We investigated the effect of temperature, variant hemoglobins, and hyperlipidemia on determination of glycosylated hemoglobin by an electrophoretic method (Clin. Chem. 26: 1598-1602, 1980). We found that: (a) temperature variations ranging from 4 to 30 degrees C were without effect on results obtained by electrophoresis; (b) concurrent determination of glycosylated hemoglobin by electrophoresis and column-chromatography in blood specimens from 150 diabetic patients yielded almost identical mean values for both procedures when operations were carried out at 22 degrees C; (c) electrophoretic determination of glycosylated hemoglobin in whole-blood hemolysate was not affected by concentration of triglycerides; and (d) unlike column-chromatographic procedures, which underestimate the percentage of glycosylated hemoglobin in patients with hemoglobin S and C, the electrophoretic method accurately determined the proportion of glycosylated hemoglobin in these hemoglobinopathies. Evidently, electrophoresis on agar gel is an excellent alternative to cation-exchange column-chromatographic methods for glycosylated hemoglobin.