We examined the relationship between consumption of fruit and vegetables during pregnancy and anthropometric measures at birth in a general population mother-infant cohort in Valencia, Spain. A total of 787 infants born between May 2004 and February 2006 were included. Fruit and vegetable consumption during pregnancy was assessed by a FFQ administered using an in-person interview. We used multiple linear regression to assess associations between fruit and vegetable intake (in quintiles) and birth weight and length adjusted for sex and gestational age, and logistic regression to assess being small for gestational age (SGA) in weight and SGA in length, defined as adjusted birth weight or length below the 10th percentile. A linear relationship was found between vegetable consumption and having a SGA (weight) and SGA (length) baby. Women in the lowest quintile of vegetable intake during the first trimester had a higher odds of having a SGA (weight) baby than women in the highest quintile [odds ratio (OR), 3.7; 95% CI: 1.5-8.9; P-trend < 0.001] and had a higher odds of having an SGA (length) baby in the third trimester (OR, 5.5; 95% CI: 1.7-17.7; P-trend = 0.04) in multivariate analysis. We found a nonmonotonic relationship between adjusted birth weight and length and vegetable consumption during the first trimester; newborns in the 2 lowest quintiles of intake had a significantly lower weight and length than those in the 4th quintile. There was no association between fruit consumption and birth outcomes. Our findings indicate that vegetable consumption throughout pregnancy may have a beneficial effect on fetal growth.