A dynamic model of a nonisothermal continuous stirred tank polymerization reactor is developed to predict product molecular weight distribution parameters. Deliberate oscillations in operating variables were programmed into the model and the resulting time-averaged reactor performance indices compared with that for optimum steady-state operation. Results indicate that periodic operation does influence polymerization reactor performance, particularly with regard to time-averaged molecular weight polydispersity. Significant resonant behavior was encountered with molecular weight properties at low frequencies.