The precondition for the antineoplastic effect of ifosfamide (Ifo) is the oxydation of the oxazaphosphorine ring system. The 'ring oxidation' by several steps leads to the formation of alkylating mustard. A second metabolic pathway results in the cytostatically inactive metabolites 2- and 3-deschlorethyl-ifosfamide (2-, and 3-d-Ifo). This 'side chain oxidation' plays a quantitatively more important role with ifosfamide than with cyclophosphamide. The urinary excretion of unmetabolized Ifo and of 2- and 3-d-Ifo, which represents the amount of Ifo not activated, has been investigated by capillary gaschromatography in 18 treatment cycles of 14 children with various therapeutic schedules (800, 2000 and 3000 mg/m2 given as bolus, over 1 h or as continuous infusion for 2 respectively 5 days). The total cumulative excretion during 12 completely sampled cycles ranged from 27 to 50% of the given Ifo. Between 14 and 34% could be detected as IFO, 9 to 29% as 3-d-Ifo and 2 to 8% as 2-d-Ifo. In 4 children the amount of 'side chain oxidized' IFO exceeded the amount of unmetabolized Ifo. 24 hours after the end of therapy excretion is nearly complete. No relationship between the excreted amount of unmetabolized Ifo, 2-d-Ifo and 3-d-Ifo and the dose schedule could be determined. There was no clearcut dependency between cumulative dose, the children got in the past, and formation of 2- and 3-d-metabolites.