352 The Glacial Theory of Prof. Agassiz. 
rocks in the channels of rivers are often worn smooth; but Dr. Buckland 
contends, and apparently on good grounds, that straight parallel strie 
and grooves never are, and cannot possibly be, produced by the action of 
gravel and stones in a stream. ‘The abrading material, say a fragment 
of rock, if it rolls along, will perhaps make occasional indentations, or 
now and then an irregular scratch ; but it cannot produce straight, par- 
allel, continued strie or grooves, unless held fast by some substance 
which prevents it from rolling, and gives its motion a determinate direc- 
tion, as the cutter in a grooving plane is kept in a fixed position by the 
wood. Now, the ice of a glacier (or iceberg) is an agent which an- 
swers this purpose admirably ; we see that it actually produces the 
effects described ; we know no other agent capable of producing them ; 
and it is therefore inferred that where well defined stri@ or grooves are 
found on rocks, we have evidence of the former existence of moving 
masses of ice. 
_ Moraines.—These afford eter evidence of the ancient existence of 
glaciers after they have disappeared. Long terraces or banks of gravel 
are occasionally found on rocks forming the sides of valleys, high above 
the bottom, and where the surface they rest on is much inclined. Ge- 
ologists have felt the difficulty of accounting for these deposits. Their 
situation is inconsistent with the idea that they were formed by running 
water ; neither could they be deposited on the margin of lakes, because 
their parts are often found not to be on the same level. Again, they 
aregfound stretching like bars across the mouths of valleys, in situations 
from which a great current, so far from depositing them, would have 
swept them away, if they had previously existed. Now, both kinds are 
well accounted for on the supposition that they were the moraines of 
glaciers ; those on the sides of the valley afi lateral, and those bar- 
ring up its mouth being terminal morain 
Again, we sometimes find one or two 0th ridges of gravel stretching 
through a wide valley lengthwise or obliquely, without discovering any 
thing in the shape of the valley to indicate why the current, if water 
was the agent, should have accumulated the movable matter here, rather 
than spread it over the surface. This also is explained, if we assume 
that it was a medial moraine. When one valley opens into another, the 
two lateral moraines on the inner sides unite, and the compound glacier, 
besides having a line of blocks and gravel on each side, has a third 
stretching along the middle, and which is therefore called a medial 
moraine. There are examples in Switzerland of glaciers with three, 
four, or six. medial moraines. Now, were the glacier to melt a 
owing to a change of climate, these three, four, or six medial mo 
would form as many ridges of gravel running along the bottom of a 
valley, or obliquely through. it, and. would resemble deposits occasion- 
ally seen in this country. 
