Many geologists have recognized that valleys, which have once been occupied by large glaciers, exhibit peculiarities which can only be ascribed to glacial erosion modify in g the earlier water-worn form of the land-surface. Indeed, anyone who allows ice-action to be locally a factor of importance in erosion, must expect that the resulting topographical forms will differ in some measure from those proper to water-erosion. Not only does ice,invirtue of its physical properties, obey different laws as regards pressure and flow, but, what is of equal importance, it fills the whole valley instead of following certain narrow channels. We may hope that, just as Gilbert and others have worked out the principles of watererosion, so the much more difficult problem of ice-erosion may some day be reduced to analysis. An attempt in this direction has been made by McGee in a paper on “Glacial Canons,” which, however, I cite here rather for its brief summary of the observed characteristics of glaciated valleys.