Missing longitudinal strength cannot reappear at very high energy

The longitudinal sum rule, computed from recent measurements of quasielastic longitudinal structure functions at high momentum transfers, is lower than the theoretical prediction by about a factor 2. Here it is shown, by comparison between the theoretical measured values of the ratio of the $f$ sum rule to the longitudinal sum rule, that any substantial fraction of missing longitudinal strength must lie at energies low enough to have been detected. This analysis is independent of any systematic experimental errors of overall scale.