Serum glucose measurements were performed on paired specimens from patients with uremia, before and after dialysis, to evaluate the automated glucose oxidase, ortho-toluidine, ferricyanide, and copper reduction methods. All methods were found to yield results statistically different from those of the glucose oxidase method whether glucose was measured before or after dialysis. Before dialysis, the mean difference between the ortho-toluidine and glucose oxidase method results was −9.0 mg. per dl., and after dialysis it was −2.8 mg. per dl.; other methods differed more from glucose oxidase. The glucose oxidase method yielded 2 mg. per dl. or less for uremic sera depleted of glucose by yeast; other methods were less specific for glucose. High uric acid concentrations depressed ortho-toluidine values. Fructose caused slight elevations in ortho-toluidine and large elevations in ferricyanide values. In view of the above findings, the glucose oxidase method is the method of choice for determining glucose values in patients with uremia. Results from 379 nonuremic sera determined simultaneously by the ortho-toluidine and ferricyanide methods were also compared. In nonuremic patients, serum fructose elevation after glucose ingestion partially accounts for the differences between the results of the ortho-toluidine and the ferricyanide methods observed.