The bioavailability of iron on Fe(III)-hydroxide-polymaltose complex was compared intraindividually with that of Fe (II)-ascorbate (iron absorption) and a Fe (II)-sulphate quick release preparation (haemoglobin regeneration test). The study was carried out in a population of 16 healthy male volunteers, phlebotomized in weekly intervals until development of an iron deficiency anaemia in order to establish a test population with only small variations of their individual body iron status. Intestinal iron absorption in fasting state as measured by 59Fe whole body retention and simultaneous estimation from plasma iron tolerance curves was low for the 59Fe (III)-complex (1.2 +/- 0.1% (means +/- SD] as compared to the 59Fe (II)-ascorbate (43.7 +/- 7.1% (means +/- SD]. Iron administration together with a test meal did not affect the absorption from the 59Fe (II)-ascorbate, whereas the 59Fe (III)-complex showed a significant increase in absorption (8.8 +/- 4.7% (means +/- SD]. Haemoglobin (Hb) regeneration after 100 mg of iron as Fe (III)-complex and Fe (II)-sulphate administered during 28 days with meals amounted to a mean of 0.68 +/- 0.2 g/l and 1.1 +/- 0.3 g/l (means +/- SD) of total daily Hb-increase and a net-Hb-increase of 0.31 +/- 0.37 g/l and 0.79 +/- 0.36 g/l (means +/- SD), respectively. There was also a small but significant rise in serum ferritin during the oral treatment in both treatment groups. The results of Hb-regeneration after treatment with the Fe (III)-complex were in the range which could be expected from the absorption measurements.(ABSTRACT TRUNCATED AT 250 WORDS)