The author studied the diagnostic usefulness of topography of flash visual evoked potential (FVEP) in evaluation of the patients with cerebrovascular diseases (CVDs) proven by brain CT scans. The patient group of 21 cases were consist of 10 cerebral infarctions and 11 intracerebral hemorrhages, that occurred at various sittes; 2 frontal lesions, 8 temporopariental lesions, 2 occipital lesions, and 8 basal ganglia lesions. As a control group, topographic studies of FVEP were done in 25 patients with no clinical eviddences of central nervous system diorders and visual symptoms. Topographic studies of FVEP with 12 recording electrodes, using Topography system 700 (San-ei), were done by binocular flashlight stimulation at the rate of 2/sec with the eyes closed. Topographic studies of FVEP were evaluated visually and the asymmetry of topography was considered as abnormal. Three types of topography were used; spatial-temporal mapping up to 300 msec of latency, latency mapping of wave II and VII (Cigenek, 1961), and amplitude mapping of wave II and VII. In both control and patient group, FVEP recorded at O1 and O2 were also obtained. The results were as follows: 1. In control group, absolute latencies and amplitudes of wave II and VII were much variable in FVEP recorded at O1and O2. Interhemispheric differences of amplitudes of wave II and VII were also much variable, but those of latencies were the least variable parameters; 1.89+1.24 msec in wave II and 2.65+3.90 msec in wave VII. Interhemispheric differences of latencies and amplitudes in patient group were interpreted as normal within the mean value plus 2 standard deviation of those of control group. 2. In patient group, abnormal FVEP recorded at O1 and O2 were observed only in 11 cases, among which 8 cases (38%) were consistent in lateralization of supratentorial lesions on brain CT scans. 3. Spatial-temporal mapping of FVEP, even in control group, was variable at each latencies, showing shifting asymimetries sidde to side and only suggested the tendency of greater amplitudes in right hemisphere and sequential transition of evoked responses. 4. In patient group, 17 cases showed abnormal latency mapping of wave II an/or wave VII, among which 13 (62%) were consistent in lateralization of supratentorial lesions. This relatively high concordance rate to brain CT findings was thought to be resulted from low variability of interhemispheric differences of latencies, and also suggested that latency mapping is more reliable method than other 2 types of topography of FVEP. 5. Amplitude mapping of wave II and/or wave VII showed 38% (8 cases) of concordance rate to brain CT findings in lateralization of supratentorial lesion. In conclusion, it is thought that topography of FVEP is a useful diagnostic test in evaluation of CVDs, and is also suggested that topography of FVEP can be used in evaluation of various supratentorial cerebral lesions functionally without such sophisticated criteria as in conventional FVEP interpretation.