The intellectual, psychoeducational, and functional status of a regional cohort of extremely low birth weight (ELBW) survivors who weighed 501 to 1000 gm at birth (n = 143) and who were born between 1977 and 1981 was compared with that of control children born at term (n = 145) who were matched for gender, age, and social class. One hundred twenty-nine ELBW survivors (90%) were available; their mean birth weight was 839 +/- 124 gm and mean gestational age 27 +/- 2.1 weeks, and 48 of them weighed less than or equal to 800 gm at birth. Both ELBW and control groups were tested at a mean unadjusted age of 8 years; 113 of 129 ELBW children completed the full test battery, eight blind children had other tests, and eight were not testable. The mean Full Scale IQ (Wechsler Intelligence Scale for Children--Revised) was 91 +/- 16 for ELBW children and 104 +/- 12 for control children (p less than 0.0001). Between 8% and 12% of the ELBW group scored in the "abnormal" range (less than or equal to -2 SD) on the Wechsler IQ and subtests, compared with 1% to 2% of the control group. The ELBW group did less well on the reading, spelling, and mathematics tests (Wide Range Achievement Test--Revised less than or equal to -2 SD: ELBW = 20% to 28%; control = 3% to 10%). The motor performance of the ELBW group (Bruininks-Oseretsky Test of Motor Proficiency less than or equal to -2 SD: ELBW = 20%; control = 1%) and their visual-motor integration (Beery Developmental Test of Visual-Motor Integration less than or equal to -2 SD: ELBW = 21%; control = 6%) were also poorer. Exclusion of 19 ELBW children with neurologic impairments or an IQ less than or equal to 70 or both did not result in significant improvement in Wechsler or achievement measures. Approximately 15% of the ELBW cohort performed in the abnormal range on the Vineland Adaptive Behavior Scales, compared with none of the control group. Although approximately two thirds of the ELBW group were performing in the normal range on intellectual measures, comparison with the control group suggests that, as a group, ELBW children were significantly disadvantaged on every measure tested.