The purpose of this study was to investigate the effects of sevoflurane concentration on canine visual evoked potentials with pattern stimulation (P-VEPs). Six clinically normal laboratory-beagle dogs were used. The minimum alveolar concentration (MAC) of sevoflurane was detected from all subjects by tail clamp method. The refractive power of the right eyes of all subjects was corrected to -2 diopters after skiascopy. For P-VEP recording, the recording and reference electrode were positioned at inion and nasion, respectively, and the earth electrode was positioned on the inner surface. To grasp the state of CNS suppression objectively, the bispectral index (BIS) value was used. The stimulus pattern size and distance for VEP recording were constant, 50.3 arc-min and 50 cm, respectively. P-VEPs and BIS values were recorded under sevoflurane in oxygen inhalational anesthesia at 0.5, 1.0, 1.5, 2.0, 2.5 and 2.75 sevoflurane MAC. For analysis of P-VEP, the P100 implicit time and N75-P100 amplitude were estimated. P-VEPs were detected at 0.5 to 1.5 MAC in all dogs, and disappeared at 2.0 MAC in four dogs and at 2.5 and 2.75 MAC in one dog each. The BIS value decreased with increasing sevoflurane MAC, and burst suppression began to appear from 1.5 MAC. There was no significant change in P100 implicit time and N75-P100 amplitude with any concentration of sevoflurane. At concentrations around 1.5 MAC, which are used routinely to immobilize dogs, sevoflurane showed no effect on P-VEP.