140 Niagara. (April, 
railway cuttings, that the till without stones was capped by 
stratified beds of clay and sand, with a few lines of small 
pebbles. 
The above sections may be taken as typical ones of the 
superficial beds that mantle the whole of the northern part 
of the States of New York and Ohio, and much of Canada, 
and I proceed to show that they are exactly what would be 
produced by the accumulation of a great mass of ice in the 
north, that gradually progressed southward, and that after- 
wards melted back again as gradually as it had ad- 
vanced. 
Let us carry ourselves, in imagination, back to the pre- 
glacial times, when the old river ran through the filled-up 
gorge from the whirlpool to St. David’s, and try to follow 
the successive steps by which it was filled up, and ulti- 
mately completely obliterated, or rather concealed. Let us 
bear in mind that the Niagara runs northward, in the di- 
rection from whence the ice came. Hall, Dana, Newberry, 
Lyell, and Ramsay, have all pointed out both from the 
scratchings of the rocks and from the transported blocks in 
the till, that the movement of the ice was from the north. 
It has also been clearly shown that the ice flowed up the 
St. Lawrence valley, from the nerth-east. It advanced up 
the slope of that great valley principally by the overflow of 
the higher parts of the ice over the lower. That there was 
some movement of the lower part of the ice from the pres- 
sure from the north is, however, sufficiently proved by the 
different formations that crop out from east to west, having 
furnished stones to the till that covers the rocks immediately 
to the south of them. Thus, according to Hall, huge blocks 
of Medina sandstone are moved southward unto the top of 
the Niagara limestone. In like manner, numerous masses 
of the Niagara limestone are drifted forward unto the Onon- 
dago salt group, and still further south, on the Chemung 
limestone, lie great numbers of immense blocks from the 
Onondago salt group to the north. ‘The size of these frag- 
ments bears a proportion to the distance they have been 
transported from the parent block, the largest being nearest 
toit. This is characteristic only of the till, and not to the 
northern boulders that are strewn over the surface, and 
which have not been transported from their distant northern 
homes by land ice. 
The immediate effect of the ice, as soon as it had dammed 
up the mouth of the valley of the St. Lawrence, must have 
been to forma great fresh water lake in front of it, on which 
it was constantly advancing. When, after filling the basin 
