The total cerebral haemoglobin concentration (tHb in mumol/l) as a major indicator of the oxygen transport capacity is investigated in neonates. Two methods to determine tHb by near infrared spectrophotometry (NIRS) have evolved so far: The first method requires a slow oxygenation change with reference to arterial oxygen saturation (tHbo-method). The second method is based on a geometrical principle and a two channel NIRS instrument (tHbg-method). The aim of this study was to compare both methods quantitatively. 15 clinically stable preterm infants needing supplemental oxygen were included in this study. For each method the measurements of three infants were excluded due to unsatisfactory measurement quality. The remaining 9 neonates had a mean gestational age of 29 (range 25.1 to 31.4) weeks, birthweight of 1272 (740 to 1690) g and a postnatal age of 2.6 (0.5 to 5) days. In each infant 6 tHbo measurements were carried out. During each tHbo measurement the mean of the continuously available tHbg (Cerebral Redox Monitor 2020, Johnson & Johnson Medical) was calculated. The mean of all successful tHbo and corresponding tHbg was determined for each infant. The mean tHbg was 151 mumol/l (range 62 to 223 mumol/l) and the mean tHbo was 59 mumol/l (27 to 113 mumol/l). The regression line between the two methods was tHbg = 1.34 x tHbo + 72 mumol/l. The r was 83.6%. The correlation suggests, that both methods can be applied to measure tHb. However, it has to be taken into account that the tHbg-method returns significantly higher values than the tHbo-method.