polishing, to the surfaces which it had formerly prepared by rasping and filing.
The calculations of the effects that have been produced by glacier agency, which are based on the assumption that the amount of material removed is the same from one year to another, are necessarily fallacious. There are not, moreover, any data from which the amount of work can be calculated that glaciers perform in any given time; but there are indications in that direction, and, so far as they go, they seem to point to the conclusion that the effects which they have produced, in the way of making hollows, are much less important than many suppose.
6. If I were asked whether the action of glacier upon rocks should be considered as chiefly destructive or conservative, I should answer, without hesitation, principally as conservative. It is destructive, certainly, to a limited extent; but, like a mason who dresses a column that is to be afterwards polished, the glacier removes a small portion of the stone upon which it works, in order that the rest may be more effectually preserved. By obliterating the inequalities of the rock, and, consequently, by reducing the area of the surfaces which are exposed to the atmosphere to a minimum, the glacier, when it retires, leaves the rock in the best possible condition to withstand the attacks of heat, cold, and water.
It has been pointed out, times without number (even by those who are in the habit of accusing glaciers of the most frightful destructiveness), that the polished surfaces which they leave behind them seem to be imperishable. All who know are agreed that centuries, nay, thousands of years, pass away, and still the roches moutonnées retain their form.
In regard to the action of the glacier, when it is in full life and activity, all are not so agreed. But when one finds evidence that glaciers which existed through vast periods of time did nothing more than round pre-existing weathered forms, dress rough and
