Dynamic susceptibility contrast magnetic resonance imaging (DSC MRI) can be used to generate high resolution maps of cerebral blood volume (CBV). To determine the test-retest reliability, CBV was measured in eight volunteers on two occasions, separated by 4 weeks. The mean ratio (scan 2/scan 1) for 72 cortical regions of interest (ROIs) was 1.03, with a coefficient of variation of 14%. The correlation between the first and second scans was 0.73 (p < 0.0001; 95%). In five hand-drawn ROIs, the mean ratio was 1.08, with a coefficient of variation of 12%. The correlation between scans was 0.84 (p < 0.0001). The data presented here support the hypothesis that DSC MRI CBV mapping has acceptable test-retest reliability.