We measured regional cerebral blood flow with the xenon 133 inhalation technique in 41 patients with major depressive disorder and 40 matched, normal controls during an eyes-closed, resting condition. The depressed group had a marked reduction in global cortical blood flow. To examine topographic abnormalities, traditional multivariate analyses were applied, as well as a new scaled subprofile model developed to identify abnormal functional neural networks in clinical samples. Both approaches indicated that the depressed sample had an abnormality in topographic distribution of blood flow, in addition to the global deficit. The scaled subprofile model identified the topographic abnormality as being due to flow reduction in the depressed patients in selective frontal, central, superior temporal, and anterior parietal regions. This pattern may reflect dysfunction in the parallel distributed cortical network involving frontal and temporoparietal polymodal association areas. The extent of this topographic abnormality, as revealed by the scaled subprofile model, was associated with both patient age and severity of depressive symptoms.