The past year's work on brain imaging in schizophrenia is reviewed. Structural studies still dominate. There is now a greater emphasis on methodology, with simple area ratios giving way to assessment of approximate volumes. Multi-structure morphometry continues to expand with particular focus on temporal lobe grey matter components and basal ganglia. There remains no clear concensus on the pattern of abnormality associated with the schizophrenic brain, the clinical correlates of structural change or the theoretical implications. There is some concensus that ventricular enlargement may antedate the onset of illness in some patients at least, though the question of progression remains open. The first reports using magnetic resonance spectroscopy have now been published, offering an exciting new dimension to the investigation of brain chemistry in health and disease.