Laser Doppler fluxmetry may improve the monitoring of cortical blood flow in neurosurgical patients. So far, however, the variability of laser Doppler readings found in the cerebral cortex has prevented a consequent usage of the technique in clinical practice. The current report analyzes the regional variability of laser Doppler readings from experimental animals. Typical frequency histograms of observed flow values display non-Gaussian distributions. A simulation technique is used to assess the number of measuring sites required for valid estimates of regional cortical flow. From a total of 990 local flow measurements from 45 rabbits random samples of sizes between 5 and 100 were repeatedly drawn to estimate the variability of median flow thus determined. The study underlines that small sample sizes below n = 20 are associated with a large variability (95th percentile: +38%) which decreases in a biphasic manner. Sample sizes above n = 25 are necessary to obtain more reliable information on regional cerebral blood flow: the 95th percentiles remain between 24 and 28% up to n = 45, and are still at 15% with n = 99.