138 SURFACE GEOLOGY. 
by Merrymeeting river. The kames on the west side of Alton bay, a 
mile and a half farther north, were formed at a later date, while the outlet 
was in this direction. When the ice-sheet had retreated nearly to the 
mouth of this bay, the outlet from its melting over the lake at the north 
was along the low area a mile east of Fort point. As.the melting of the 
ice advanced towards the north-west, the kame-like terraces near West 
Alton, and those in the north-east part of Gilford, were probably depos- 
ited at the mouths of glacial rivers. Their height is that which the lake 
held when its outlet was to the Cochecho valley. The series of kames in 
Tuftonborough and Wolfeborough (p. 127) was probably formed at nearly 
the same time by a glacial river from the north-west, after the ice had 
disappeared from the south end of the lake and from the basin of Smith’s 
pond. The kames between Davis island and Lily pond indicate that the 
drainage from the ice-sheet was by this avenue before it was melted at 
the present outlet a mile farther west. A kame on the north side of Little 
Squam lake marks the outflow from the melting ice-sheet over that basin. 
The other deposits of modified drift about these lakes have been 
brought down by short streams, and are scanty in amount because the 
principal drainage of this area in the Champlain period was outward on 
all sides. They appear to have been formed in the same way that deltas 
are spread out nearly level at the mouths of tributary streams, often at an 
elevation much above the floods in the main valley. The height of the 
lakes during this deposition may therefore have been the same as now. 
If they had ever stood for any long period at a greater height, the hill- 
sides of till would be marked by a line like that of the present shore. 
This is mostly composed of till, which presents a wall of boulders four or 
five feet high, its finer portion having been washed away by the waves. 
MopiFieED DriFT ALONG MAGALLOWAY AND ANDROSCOGGIN RIVERS. 
Mr, J. H. Huntington has kindly supplied information in regard to the 
modified drift of the Magalloway and the upper portion of Androscoggin 
river. He has also mapped the alluvial areas along the Upper Ammo- 
noosuc river. The general geological map in the atlas shows the extent 
of these deposits in New Hampshire, so far as definite boundaries can be 
drawn.* 
* The Androscoggin river system is noticed in Vol. I, on pp. 224-226, 301, 309-311, and 322. 
