The technique and results of per-operative angiography during surgery for intracranial vascular lesions in 50 patients are described. Carotid territory lesions are shown by percutaneous cannulation of the common carotid artery, after craniotomy and during the progress of the operation. Posterior fossa or extensive lesions are shown by catheterisation of the vertebral or subclavian arteries, before craniotomy. Screening is used for positioning, employing a mobile image intensifier and closed circuit television. An initial angiogram is taken routinely after surgical exposure but before obliteration of the lesion. Check angiograms are taken following surgical treatment, to confirm complete obliteration of the lesion and exclude inadvertent vascular occlusion. Post-operative angiography is employed, where possible, to assess the efficacy of per-operative angiography. The technique was successful on 49 occasions. Film quality compared favourably with pre-operative and post-operative angiography. No significant complication occurred. Eleven arterio-venous malformations were completely or partly removed using angiography during the progress of dissection and for final confirmation. The findings were confirmed by post-operative angiography. Of 38 aneurysms investigated, 32 showed complete obliteration at operation and 4 deliberate but incomplete obliteration. Two aneurysms were inadvertently incompletely obliterated, one of which showed complete obliteration following adjustment of the surgical clip, 67% of aneurysms showed no major branch occlusion following surgery, and a further 11% showed known and deliberate occlusion. 22% showed inadvertent occlusion of a major branch. The clip was adjusted on 5 occasions, with relief of occlusion in 3, but no relief in 2. Post-operative angiography confirmed the per-operative angiogram findings in all but 2 cases. The technique was also used during surgical treatment of one carotico-cavernous fistula. Per-operative angiography was helpful in confirming surgical obliteration, showing the effect of clip adjustment and excluding major arterial occlusion at the time of operation, and reducing the need for routine post-operative angiography or further craniotomy. The technique and results of per-operative angiography during surgery for intracranial vascular lesions in 50 patients are described. Carotid territory lesions are shown by percutaneous cannulation of the common carotid artery, after craniotomy and during the progress of the operation. Posterior fossa or extensive lesions are shown by catheterisation of the vertebral or subclavian arteries, before craniotomy. Screening is used for positioning, employing a mobile image intensifier and closed circuit television. An initial angiogram is taken routinely after surgical exposure but before obliteration of the lesion. Check angiograms are taken following surgical treatment, to confirm complete obliteration of the lesion and exclude inadvertent vascular occlusion. Post-operative angiography is employed, where possible, to assess the efficacy of per-operative angiography. The technique was successful on 49 occasions. Film quality compared favourably with pre-operative and post-operative angiography. No significant complication occurred. Eleven arterio-venous malformations were completely or partly removed using angiography during the progress of dissection and for final confirmation. The findings were confirmed by post-operative angiography. Of 38 aneurysms investigated, 32 showed complete obliteration at operation and 4 deliberate but incomplete obliteration. Two aneurysms were inadvertently incompletely obliterated, one of which showed complete obliteration following adjustment of the surgical clip, 67% of aneurysms showed no major branch occlusion following surgery, and a further 11% showed known and deliberate occlusion. 22% showed inadvertent occlusion of a major branch. The clip was adjusted on 5 occasions, with relief of occlusion in 3, but no relief in 2. Post-operative angiography confirmed the per-operative angiogram findings in all but 2 cases. The technique was also used during surgical treatment of one carotico-cavernous fistula. Per-operative angiography was helpful in confirming surgical obliteration, showing the effect of clip adjustment and excluding major arterial occlusion at the time of operation, and reducing the need for routine post-operative angiography or further craniotomy.