This conference resulted in a foundation upon which consensus will be built in the future. Further, it provided a forum for the articulation of the critical need for carbohydrates in human nutrition. We have come a long way in the past 25 years in the food and nutrition sciences and this conference well illustrated that point. At the White House Conference on Nutrition in 1969, macronutrients were never even mentioned as being necessary to nutrition; only micronutrients were emphasized. In 1977 the term complex carbohydrate was used without definition in the Dietary Goals for the United States. This conference encapsulated the tremendous strides that have been made concerning the chemical, nutritional, biologic, and physiologic importance of carbohydrates in health and disease. It is fascinating to review the recommendations concerning complex carbohydrate from 1977 on. Recommendations were made by the Senate Select Committee in 1977; the Dietary Guidelines of 1980, 1985, and 1990; the Surgeon General's report of 1979 and 1988; the 10th edition of Recommended Dietary Allowances; the 1989 National Academy of Sciences Diet and Health: Implications for Reducing Chronic Disease Risk; the 1991 FDA proposal for labeling; and the 1992 Food Guide Pyramid. Indeed, as Dr. Joanne Slavin pointed out, "there is a wide range of consensus that increased complex carbohydrate is consistent with good health," and certainly as Dr. Gary Henderson said, "the term complex carbohydrate has equity with consumers." With this level of apparent clarity it is difficult at first to identify a problem for the food label. After all, if consumer health would benefit, why not simply do it? However, a problem does exist in that carbohydrates may be classified chemically in groups that are difficult to analyze and to define as to their physiologic function in the body. Indeed, complex carbohydrates defined by chain length may have molecular structures less complex than those of smaller units and may vary in solubility and digestibility. A continuum does exist in each of the chemical, analytical, and physiologic characteristics of carbohydrates. Scientists do not like a continuum and regulators probably like it even less, but such a continuum holds great promise for future consensus. This continuum provided the thoughtful alternatives presented above by Drs. Jonathan DeVries, Dennis Gordon, and Alison Stephen. Dr. DeVries suggested a scheme of complex carbohydrates, sugars, and sugar alochols, while Dr. Gordon suggested complex carbohydrates, dietary fiber, and simple carbohydrates, and Dr. Stephen suggested starch, sugar, and nonstarch polysaccharides.(ABSTRACT TRUNCATED AT 400 WORDS)