SummaryThe literature on the cortisol concentration in cord plasma, the transplacental transfer of cortisol, and the plasma protein‐binding of cortisol is reviewed. During pregnancy there is a progressive rise in the plasma levels of 17‐hydroxycorticosteroids. It has been demonstrated that cortisol can cross the placenta, and that the concentration of 17‐hydroxycorticosteroids in cord blood at birth are merely the reflection of maternal levels. The ratio of maternal to foetal plasma levels lies between 5:1 and 2:1. In cases of vaginal delivery both the maternal and cord plasma levels are elevated in comparison to those in cases of elective Caesarean section. The concentrations of 17‐hydroxycorticosteroids in plasma from diabetic mothers and in cord plasma from their infants are not significantly different from the levels in nondiabetic mothers and their infants. Studies primarily designed to determine the 17‐hydroxycorticosteroid level in cord blood from premature infants have apparently not been carried out. Most of the circulating cortisol is bound to a specific a‐globulin (the corticosteroid‐binding globulin, CBG), and there is only a minor freely diffusible, and presumably biologically active, fraction. During pregnancy there is a rise in the CBG‐binding capacity of maternal plasma, whereas the CBG‐binding capacity of cord plasma from full‐term infants is less than that found in normal adults and children. There are no available data in the literature on the CBG‐binding in cord plasma from premature infants or infants of diabetic mothers.The percentage of CBG‐bound cortisol in diluted plasma and the total plasma cortisol were determined in cord plasma from 9 full‐term infants, 9 premature infants, and 11 infants of diabetic mothers, and in plasma from their mothers at delivery.The principles of the method and the calculations are outlined. The percentage of CBG‐bound cortisol in diluted plasma was determined by dialysis against human albumin solution. The total plasma cortisol was determined fluorometrically after chromatography on silica gel. An evaluation of the method seemed to indicate reasonable degree of specificity, precision, and accuracy.The percentage of CBG‐bound cortisol in diluted maternal plasma was invariably higher than that found in the cord plasma. In diluted maternal plasma the percentage of CBG‐bound cortisol was significantly higher in mothers who gave birth to full‐term infants than in mothers who gave birth to premature infants, and there was furthermore a positive correlation between this percentage and the period of gestation at delivery. Similarly the percentage of CBG‐bound cortisol in diluted cord plasma was significantly higher in full‐term than in premature infants, and there was a very good positive correlation between this percentape and the gestational age of the infants. There was, moreover, a good correlation between the percentage of CBG‐bound cortisol in diluted maternal plasma and the percentage found in diluted cord plasma from the corresponding infant. The percentage of CBG‐bound cortisol, both in diluted plasma from the diabetic mothers and in diluted cord plasma from their infants, fell within the normal range that would be expected from the period of gestation at delivery.The concentration of total plasma cortisol in maternal plasma was invariably higher than that found in cord plasma. The levels in plasma from diabetic mothers and in cord plasma from their infants were not significantly different from those in nondiabetic mothers and their infants.The calculated non‐CBG‐bound cortisol concentration was significantly higher in mothers delivered prior to term than in mothers delivered at term. Similarly, the premature infants yielded a mean value of non‐CBG‐bound cortisol in cord plasma which was significantly higher than that found in full‐term infants. The mean concentration of non‐CBG‐bound cortisol in maternal and cord plasma from the diabetic mothers and their infants fell between those of the two categories of nondiabetic mothers and their infants. There was no obvious relationship between the severity of maternal diabetes, or degree of ‘Cushingoid’ appearance of the infants at birth, and the magnitude of the total plasma cortisol concentration, or the non‐CBG‐bound cortisol concentration, either in maternal or cord plasma.With regard to the total plasma cortisol the mean ratio of maternal to cord concentrations was approximately 3: 1, whereas the mean ratio of maternal to cord non‐CBG‐bound cortisol concentrations was about 1:1.It is suggested that the low CBG‐binding capacity in cord plasma, and the correlation of this capacity with gestational age, might reflect an increasing supply of oestrogens acting upon a deficient CBG synthesis. The concentration of non‐CBG‐bound cortisol was essentially the same in. maternal and cord dasma. in sham con‐ trast to the concentration gradient of total cortisol between maternal and cord plasma. It appears that this concentration gradient occurs as a result of an equilibrium in which there is a higher CBG‐binding capacity on the maternal side than on the foetal side of the placenta. During foetal life this equilibrium system allows the foetal cortisol concentration to be under the control of maternal homeostasis. At delivery the maternal plasma cortisol rises to a level substantially above the CBG‐binding capacity. The non‐CBG‐bound cortisol level rises correspondingly, and more cortisol is transferred to the foetal side of the placenta. In this way the foetus is supplied with a surplus of maternal cortisol at birth. The lower CBG‐binding capacity in premature infants is compensated for by a higher proportion of non‐CBG‐bound, and presumably biologically active, cortisol than that in full‐term infants, but in both categories of infants this cortisol fraction is considerably higher than that found under physiological conditions later in life.It appears that the present study of the CBG‐binding of cortisol, of the total cortisol concentration, and of the non‐CBG‐bound cortisol concentration in cord plasma has not supported the suggestion that the ‘Cushingoid’ appearance of some infants of diabetic mothers could be attributed to an increased exposure to adreno‐corticosteroids during foetal life. However, both the CBG‐binding of cortisol and the total cortisol concentration in maternal and cord plasma are modified during parturition to such an extent that the findings after delivery hardly permit of any conclusion as to the condition in utero prior to labor. The finding of an almost equal concentration of non‐CBG‐bound cortisol in maternal and cord plasma suggests that serially performed measurements of this cortisol fraction during pregnancy might prove useful in providing information on the concentration of biologically