In the middle of the 19th century physicians performing autopsy could recognise cerebral softening through arterial occlusion, and distinguish it from intracerebral haemorrhage (apoplexy). During life diagnostic possibilities were limited, until the introduction of arteriography in 1927. A low blood pressure and embolism from the heart were considered possible causes of brain softening. After the turn of the century, when blood pressure could be measured, cerebral haemorrhage was generally attributed to arterial hypertension. The anatomical substrate of brain haemorrhage was controversial: bursting of a microaneurysm or a sclerotic artery, and primary infarction followed by increased permeability of vessel walls were all proposed as intermediate events. The subject of subarachnoid haemorrhage was first mentioned in the 1930s. Migraine and toxic factors were initially implicated as causes, later especially aneurysms. In the 19th century treatment of stroke would often involve blood-letting and application of dressings. Lumbar puncture was used in the 1920s and 30s for subarachnoid or intracerebral haemorrhages. In 1950 the first report about surgical occlusion of cerebral aneurysms appeared in the Journal.