The purpose of this study was to examine the effects of overall dietary variety, variety among major food groups, and variety within major food groups on dietary quality. Nutritional adequacy, one aspect of dietary quality, was measured by a Mean Adequacy Ratio (MAR)--an index of the percent of recommended intake for 11 nutrients. Other dietary quality measures included the percent of calories from fat and sugar and total intakes of energy, cholesterol, and sodium. A study sample of 3,701 individuals was selected from USDA's 1977-78 Nationwide Food Consumption Survey, excluding pregnant and lactating women and children under 1 year of age. Multiple regression analyses were performed to examine the relationships between each type of variety and each diet quality measure, controlling for age, sex, the number of foods, and all of their two-way interactions with variety. The variety terms added a significant increment to the variation in MAR that was explained by each of the models. Variety among five major food groups explained as much of the variation in MAR as did variety within those groups. Thus, dietary variety might best be defined as simply including foods from each of the major groups. Increases in this type of variety were associated with greater increases in MARs for females than for males and for persons with lower vs. higher numbers of foods. None of the types of variety could account for a sizeable proportion of the variation in the intakes of energy, fat, sugar, sodium, or cholesterol. That is, those measures were not related in any appreciable way to variety per se--either to the expansion or to the restriction of food choices. The key to limiting the intake of those constituents may be to selectively alter the scope of food choices to include more of some foods and less of others.