MECHANICAL ACTION OF GLACIERS. 147 
withdrawn ; and ice is elastic, as elastic, apparently, as is 
glass. But there is also another phase of the same pheno- 
menon which presents itself. It is alleged, and it has 
been satisfactorily demonstrated, that in the course of 
passage through a narrow strait or over a steep precipice, 
a sheet of ice becomes broken up into an infinite number 
of small pieces, admitting thus of an easy passage of the 
mass ; and that these become frozen together again on 
their escape from the pressure. And thus, like the 
tenacious substances which have been named, honey and 
tar, the ice flows on in an apparently solid, as do these in 
a semifluid form or consistence. 
But in doing so the friction on the bottom and sides of 
the channel is great, In many places—I had almost said 
in all countries in these northern latitudes, including our 
own—there may be seen hard rocks on mountain sides, 
and in some cases on mountain tops, marked with striae, 
fine parallel hairstrokes, which are attributed to the 
passage of ice in a state of flux. Besides these there are 
found everywhere what are called boulders—large masses 
of rock, which have been torn from sides of mountains by 
passing glaciers, borne along by the moving stream, em- 
bodied in it it may he, and deposited where the ice melt- 
ing could no longer sustain it. On the surface of the 
glacier, moreover, there are often seen longitudinal streaks 
of débris which have fallen upon it from higher situated 
mountain sides as the glacier passed. Moraines, linear 
deposits of stones and rubbish across valleys, are the pro- 
duce of such, carried down to the lower edge of the glacier, 
and dropt as this melted away through the heat. In 
some valleys there are a succession of such moraines, 
separated by greater or less distances. These indicate 
what had at successive periods been the extremity of a 
glacier previously existing there, which extremity in exist- 
ing glaciers may be shown to have alternately receded and 
advanced, and again, it may be, receded and again ad- 
vanced, only again to recede, after more or less protracted 
periods of stationary limit, according as the local tempera- 
