The Department of Health (1992) has recently stated that 'Nutritional reviews concerning elderly people are especially constrained by lack of data', and that much of the emphasis in the nutritional literature has been placed on the study of institutionalized, and often chronically ill, elderly subjects rather than the non-institutionalized elderly who form the majority of this population. The present study presents information on the dietary intake and biochemical status of non-institutionalized elderly subjects (68-73 and 74-90 years) and compares such data with those obtained for adult (20-64 years) and adolescent (13-14 years) populations living within the same community. Nutrient intakes and appropriate biochemical measurements of nutrient status, performed on fasting blood samples, were statistically examined and have been discussed in relation to potential age-related influences. The nutrient intake of elderly subjects was on a par with adolescents of corresponding sex but generally lower than that of adult counterparts. There were several significant differences in biochemical measurements of nutrient status between age groups. In general these did not suggest progressive age-related trends. However, there were significant suggestions of age-related increases in whole-blood glutathione peroxidase (EC 1.11.1.9) activity, serum ferritin, plasma cholesterol, LDL and triacylglycerol concentrations and decreases in plasma HDL and ascorbic acid concentrations. The significance of these differences is discussed. An age-related difference (suggestive of a decline) in vitamin C status together with a difference (suggestive of an increase) in glutathione peroxidase activity may indicate an imbalance in the regulation of O2-derived free-radicals with ageing. These observations are worthy of a further study in the light of current thinking which relates the induction of a number of diseases to oxidative damage.