Regional rates of apparent glucose utilization (GU) in metastatic Walker 256 (WL-256) brain tumors produced by the intracarotid injection of WL-256 tumor cells in rats were measured using 14C-deoxyglucose and quantitative autoradiography. Apparent glucose utilization was uniform within individual small and medium size tumors without necrosis, varied considerably among different tumors within this group, and did not correlate with tumor size or location. High values of GU in medium and large-size tumors correlated with viable-appearing tissue in contrast to necrotic tissue and were always 1.3 to 3 times higher than that of adjacent and contralateral nontumorous brain. The apparent net extraction of glucose (En*) in viable tumor regions was estimated to be several fold higher than that in remote brain tissue; analysis of this data for medium and large tumors indicates that the calculated values of GU and En* overestimate the actual rates of utilization and net extraction of glucose. Local cerebral glucose utilization (LCGU) was higher than normal adjacent to small tumors and lower than normal adjacent to larger tumors. The LCGU in many gray-matter structures remote from the intracerebral tumors was reduced and roughly proportional to the metastatic tumor burden. The comparatively high uptake of 2-deoxyglucose by viable tumor cells has diagnostic value and suggests that appropriate glucose analogues could be developed to produce a tumor-selective inhibition of glycolysis and tumoricidal effect.