Investigation of the relationship between iron deficiency, iron supplementation and susceptibility to infection, was suggested by the author's initial observations of an association of anaemia with serious bacterial infections in infancy in Papua New Guinea. The bulk of previous longitudinal clinical intervention studies in infancy showed beneficial effects of iron supplementation. However, defects of control and design and recording in these studies and contradictory anecdotal reports left the question unresolved. A prospective, placebo-controlled, randomised, double-blind trial of iron prophylaxis (3ml intramuscular iron dextran = 150mg Fe) to two month old infants was carried out on the North Coast of Papua New Guinea where there is high transmission of malaria. A literature review, pilot studies, protocol, demography, geography and laboratory methods developed are described. Findings indicate that the placebo control group became relatively iron deficient over the first year of life and that the iron dextran group had adequate, although not excessive iron stores and a higher mean haemoglobin; however, the prevalence and effects of malaria recorded in the field were higher in the iron dextran group. Analysis of field and hospital infectious morbidity in the trial indicated a deleterious association with iron dextran for all causes including respiratory infections (the main single reason for admission). Total duration of hospitalisation was significantly increased in the iron dextran group. Analysis of other factors showed (1) a higher admission rate associated with low weight-for-height recorded at the start of the trial; (2) a significant positive correlation between birth haemoglobin and hospital morbidity rates; (3) increased malaria rates in primiparous mothers of the cohort infants who received iron infusion during pregnancy; (4) lower relative risk of malaria associated with iron prophylaxis in individuals with alpha thalassaemia, which was found to be highly prevalent in this region. In conclusion, it is suggested that policies of iron supplementation, total dose iron injection and routine presumptive iron therapy for anaemia which are widely in practice in malaria endemic areas should be closely reviewed.