&NA;Treatment of gliomas remains disappointing in spite of a great number of experimental biological data and of randomized therapeutic studies. This could be partly explained by the inefficiency of our conventional methods to assess the regional metabolism of these tumors. The use of positron emission tomography (PET) brings encouraging possibilities in this field. We report our preliminary experience of measuring regional cerebral methionine uptake with PET after intravenous injection of [11C]l‐methionine. Twenty‐two patients with histologically confirmed gliomas were studied. An ECAT II positron emission tomograph was used for scanning. The position of the plane was chosen to include a major section of the tumor in the reconstructed brain slice. The protocol required a two‐step examination: 1) after injection of 15 to 25 mCi of [11C]l‐methionine, 12 scans were performed over a period of 46 minutes; and 2) 18 hours later, regional cerebral blood volume was measured in the same slice after intravenous injection of 2 to 4 mCi of68GaCl3. The tumoral region of interest was determined as being the area of maximum activity. For each patient we calculated the ratio, R, between the activity in this tumor region of interest and the activity in the contralateral healthy symmetric region of interest which was used as an “internal standard” for the same patient. We correlated the ratio R with the histological grading. In 22 patients, mean values of R were calculated for each tumor: Grade II (n = 5): R = 1.04 ± 0.27; Grade III (n = 5): R = 1.68 ± 0.22; and Grade IV (n = 12): R= 2.33 ± 0.86. Correlations between hypermetabolism and grading are highly significant for Grades II/III (P= 0.0045) and II/IV (P= 0.00019), but not for Grades II/IV (P= 0.3141). We assessed the effects of radiotherapy: in 3 glioblastomas studied before and after radiotherapy, a significant decrease of R was observed: from 2.84 ± 1.43 to 1.35 ± 0.05. In one Grade II glioma, there was no change. Last, we assessed the early effects of intra‐arterial chemotherapy: 8 patients were studied before and after the first course of treatment with an intracarotid injection of 1,3‐bis‐(2‐chloroethyl)‐l‐nitrosourea (BCNU). Early modifications of R were correlated with clinical evolution after completion of treatment (intracarotid BCNU every 6 weeks). Despite a patient in whom a dramatic decrease of R was followed by a long survival with normalization of the computed tomographic scan, no significant correlation could be displayed between modifications in R and the length of survival. This approach will have to be developed in future studies. (Neurosurgery25:720‐728, 1989)