Women who deliver preterm (<37 completed weeks' gestation) are at high risk for recurrence. This has prompted exploration of candidate genes (both maternal and fetal) associated with preterm delivery. Epidemiologists can use recurrence patterns of preterm delivery across generations to assess the relative contributions of maternal and fetal genes. The authors used data from the Medical Birth Registry of Norway (1967-2004) to identify 191,282 mothers and 127,830 fathers who subsequently had at least one singleton offspring. The authors stratified parents according to whether or not they had been born preterm and calculated the risk of preterm delivery among their firstborn. Mothers born preterm had a relative risk for preterm delivery of 1.54 (95% confidence interval (CI): 1.42, 1.67). This association was weaker for fathers born preterm (relative risk (RR) = 1.12, 95% CI: 1.01, 1.25). Among early preterm births (<35 weeks), the effect became stronger for mothers (RR = 1.85, 95% CI: 1.52, 2.27) and weaker for fathers (RR = 1.06, 95% CI: 0.77, 1.44). These data suggest that paternal genes have little, if any, effect on preterm delivery risk. This argues against major contributions of fetal genes inherited from either parent. The increased risk of preterm delivery among mothers born preterm is consistent with heritable maternal phenotypes that confer a propensity to deliver preterm.