Objective: It has been argued that semantic verbal fluency may be a promising cognitive endophenotype in schizophrenia and possibly a trait-like cognitive measure of the illness. The present study examined the stability of verbal fluency performance in schizophrenia over-time. It was hypothesized that verbal fluency performance would be stable over 1 year. Method: Data was extracted from a larger study that followed 53 out-patients (16 female; mean age = 39.2 years) diagnosed with schizophrenia or schizoaffective disorder at baseline, 6, and 12 months. At each testing interval, phonemic (i.e., F, A, S) and semantic (i.e., animals) fluency tasks were administered as per Benton (1968) and Butters et al. (1987), respectively. Traditional scoring (i.e., total correct words in one minute/trial) along with phonemic clustering, semantic clustering, phonemic switching, and semantic switching component scores were derived and compared across time points. Results: Linear regression analyses were performed to examine clustering and switching patterns for semantic and phonemic fluency longitudinally. No significant differences (p > 0.05) were found over time. Furthermore, no significant changes in performance were observed over time for both semantic and phonemic mean cluster size or number of switches. Moreover, partial correlation analysis revealed no statistical relationship between phonemic fluency and years of illness. Conclusion(s): The results support the hypothesis in that the component verbal fluency performance across the three time points was stable. Accordingly, the results provide further substantiation to the assertion of verbal fluency as a possible cognitive endophenotype in schizophrenia.