Some evidence suggests that light-to-moderate alcohol consumption during pregnancy is associated with adverse outcomes in the offspring, but the precise biological mechanisms underlying such associations are currently unknown. Epigenetic modifications have been suggested as one potential explanation. Within the Pregnancy and Childhood Epigenetics (PACE) consortium, we performed meta-analysis to combine information from six population-based birth cohort studies to investigate DNA methylation at over 450,000 sites in the cord blood of newborns differentially exposed to alcohol in utero. We were primarily interested in the effects of sustained consumption throughout pregnancy (data available for five cohorts, 3,075 mother-child pairs), which represents a prolonged prenatal exposure to alcohol, but we also explored binge-drinking and timing-specific exposures. In addition to looking for differential methylation at individual CpG sites, we also used two different methods, Comb-P and DMRcate, to identify differentially methylated regions (DMRs). We found no strong evidence of association between any of our alcohol exposure measures and DNA methylation at any individual CpG site. Using Comb-P, we identified 19 DMRs in the offspring of mothers who drank throughout pregnancy compared to the offspring of mothers who gave up drinking at the start of pregnancy, but these were not validated using DMRcate. In this multi-cohort study of the general population we found no evidence that maternal alcohol consumption during pregnancy is associated with offspring cord blood DNA methylation, which is in stark contrast to the multiple, strong associations that previous studies have found for maternal smoking. However, it is possible that a combination of a larger sample size, higher doses, different timings of exposure and a more global assessment of genomic DNA methylation might show evidence of association.