Schizophrenia is related to physiological and anatomical disorders in the brain (Andreasen, Flaum, Swayze, Tyrell, & Arndt, 1990; Andreasen, 1985; Crow, 1980). Physiological changes are more closely associated with positive symptoms, while anatomical changes are more closely linked to negative symptoms. Physiological alterations thought to contribute to schizophrenia include increased levels of subcortical dopamine (i.e., the dopamine hypothesis proposed by Matthysse [1977]) and reductions in glucose utilization and central amine activity (Rao & Moller, 1994). A future column will address the neurophysiology and neurochemistry of schizophrenia. Many anatomical changes occurring in schizophrenia are microscopic. An exception is the ventricular enlargement (or increased ventricular brain ratios [VBRs]) found in patients with chronic schizophrenia (Andreasen, 1985; Flaum et al., 1995; Lewine, Hudgins, Brown, Caudle, Risch, 1985; Rao & Moller, 1994; Vita et al., 1994, 1995). This observation first grew out of postmortem findings that suggested a significant increase in ventricular size among schizophrenic patients. CT and now MRI studies confirm these earlier reports. Ventricular enlargement has two etiologies: 1) ventricles expand like a balloon due to increased levels of cerebrospinal fluid and 2) ventricles fill a vacuum when periventricular tissue dies, atrophies, or does not fully develop. The former occurs in hydrocephalus while the latter is prominent in Alzheimer's disease (cell death and atrophy) and in schizophrenia (developmental lag). It is important to note that unlike Alzheimer's disease, the ventricular enlargement found among schizophrenic patients is not thought to be associated with neurodegeneration (Marsh, Suddath, Higgins, & Weinberger, 1994; Roberts, Leigh, & Weinberger, 1993; Sponheim, Iacono, & Beiser, 1991). The photograph on the left in Figure 1 is that of a sized ventricle in a person not afflicted with schizophrenia. The photograph on the right shows the enlarged ventricles of a person suffering from schizophrenia. The difference in ventricular size is readily apparent in these photographs but not all cases are so compelling because approximately 50 percent of schizophrenia patients fall within the range of control or normal subjects (Cannon & Marco, 1994). The search for a gross anatomical marker of schizophrenia is further compromised by ventricular variability; that is, ventricle size varies among nonschizophrenic individuals as well. Historically then, finding acceptable control ventricles has been a major challenge to schizophrenia researchers. [Figure 1 ILLUSTRATION OMITTED] The search for adequate controls has led researchers to what might be the control-the monozygotic, unaffected twin of patients with schizophrenia. As a group, first-degree relatives have a high (70% to 100%) sensitivity ratio for ventricular enlargement (Cannon & Marco, 1994), and it is consistent with genetic theory to assume monozygotic twins occupy the higher end of this range. As can be seen in Figure 2, the affected twin has a slightly larger ventricle than does the unaffected twin. However, without the perfect control, the schizo-phrenic twin would not present with a discernably larger ventricle. In fact, when compared to the affected patient in Figure 1, the affected twin in Figure 2 would appear to be well within the parameters of ventricular size. Compositely, Figures 1 and 2 suggest a continuous distribution of pathology (from seemingly slight to gross pathoanatomical changes) among individuals with schizophrenia. Hence, even small losses of periventricular brain tissue can have profound impact on mental function. [Figure 2 ILLUSTRATION OMITTED] Other anatomical changes are microscopic but aggregately can manifest in such proportion as to be visible to the unaided eye. A majority of studies find schizophrenia patients, as a group, have significantly smaller limbic structures (Marsh et al. …