In order to know the influence of intravenous injection of 50 per cent glucose solutions on the sugar contents of the cerebrospinal fluid, after estimating the error of the laboratory method the authors studied in detail such effects in three groups of patients: group 1, for controlling the results (3 cases) ; group 2, including 5 patients with normal and 5 with abnormal (presenting meningitic and/or parenchymatous changes) fluids; group 3, including 10 patients with normal and 10 with abnormal fluids. In group 1 no glucose was injected; in group 2, 10 gm were administered; in group 3, 50 gm were injected. The method of study consisted essentially of: a) simultaneous withdrawal of blood and fluid, in fasting conditions, for determination of the initial levels; b) intravenous injection of 50 percent glucose solution; c) 15, 30, 45, 60, 120, 180 and 240 minutes later, other samples of blood and fluid were withdrawn from each patient. The results were submitted to statistical analysis. The oscilations of the average levels obtained in groups 1 and 2 were comparable to those due to the experimental error. The differences between the initial mean and the average contents 15 and 30 minutes after the injection of 10 gm glucose were not statistically significant; in the patients receiving 50 gm, only after 30 minutes the average content differed significantly from the initial mean. The linear regression equations relating time to the blood and cerebrospinal fluid sugar contents were calculated. The ratios between fluid and blood sugar contents were analysed. In 10 cases of group 3 the blood and fluid sugar contents before and 24 hours after the glucose injection were compared. The authors draw the following conclusions: 1. The intravenous injection of 10 gm glucose does not influence significantly the cerebrospinal fluid sugar content. 2. After the intravenous injection of 50 gm glucose it is observed, during the first hour, a uniform increase of the fluid sugar contents, mostly in the patients with abnormal fluid, probably owing to greater permeability of the blood-brain barrier. 3. The ratios between cerebrospinal fluid and blood sugar contents (CSF/B) showed progressive increase in the patients with abnormal fluid 15 to 120 minutes after the injection of 10 gm glucose. 4. The CSF/B ratios showed progressive increase both in the patients with normal and abnormal fluid, 15 to 120 minutes after the injection of 50 gm glucose; the difference between the regression coefficientes of these two groups was not significant. 5. The average initial ratio between the cerebrospinal fluid and the blood sugar contents, in 29 patients, was 0.617 ± 0.129. 6. The increase of the cerebrospinal fluid sugar content after the intravenous injection of 50 gm glucose lasted 4 hours at least. 7. There was no significant difference between the blood and cerebrospinal fluid sugar contents before and 24 hours after the intravenous injection of 50 gm glucose.