In the last 40 years, blood flow and metabolism in the brain have been measured by many methods of varying resolution and reliability. Modern methods have helped to answer some basic pathophysiologic questions, in particular disproving ongoing global ischemia as a cause for dementia. But much of this physiologic work is confounded by swiftly changing clinical and pathologic understandings of ischemic and other forms of dementia: the field remains limited as much by unresolved clinical questions as by technologic feasibility.