The purpose of this study was to assess whether intrauterine growth retardation was associated with an increased risk of sudden infant death syndrome (SIDS). A total of 148 SIDS cases were identified from the Upstate New York (exclusive of New York City) live birth cohort for 1974 (n = 132,948). Dead controls represented all other sudden deaths (n = 114). Live controls were randomly selected and matched to cases on mother's age, race, parity, and residence and infant's birth date (n = 355). Data were collected from vital certificates (response, 97%), medical records (89%), and autopsy reports (100%). Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated with the use of logistic regression techniques to control for confounding. With live controls, significant risks were observed for gestations less than 37 weeks (OR = 2.2, CI 1.2-4.1), birth weights less than 2,500 g (OR = 2.5, CI 1.3-5.0) and birth lengths less than or equal to 47.0 cm (OR = 3.4, CI 1.8-6.4). Birth length less than or equal to 47.0 cm was the only significant risk factor observed when dead controls were used (OR = 2.9, CI 1.3-6.8). Risk decreased with increasing gestation and birth size. Postterm infants (greater than or equal to 42 weeks) were at lowest risk (live controls OR = 0.9, CI 0.5-1.6; dead controls OR = 0.6, CI 0.3-1.1). When gestational age was controlled for, SIDS infants were found to have reductions in both weight and length; this suggests that responsible mechanisms begin early in pregnancy.