A routine method to determine total haemoglobin mass (tHb) in clinical practice and sports medicine is non-existent. Radioactive tracers or other dilution procedures like the common CO-rebreathing method (Proc(com)) are impractical, the latter in particular because of the relatively long time of respiration. According to the multicompartment model of Bruce and Bruce (J Appl Physiol 95:1235-1247, 2003) the respiration time can be considerably reduced by inhaling a CO-bolus instead of the commonly used gas mixture. The aim of this study was to evaluate this theoretical concept in practice. The kinetics of the HbCO formation were compared in arterialised blood sampled from an hyperaemic earlobe after inhaling a CO-bolus (Proc(new)) for 2 min and a CO-O(2) mixture (Proc(com)) for approximately 10 min. The reliability of Proc(new) was checked in three consecutive tests, and phlebotomy was used to determine the validity. VO(2max) was determined with and without previous application of Proc(new) and the half-time of HbCO was registered also in arterialised blood after resting quietly and after the VO(2max) test. Proc(new) yielded virtual identical tHb values compared to Proc(com) when HbCO determined 5 min after starting CO-rebreathing was used for calculation. The typical error of Proc(new) was 1.7%, corresponding to a limit of agreement (95%) of 3.3%. The loss of 95 g (19) haemoglobin was detected with an accuracy of 9 g (12). After application of Proc(new) VO(2max) was reduced by 3.0% (3.7) (P=0.022) and half-time was lowered from 132 min (77) to 89 min (23) after the VO(2max) test. Inhaling a CO-bolus markedly simplifies the CO-rebreathing method without reducing validity and reliability and can be used for routine determination of tHb for various indications.