Iron chelation therapy for patients maintained on a regular transfusion regime is at present best carried out by means of daily infusions of desferrioxamine (Hussain et al 1977; Pippard et al 1978) but this is onerous for the patient and has social and economic disadvantages. Many recent attempts to provide more effective drugs for iron chelation have been summarized by Jacobs (1979) and increasing attention is now being paid to the possibility of oral iron chelation therapy. Hoy et al (1979) showed that when isonicotinic acid hydrazide (INH) and pyridoxal are mixed in equimolar amounts a hydrazone is formed which chelates iron, and oral administration of this compound to rats results in an eightfold increase in faecal iron excretion. It is effective on repeated administration (Cikrt et al 1980), the main route of iron excretion being through the bile. Long term studies in the rat have not been successful in reducing the iron load of test animals and this appears to be related both to their high dietary iron content and instability of the hydrazone. Its effective shelf life at room temperature is no longer than one month and this is a considerable disadvantage from a therapeutic point of view. Pyridoxal is known to form a Schiff base with many amino acids and its reactivity has led us to examine complexes of pyridoxal with a number of substances in an attempt to find an alternative iron chelator of greater stability than the INH complex and of comparable effectiveness on oral administration. The screening procedures used were the effects on Chang cell iron metabolism (White et al 1976) and on iron excretion in the rat (Hoy et al 1979).