This experimental vasospasm model was made by giving the autologous blood into the cisterna magna. Later Nimodipine (ND) was administered by intracisternal or subcutaneous (s.c.) way in defined times. After that, in the light microscopic examination, the wall thickness (W), lumen (L) and diameter (D) of basilar artery were measured and diameter/wall thickness (D/W), diameter/lumen (D/L) ratios were studied and compared with control group with SAH.In the first group, intracisternal ND was administered immediately after experimental SAH, ND doses were10 pg/kg (n: 5), 20 pg/kg (n:4), and 50 pg/kg (n:5). There was no significant difference in the measurements between study group and control (n:9). In the second group which received s.c. 100 pg/kg t.i.d. for 72 hours, the wall thickness significantly decreased when compared to control group (p<0.01) but the other parameters did not change.The double-hemorrhage group which received samedose (100 pg/kg) of ND showed significant increase in the D/W ratio when compared to matching control group (p<0.01). But there was no change in other parameters.Electron microscopy was made only in rats with SAH taking 100 pg/kg t.i.d. ND for 72 hours. Ultrastructural changes were endothelial changes, subendothelial blebs, corrugation of lamina elastica interna, endothelial denudation and intracytoplasmic vacuoles in both control group with SAH and ND receiving ones. Intracytoplasmic smooth-muscle vacuoles were less prominent in the rats receiving ND.These findings may indicate that ND shows its effectiveness on experimental vasospasm by decreasing the wall thickness and preventing some morphologic and ultrastructural changes.