The genetically dystonic rat is insensitive to the tremorogenic effects of harmaline. This behavioral deficit has been linked to a defect in the olivocerebellar pathway, since few Purkinje cells of dystonic rats show a normal increase in rhythmic complex spike activity following harmaline. In normal rats, the Purkinje cell response to harmaline and tremor are initiated by a rhythmic increase in neuronal firing in the caudal inferior olive. The present single unit recording study was conducted, therefore, to determine if the inferior olive of the dystonic rat is activated by harmaline. Olivary unit responses to harmaline were also examined in normal rats made tolerant to harmaline tremor. These rats are behaviorally insensitive to harmaline and also fail to display rhythmic complex spike activity but do not have the motor deficits of the mutant rats. The spontaneous firing rate of neurons in the caudal and rostral inferior olive of the dystonic rat was significantly slower than that of phenotypically normal littermates. Despite this, all cells recorded in the caudal portion of the medial accessory olive of both dystonic and normal rats showed increased rhythmic activity following harmaline injection. Thus, the failure of the mutants to show harmaline tremor is not due to a failure of the drug to activate cells in the olive. Rather, the data suggest a defect in the subsequent transmission of this information. Unlike the control and dystonic rats, harmaline-tolerant rats failed to show sustained rhythmic activity in the inferior olive. These findings suggest that chronic treatment with harmaline may interfere with harmaline tremor at the level of the inferior olive.