Propylene oxide (PO) is used as an intermediate in the chemical industry. Human exposure to PO may occur in the work place. Propylene, an important industrial chemical and a component of, for example, car exhausts and cigarette smoke, is another source of PO exposure. Once taken up in the organism, this epoxide alkylates macromolecules, such as haemoglobin and DNA. The aim of the present investigation was to compare two methods for determination of in vivo dose, the steady state concentration of PO in blood of exposed rats and the level of haemoglobin adducts. Male Fischer 344 rats were exposed for 4 weeks (6 h/day, 5 days/week) to PO at a mean atmospheric concentration of 500 ppm (19.9 micromol/l). Immediately after the last exposure blood was collected in order to determine the steady state concentration of PO. Free PO was measured in blood samples of three animals by means of a head space method to be 37 +/- 2 micromol/l blood (mean +/- S.D.). Blood samples were also harvested for the measurement of haemoglobin adducts. N-2-Hydroxypropyl adducts with N-terminal valine in haemoglobin were quantified using the N-alkyl Edman method with globin containing adducts of deuterium-substituted PO as an internal standard and N-D,L-2-hydroxypropyl-Val-Leu-anilide as a reference compound. Tandem mass spectrometry was used for adduct quantification. The adduct levels were < 0.02 and 77.7 +/- 4.7 nmol/g globin (mean +/- S.D.) in control animals (n = 7) and in exposed animals (n = 34), respectively. The adduct levels expected at the end of exposure were calculated to be 71.7 +/- 4.1 nmol/g globin (mean +/- S.D.) using the measured steady state concentration of PO in blood and taking into account the growth of animals, the life span of erythrocytes, the exposure conditions and the second order rate constant for adduct formation. The good agreement between the estimated and measured adduct levels indicates that both end-points investigated are suitable for biological monitoring.