Objective: To evaluate the influence of acid-base status at birth and Apgar scores on survival in very low birth weight infants. Methods: We evaluated 1073 infants born alive and weighing 500–1000 g during 1979–1991; 658 had umbilical artery gas values examined. Apgar scores were assigned at 1 and 5 minutes after birth. Umbilical artery blood samples were collected at delivery for pH, carbon dioxide pressure (PCO2), and bicarbonate. Infants were grouped at 23–24, 25–26, 27–28, and 29 weeks or more. Using survival as the dependent variable, multiple logistic regression analyses were performed controlling for gestational age, birth weight, plurality, antenatal glucocorticoid use, mode of delivery, and year of birth, as well as for Apgar scores and cord blood gases. Results: In every gestational age grouping, compared with infants with a pH lower than 7.05, survival was higher in infants with an umbilical artery pH of 7.05 or higher, significantly so at 27–28 weeks. There was no consistent relationship between umbilical artery PCO2 or bicarbonate and survival. However, with the exception of the 1-minute Apgar score at 23–24 weeks, the relationship of Apgar scores to survival was significant in all gestational age periods. Using multiple logistic regression analyses, the only significant relationships between any of the cord blood gases, Apgar scores, and mortality involved low 1-minute (odds ration [OR] 2.7 [95% confidence interval (CI) 2.0–3.6]) and low 5-minute Apgar scores (OR 2.8 [95% CI 2.0–3.8]) and a bicarbonate less than 21 mEq/L (OR 1.6 [95% CI 1.1–2.4]). Conclusion: One- and 5-minute Apgar scores are better predictors of survival than umbilical artery blood gases in neonates weighing 500–1000 g at birth. © 1996 Published by Elsevier Science Inc.