The middle cerebral artery was occluded at its origin via subtemporal approach by microsurgical technique in 24 dogs. In 8 of these 24 dogs, end-to-side anastomosis between the maxillary artery and a branch of the middle cerebral artery (MA-MCA anastomosis) was made 4 hours after MCA occlusion. In 5 dogs, MA-MCA anastomosis was performed under microscopic control 3 weeks after MCA occlusion. Remaining 11 dogs without shunt operation were used as control animals. All the animals were clinically observed every day until sacrifice. In the control animals, common carotid angiography was performed between the 2nd and the 5th postoperative weeks. The treated animals were studied by selective external carotid angiography 2 weeks after MA-MCA anastomosis. After sacrifice, transcarotid perfusion with 10% formalin solution was carried out and the brain was carefully removed. Each brain was additionally fixed in 10% formalin, sectioned, stained and examined pathologically. Clinical evaluation in all the control animals showed mild to severe neurological deficits or death. On the other hand, the animals with patent bypass in acute stage demonstrated no neurological deficits. Gross and microscopic evaluation of the brains showed that the permanent occlusion produced a medium or large-sized infarct in the occluded MCA territory, and the patent prompt bypass usually caused no or only microscopic infarct. In the patent delayed bypass, the size of infarct seemed smaller than that in the untreated animals. No hemorrhagic infarct was found in treated animals with either prompt or delayed bypass. In general, it seemed that the animals with patent bypass fared better than untreated animals both clinically and pathologically. The experimental data suggest that reestablishment of blood flow by extra-intracranial anastomosis, particularly within 4 hours after MCA occlusion, may lead to a significant restoration of neurological function without pathological damage of the brain.