of adolescent attention/memory performance and its relationship with prenatal alcohol exposure in a population-based, longitudinal, prospective study (n = 462) involving substantial covariate control and "blind" examiners. Prenatal alcohol exposure was significantly related to attention/memory deficits in a dose-dependent fashion. A latent variable reflecting 13 measures of maternal drinking was correlated 0.26 with a latent variable representing 52 scores from 6 tests measuring various components of attention and short-term memory performance. The number of drinks/occasion was the strongest alcohol predictor. Fluctuating attentional states, problems with response inhibition, and spatial learning showed the strongest association with prenatal alcohol exposure. A latent variable reflecting the pattern of attention/memory deficits observed at 14 years correlated 0.67 with a composite pattern of deficits previously detected on neurobehavioral tests administered during the first 7 years of life. The 14-year attention/memory deficits observed in the present study appear to be the adolescent sequelae of deficits observed earlier in development. As is usual in such studies, not all exposed offspring showed deficits.