Are serious growth delays caused by malnutrition and neglect permanent or reversible? The effects of institutionalization and international adoption on children's physical growth are estimated with meta-analysis. Studies with sufficient data to compute differences between adoptees and the reference population (33 papers with 122 study outcomes) were collected through Web of Science, ERIC (Education Resource Information Center), PsycINFO (Psychological Literature), and Medline (U.S. National Library of Medicine) (1956-2006). The influence of pre- and postadoption care on height, weight, and head circumference was tested. Effect sizes (d) and confidence intervals (CIs) around the point estimate for the growth lag indices were computed. The more time spent in institutional care, the more the children lagged behind in physical growth (d = 1.71, 95% CI: 0.82-2.60, n = 893). At adoptive placement, the children showed large delays in height, weight, and head circumference (d = -2.39 to -2.60; n = 1331-3753). Although after adoption, they showed almost complete catch-up of height (d = -0.57, 95% CI: -0.87 to -0.27, n = 3437 adoptees) and weight (d = -0.72, 95% CI: -1.04 to -0,39, n = 3259 adoptees), catch-up of head circumference seemed slower and remained incomplete (d = -1.56, 95% CI: -2.27 to -0.85, n = 527). Later age at arrival was related to less complete catch-up of height and weight. International adoption leads to substantial catch-up of height and weight but not of head circumference, demonstrating differential plasticity of children's physical growth.