Placental hemosiderin deposition representing intrauterine bleeding at least 24–48 h before delivery is detected frequently in prematurity. The objective of this study was to assess incidence and site of histologic evidence of intrauterine bleeding in association with fetal growth in prematurity. Placentas of consecutive nonanomalous singleton liveborns delivered <32 weeks of gestation were studied for the presence of hemosiderin in decidua of the placental basal plate or extraplacental membranes (confirmed by Prussian blue stain). Cases of placenta previa, clinical abruption, or coagulopathy and cases in whom obstetric and neonatal gestational age assessment differed by >2 weeks were excluded. A single reviewer blinded to clinical data except for gestational age at delivery assessed the presence of decidual hemosiderin. Statistical analysis included ANOVA, and Mann-Whitney U test with p < 0.05 considered significant. The study included 352 patients delivered for principal indication of premature rupture of membranes (PROM) or preterm labor (PTL) and 78 patients delivered for preeclampsia between 1989 and 1994. Mean birth weight percentiles for neonates delivered following PROM/PTL versus preeclampsia were: no decidual hemosiderin 42 ± 25 versus 17.4 ± 25, extraplacental membrane hemosiderin 42 ± 25 versus 9.2 ± 10, placental basal plate hemosiderin 42 ± 25 versus 17 ± 24, and hemosiderin in both sites 27 ± 21 versus 6.4 ± 10 (p = 0.02). Hemosiderin deposition in both placental basal plate and extraplacental decidua is associated with significantly lower mean birth weight percentiles in PROM/PTL at less than 32 weeks of gestation. We postulate that in these patients placental disruption which accompanies decidual bleeding may explain the relatively impaired fetal growth. In preeclampsia, hemosiderin depositions are not associated with further impaired fetal growth.