The response of flat, thin, parallel, metal fuel elements to the loads imposed by the flow of coolant through reactor core passages is examined for the existence of plate divergence at velocities above a “critical” value. It is shown that small modifications of the simplifying assumptions used in the analysis produce a great difference in the conclusions regarding the possibility of divergence and the interpretation of the “critical” coolant velocity. The basic assumptions are the same as those of Miller (1), except that fluid inertia effects are included in the analysis of periodically supported plates. Although agreement exists between the results of the dynamic model of Section I and that of “neutral equilibrium” used by Miller, the additional consideration of fluid inertia leads to a different interpretation of “critical” velocity for periodically supported plates treated in Section II.