The development of transoesophageal echocardiography as a relatively non-invasive means of imaging the aortic arch has led to the discovery of this region as a potential cause of stroke. There is in vivo and autopsy evidence that aortic arch atheroma is an independent risk factor for ischaemic stroke and that the risk is of the same order of magnitude as the well established risk factors such as hypertension, carotid stenosis and atrial fibrillation. The most common mechanism of stroke development is probably by virtue of embolic debris lodging within the cerebral circulation. Transoesophageal echocardiography as a means of detecting aortic arch atheroma is indicated in patients with embolic stroke in whom no obvious source of embolism from large arteries or heart can be found. While there is no consensus on how best to manage patients with aortic arch atheroma in a primary or secondary prevention setting, antiplatelet therapy, anticoagulants, thrombolytic agents and surgery have been suggested or used. Further improvements in non-invasive imaging of the aortic arch are likely to lead to a better understanding of its place in the genesis of ischaemic stroke and pave the way for trials of therapy.