We used 133Xe dynamic single-photon emission computed tomography (DSPECT) to measure the resting cerebellar blood flow in 17 neuroleptic-free schizophrenic and schizophreniform patients and 13 normal control subjects. A subset of these subjects (11 patients and 7 control subjects) additionally underwent activation studies during the Wisconsin Card Sorting (WCS) and Number Matching (NM) tests. Baseline relative cerebellar blood flow was significantly lower in older patients than in age-matched control subjects. For absolute cerebellar flow, there was a significant difference between patients and control subjects in the overall activation response (patients: NM 13.4% increase, WCS 15.7% increase; control subjects: NM 3.1% decrease, WCS 0.0% change). This difference was more pronounced in older subjects. Cerebral blood flow significantly increased during NM (patients: 21.3% increase, control subjects: 6.5% increase) and WCS (patients: 16.5% increase, control subjects: 9.7% increase). The difference in the magnitude of cerebral NM activation between schizophrenic patients and control subjects, although not statistically significant, may call into question the appropriateness of using NM as a control task in schizophrenic patients. Finally, we found no differences between the effects of WCS and NM on cerebellar or cerebral blood flow. Because of the small number of subjects in each group, the results of this study should be interpreted cautiously.