food intake of a population as indices of its health and well-being has been the subject of a lively debate in recent years. One major focus of the discussion is the question raised by Sukhatme (1978, 1981, 1982) about the possibility of defining objectively the quantum of calories, protein and other nutrients necessary for people of specified age, sex and activity levels to maintain normal health. He argued that apart from the variation in food requirement between individuals falling within the same age-sex activity category (which has been well recognised), there are self-regulatory mechanisms for adjusting energy expenditure to intake which enable a given individual to maintain a normal level of activity without any significant change in body weight or loss of health despite day- to-day and week-to-week variation in the level of food intake or, to be more precise, calorie intake. Sukhatme has also cited some evidence to show that intra-individual differences are much more important than inter- individual variations. On this basis he questioned the use of average nutritional norms as the basis for judging the incidence of poverty and under-nourishment, and for deciding the policy intervention appropriate to alleviating these conditions. While the existence of intra-individual variation is now generally accepted, the factors responsible for them and, even more so, their implication for the use of average nutritional norms for measuring poverty and under-nourishment continue to be matters of controversy. (See, for example, Dandekar, 1981; Krishnaji, 1981; Srinivasan 1977; and Gopalan, 1983.) Specifically, there is some question whether the claim about intra-individual variation being more important than variations between individuals is valid for a group which does not normally get enough to eat. Mechanisms by which individuals can maintain weight, health and normal activity in the face of day-to-day variations cannot be independent of the long term mean level of intake. In any case, as Sukhatme himself recognises, the importance of intra- individual variations fails as the period over which intake observations are taken increases till eventually we are left only with variations between individuals. It has of course been suggested that biological adaptation arising from inter-actions between genetic and environmental factors can take place even over the long-run, i e, in the face of changes in the sustained level of food intake by individuals. "If nutrient constraints are encountered at a given rate of growth, the rate is slowed down to bring the nutrient demand into equilibrium with nutrient supply. By thus regulating the speed of the internal physiological 'clock' short-run equilibrium is established and the ultimate size and shape of the adult may be moulded to its environment" (Seckler 1979:5). Low levels of sustained intake may on this reasoning result in a population of smaller physical stature which may nevertheless show no greater signs of ill health or clinical malnutrition than a better fed population. But clearly this is a long term process and one which has definite limits. Whether indeed sustained low levels of intake or significant reductions in the sustained level of intake leads to mere adaptation in size without impairing health or activity is an important question of fact but one which has not been adequately explored and certainly not conclusively.1 This paper examines, on the basis of some Indian survey data, the relation between sustained mean food intake of population in different regions and their mean size and health status. We recognise that these relations are inherently complex and the relevant information of the requisite quality necessary to unravel them empirically is also difficult to get. Nevertheless some information is available, and the question of how sustained low intakes and significant reduction in mean intakes affects people is particularly important in our context because a very high proportion of the population live close to the margin of subsistence and the per capita food grain production (the major component of food) has shown a declining trend in as many as eight major states during the last two decades.2 THE AVAILABLE DATA The National Institute of Nutrition (NIN) has conducted several detailed field surveys of food intake, nutritional status and anthropometric measurements in different parts of India over the last 2-3 decades. Some of these enquiries are of limited scope being confined to particular towns/villages or sections of the population. But the Institute has also carried out two large-scale sample surveys which give anthropometric data for a number of states at two points of time. This body of data provides a basis for studying the size-intake relation and, with heavy qualification, changes in size over time.
