This paper addresses the expected impact of use of macronutrient substitutes by individuals without strong motivation to control macronutrient or energy intake when such modified foods serve as replacements of foods normally ingested rather than as "add-ons". A basic premise of the paper is that individuals are likely to replace a substantial part of the energy equivalent of the original substitution. It further assumes that the macronutrient composition of the additional foods consumed will reflect normal food selection behaviours. The paper derives a description of the expected selection behaviour from examination of the within subject variance in energy and macronutrient intake of 29 subjects followed for 365 consecutive days (the Beltsville One Year Dietary Intake Study). Patterns observed in these subjects were validated through examination of associations between macronutrient intake and energy intake in 600 women, each of whom contributed 6 days of dietary data (USDA CSFII-85). As presently proposed the model suggests that the use of non-caloric fat replacements, by subjects without strong motivation to control fat or energy intake, can be expected to result in a net decrease (less than original substitution) in fat intake and net increases in carbohydrate and protein intakes. Conversely, use of carbohydrate replacements in core foods can be expected to result in net increases in fat and protein intakes and a partial decrease in carbohydrate intake. The magnitude of these net changes is seen to be a function of the extent of replacement of energy.