(A) Schematic of crossover experiment. Pregnant germ-free dams received cultured microbial collections from a preterm infant with poor growth in late gestation by gavage to facilitate perinatal colonization of neonates. On P2 and P4, cultured bacterial collections from an infant with appropriate growth were applied topically on the mother to facilitate transmission of strains to the suckling neonates. Weight and length on P20–21 were compared with mice treated with bacteria from an extremely preterm infant with appropriate growth alone and mice treated with bacteria from an extremely preterm infant with poor growth alone. (B) The proportion of ASVs in the fecal microbiotas of mouse pups in the crossover group that were present in the fecal microbiota of the preterm infant donor with poor growth (red), the preterm infant donor with appropriate growth (gray), or present in both donor infant microbiotas (pink). (C) The microbiotas of pups in the crossover group had greater similarity (lower Jensen-Shannon divergence) to pups treated with the appropriate growth microbiotas alone than pups treated with the poor growth microbiotas alone. (D) Relative abundances of bacterial taxa at the lowest assigned taxonomic level in the pups that received the poor growth microbiotas alone, pups in the crossover group before and after topical introduction of microbial consortia from infants with appropriate growth, and pups that received the appropriate growth microbiotas alone. (E) Mice in the crossover group (18 pups in 3 litters) had significantly higher weight and length than mice who received only bacteria from infants with poor growth (5 litters) but not significantly different from mice who received bacteria from infants with appropriate growth (7 litters). Appr, appropriate. **P < 0.01; ***P < 0.001 by Wilcoxon’s rank-sum tests with Benjamini-Hochberg correction.