4^7 
natural arch, by which the waters escape from it. The 
respective materials of moraines are to be distinguished by 
the colour of the rocks, and thus the several fragments can 
be traced to their origin. Lateral moraines are deposited on 
the cultivated fields, and the still more permanent record of 
the rounded and polished rocks is apparent on the rocks con- 
tiguous to the ice. The glaciers sometimes advance and 
sometimes recede : when they advance, the moraines are im- 
pelled forward ; but when they recede, the moraines are left 
They form, however, much less certain and durable records 
than the polished condition of the rocks ; and it is important 
to bear this in mind when we come to consider the evidence 
of glaciers in Great Britain ; at present, suffice it to observe, 
that there is strong reason to believe that they were rounded 
and polished by glacial action — by the grinding of ice, not 
by torrents of water ; and that these rocks, lying far above 
the level of the present glaciers, furnish strong evidence of 
gl^-ciers having formerly extended over vast tracts where they 
no longer exist. 
In order that this grinding and polishing process may be 
rendered apparent, Professor Agassiz gives a view of the 
inferior extremity of the Glacier of Zermatt, in which the 
rugged appearance of the ice, when exposed to atmospheric 
influence, is apparent. As the ice is here detached in many 
places from the rocks, Agassiz was enabled to penetrate 
beneath its surface, and thus observe the manner in which the 
polish of the rocks is effected by the movement of the ice, which, 
in expanding, as it constantly does, from causes detailed in 
the published description by Agassiz, acts like a rasp upon 
the surface, while the gravel and abraded portions of rock 
adhere to the under surface of the ice, and determine the 
furrows, grooves, and scratches. Many beautiful examples 
of the rounded bosses or domes, into which the rocks are 
worn, are also shown. These rocks are situated on the very 
