Abstract
It is pointed out that the observation of plasmon energy-loss peaks and determination of peak breadths in inelastic-electron-scattering experiments cannot be used to distinguish metallic or insulating materials. The recent claim, based on inelastic-electron-scattering data, by MacRae, Müller, Lander, Morrison, and Phillips that the initial (low-density) second layer of cesium adsorbed on a tungsten (100) surface is in an insulating state is considered not proved.