Seventy cases of intracerebral hematomas were analysed, forty-six of spontaneous origin and remainder of traumatic one, which had been experienced in Seoul National University Hospital between March, 1970 and June, 1975. The results were as follows: A. Among forth-six cases of spontaneous intracerebral hematomas, twenty-six cases(52%) were primary intracerebral hematomas, normotensive or hypertensive, seven aneurismal ruptures(15%), seven arteriovenous malformations(15%), two cerebral rete mirabilies, one tumor bleeding, one leukemia, one idiopathic thrombocytopenic and one due to other blood dyscrasia. 1. Among twenty-six cases of primary intracerebral hematomas, angiography was performed in all and fourteen cases were operated. The sites of hemorrhage were as follows: a. Putaminal hemorrhage:17 cases(65%)(two cases associated with intraventricular hematoma). b. Thalamic hemorrhage:5 cases(19%)(two cases associated with intraventricular hematoma). c. Subcortical hemorrhage:4 cases(16%). 2. Overall mortality among twenty-six cases was forty-six per cent;operative mortality was forty-three per cent, and non-operative, fifty per cent. 3. Age, mode of onset, level of consciousness at the time of surgery and the site of hematoma were all important factors for surgical result. Surgical intervention seemed to be most favorable in patients who had verbal communication with subcortical or localized putaminal hematoma. 4. Normotensive group was worse than hypertensive one with conservative treatment, but the former was more benefited than the latter with surgery. 5. Intracerebral hematomas due to aneurismal rupture, arteriovenous malformation and others were also influenced with above mentioned factors, but the nature of underlined disease was of prime importance. B. Among twenty-four operated cases of traumatic intracerebral hematomas: 1. Five cases(21%) showed classical lucid interval with variable duration from thirty minutes to forty-eight hours. 2. Temporal lobe was most frequently involved(63%). 3. Lateral blow was predominant than axial low, and contrecoup than coup lesion. 4. Overall mortality was twenty-nine percent. 5. Age, lucid interval, preoperative level of consciousness and associated cerebral contusion or subdural hematoma were important factors for prognosis, but the preoperative level of consciousness was of prime importance.