OBJECTIVE: Previously we found that certain fetal imprinted genes represented as an allele score are associated with maternal pregnancy glucose concentrations. Recently it was reported that fetal polymorphisms with strong associations with birth weight tend to mediate these independently of increases in maternal pregnancy glucose concentrations. We therefore investigated whether potential associations between the fetal allele score and birth weight were related to maternal glucose concentrations in the Cambridge Baby Growth Study. RESULTS: The fetal imprinted gene allele score was positively associated with birth weight (β = 63 (17-109) g/risk allele, β' = 0.113, p = 7.6 × 10-3, n = 405). This association was partially attenuated by adjusting for maternal glucose concentrations (β = 50 (4-95) g/risk allele, β' = 0.089, p = 0.03, n = 405). The allele score was also positively associated with risk of being large for gestational age at birth (odds ratio 1.60 (1.19-2.15) per risk allele, p = 2.1 × 10-3, n = 660) and negatively associated with risk of being small for gestational age at birth (odds ratio 0.65 (0.44-0.96) per risk allele, p = 0.03, n = 660). The large for gestational age at birth association was also partially attenuated by maternal glucose concentrations. These results suggest that associations between the fetal imprinted gene allele score and size at birth are mediated through both glucose-dependent and glucose-independent mechanisms.