128 
GLACIAL PHENOMENA IN MAINE. 
from east to west, at right angles with it. 
More favorable circumstances for the study of 
glacial erosion could hardly be found. On 
comparing the marks and polished surfaces 
which pass at right angles over the edges of 
these upturned slate beds in the bottom of the 
valley as well as upon its sides, they are found 
to have exactly the same direction due north as 
the valley itself. Evidently, then, the agent 
which produced them must have been instru¬ 
mental in shaping this trough, as it moved 
down the valley, before it could follow its 
path unimpeded by any inequalities of surface. 
Had it been a fluid mass, it would have fitted 
itself to the lay of the land: it would have 
followed the vertical edges of the strata, work¬ 
ing its way in between them, instead of cutting 
them all to one evenly rounded surface, as it 
has done. Indeed, it would seem as if this 
place were meant to facilitate the task of the 
investigator. It presents the data for an im¬ 
mediate comparison between the action of 
water and that of ice, the limit of the former 
being distinctly visible in the narrow furrow at 
the bottom of the valley in which the river has 
cut its bed. This furrow is sunk somewhat 
