A prospective study of the relationships among fetal heart rate pattern, meconium staining of the amniotic fluid, umbilical cord artery pH, and Apgar score was carried out in 1219 consecutive births. Interpretable cardiotocogram patterns and cord arterial pH and blood gas analysis were obtained in 698 cases. The sensitivity of an abnormal cardiotocogram at any time for acidosis (more than 1 SD below the mean, pH less than 7.17) was 80%, and for severe acidosis (more than 2 SDs below the mean, pH less than 7.085) was 83%. However, the predictive value was low, and 32% of fetuses had an abnormal cardiotocogram but no acidosis. If only cardiotocogram abnormality in the first stage of labor was considered, sensitivity was still 47% for acidosis and 67% for severe acidosis, and the false-positive rate was reduced to only 14%. We attempted to improve the prediction of acidosis by including meconium staining of the amniotic fluid, but 65% of the variation in umbilical cord artery pH and 72 and 86% of the variation in 1- and 5-minute Apgar scores, respectively, remained unexplained. In light of these poor correlations, the current practice of considering cardiotocogram abnormality, meconium staining of the amniotic fluid, acidosis, and low Apgar scores as indicating one single disorder, "fetal distress," is not valid.