OBJECTIVE： To study the dose-time-effect relationship of Tibetan medicine Rannasangpei in cerebral ischemic- reperfusion injury model rats with intragastric administration. METHODS： The rats were randomly divided into sham operation group （normal saline， 10 mL/kg）， model control group （normal saline， 10 mL/kg）， positive control group （nimodipine， 30      mg/kg）， Rannasangpei different dose groups （0.52， 1.04， 2.08， 4.17， 8.33， 16.67， 33.34， 66.68， 133.36， 266.72 and 533.44    mg/kg）， with 18 rats in each group. Each group was given relevant medicine intragastrically once； 25 min after intragastric administration， cerebral ischemic-reperfusion injury model was established with suture-occluded method in those groups except for sham operation group. 24， 48， 72 h after cerebral ischemia， neuroethology of rats were graded in each group. The rate of cerebral infraction was detected to evaluate the optimal effective time， the optimal dose （Dmax） and maximal effect （the rate of minimum cerebral infraction， Emax） of Ratnasampil at different periods of cerebral ischemia. Dose-time-effect relationship of Rannasangpei dose with the rate of cerebral infraction was fitted with Thermo Kinetica 5.1 software. The area under curve （AUClast） and retention dose （MRTlast） of dose-effect curve were calculated， and detect the levels of SOD and MDA. RESULTS： Compared with sham operation group， the neurobehavior of model group was significantly abnormal （P＜0.05 or P＜0.01）， and the rate of cerebral infarction was significantly increased （P＜0.01）； the level of SOD was decreased significantly （P＜0.01， 48 h）， and the level MDA was increased significantly （P＜0.05， 48 h）. Compared with model control group， there was no significant change in neurobehavioral abnormalities in the nimodipine group （P＞0.05）， and the rate of cerebral infraction was decreased significantly （P＜0.01， 24， 48 h）. The level of SOD in rats were increased significantly （P＜0.01， 48 h）， while the level MDA decreased significantly （P＜0.05， 48 h）. Rannasangpei 2.08-33.34 mg/kg could significantly improved neurobehavioral abnormalities （P＜0.05， 24 h）； 24 h after cerebral ischemia， the rate of cerebral infraction was decreased significantly in Rannasangpei 4.17-133.36 mg/kg group （the lowest is 33.34 mg/kg group， P＜0.05 or P＜0.01）. The level of SOD in rats were increased significantly in 33.34-533.44 mg/kg groups （P＜0.05）. the level MDA was decreased significantly in 0.52-2.08， 8.33， 33.34， 266.72 and 533.44 mg/kg groups （P＜0.05）. Dmax was 33.34 mg/kg， Emax was 3.02％， AUClast was 5 141.76 mg/kg and MRTlast was 329.161 mg/kg. 48 h after cerebral ischemia， the rate of cerebral infraction was decreased significantly in Rannasangpei 2.08-133.36 mg/kg groups （the lowest is 66.68 mg/kg group， P＜0.05 or P＜0.01）， while the level of SOD was increased significantly in 1.04-533.44（except for 4.17）mg/kg groups （P＜0.05）. The level of MDA was decreased significantly in 16.67-66.68， 533.44 mg/kg groups （P＜0.05）， Dmax was 66.68 mg/kg， Emax was 2.13％， AUClast was    5 219.36 mg/kg and MRTlast was 340.521 mg/kg. 72 h after cerebral ischemia， the rate of cerebral infraction and the level of MDA had no significant decreased in Rannasangpei groups （P＞0.05）， and the levels of SOD had no significant increase （except for 0.52 mg/kg group， P＞0.05）. CONCLUSIONS： The optimal effective time of Rannasangpei for the treatment of cerebral ischemia-reperfusion injury in rats is 48 h， and the Dmax is 66.68 mg/kg. The improvement mechanism may be related to increase the level of SOD and decrease the level of MDA.