Measurement of glycosylated hemoglobins in diabetic patients has been available to clinicians for about five years. Such measurements correlate with mean serum glucose determinations over time; therefore, they have stimulated a number of studies to determine (1) if these assays are useful in diagnosing diabetes, (2) the clinical utility of determinations of minor hemoglobins in monitoring diabetic control, and (3) the relationship of glucose "control" (as indicated by concentrations of glycosylated hemoglobins) to abnormalities or "sequelae" of the diabetic state. High concentrations of glycosylated hemoglobins are highly specific for diabetes, and positive findings provide a useful diagnostic test. However, this measurement is less sensitive than a glucose tolerance test. As a clinical tool, these hemoglobins are most useful in labile diabetes, i.e., juvenile-onset diabetes and diabetes in pregnancy. In adult-onset diabetes, the fasting serum glucose concentration is apt to correlate well with the concentration of hemoglobins A1a-c. A correlation between several abnormalities associated with diabetes mellitus and concentrations of hemoglobins A1a-c have been reported. These abnormalities include abnormalities of the erythrocyte, leukocyte, platelet, and coagulation cascade and hormonal profiles in juvenile-onset diabetes and diabetes in pregnancy. In addition, correlation have been reported between certain risk factors or abnormalities, associated with vascular disease and concentrations of minor hemoglobins, including lipid profiles microvascular disease as reflected by retinal changes and quadriceps capillary basement membrane thickening, and macrovascular disease as reflected by pulse volume recordings. These studies have led to a reevaluation of the role of glucose "control" in contributing to diabetic sequelae, and, thus, have stimulated new approaches to the management of diabetes.