In response to the Deaton–Dreze (2009) explanation of a downward shift in the calorie Engel curve in terms of lower requirements due to health improvements and lower activity levels, we have developed an alternative explanation of changes in the consumption of calories, protein and fats over the period 1993–2004. This explanation is embedded in a standard demand theory framework, with food prices and expenditure (as a proxy for income) cast in a pivotal role. Based on different experiments, robust demand functions are estimated for each of three nutrients viz. calories, protein and fats, separately for rural and urban areas. Our results show consistently robust food price and expenditure effects. Besides, shifts in food price elasticities over time are captured. Over and above these effects, there are shifts in demands due to factors other than those specified in the demand equation. In the context of calories, for example, it is plausible that part of the reduction in their consumption was due to health improvements and less strenuous activity levels — especially but not necessarily confined to rural areas. So, while the Deaton–Dreze (2009) explanation is not rejected, it is arguable that it is complementary to the demand-based explanations Key words: Nutrients, Prices, Expenditure, .Demand, Rural, Urban, India.