Since every component of the evoked responses are considered to be related to topographically specific neural structures, it is possible that the location and severity of brain dysfunction could be accurately defined by careful analysis of evoked responses. The main objective of this experiment was to replicate some of the mechanisms involved in human brain injuries in cat and observe the effect of focal hemispheric brain injury on regional cerebral blood flow(rCBF) and somatosensory evoked potential(SEP) and to evaluate the effects of mannitol on them. Thirty adult cats weighing 2.5 to 4.2kg, were used in this study. The animals were divided into 3 groups of 10 cats each:(1) mild injury, (2) severe injury and (3) mannitol treated severe injury group. A mild injury was produced by moving the drill along a predetermined pathway through the right parietal hole at 50 cycle per minute for 2 seconds and a severe injury was produced in a similar fashion at 200 cycle per minute for 3 seconds. A mannitol treated group was produced in a same method as the severe injury group. The rCBF and SEP measurements were performed immediately after injury in each animal, at 30 minutes, 60 minutes and the final flow at 90 minutes by Pasztor(1973) hydrogen clearance technique. The results obtained were as follows. 1) After focal cerebral hemispheric injury, there were rapid rise in intracranial pressure, bradycardia, changes in blood pressure and marked alteration in respiration which are neurogenically mediated. 2) Normal control flows(rCBF, ml/100g/min) were 30.7+/-5.9 in right frontal, 35.2+/-6.7 in right parietal, 27.9+/-6.8 in left frontal, and 35.2+/-7.3 in left parietal lobes. 3) Sequential changes of the rCBF after focal cereral hemispheric injury were as follows. (1) Mild focal hemispheric injury resulted in a reduction of flow to 30% of control flow(RF:18.8+/-3.7, RP:25.0+/-7.8ml/100g/min) at injury resulted in a reduction of flow to 30% of control folw(RF:18.8+/-3.7, RP:25.0+/-7.8ml/100g/min) at injury site after immediate injury. (2) Severe focal hemispheric injury resulted in a reduction of flow to 50% of control flow(RF:20.4+/-10.9, RP:18.8+/-7.6ml/100g/min) at injury site after immediate injury. (3) Mannitol-treated severe injury resulted also in a reduction of flow to 50% of control flow at the injury site after immediate injury, but at 90 minutes the flow was 75% of the control flow. 4) A close correlation was found between cortical-evoked potentials and flow, suggesting a threshold relationship both on injury and non-injury areas. (1) The SEP was present shortly after injury though markedly altered in shape and the early components(No, N1) of the SEP were suppressed first. (2) It was also noted that the amplitude of the SEP was much smaller, perhaps due to direct ijury on the injured area. (3) The SEP disappeared if the rCBF in either hemisphere fell below 15ml/100gm/min. 5) It might be inferred from these results that adequate flow was vital for the preservation and return of electrical activity following brain injury.