The authors report five cases of chronic epidural haematoma diagnosed and treated surgically in the last four years. Age, source and rate of bleeding, clot volume, adherence to the dura, individual tolerance and location of the haematoma have all been incriminated in the past to explain the chronic nature of these lesions. The location of the clot seems to us the most important causative factor. We emphasize the value of CT in the diagnosis of these lesions.