292 SURFACE GEOLOGY. 
noted in southern New Hampshire. The portions of these which con- 
sist of lower till may average equal to a depth for each of fifty feet upon 
an area of one tenth of a square mile. This indicates that the lower till, 
accumulated in such masses, would form a layer perhaps six inches thick, 
if evenly spread over the whole state. These hills and slopes are only 
found, however, upon small portions of this area, and for the districts of 
their greatest abundance in Cheshire and Hillsborough counties, as in 
portions of Walpole, Chesterfield, Dublin, Jaffrey, Rindge, New Ipswich, 
Temple, Wilton, New Boston, and Goffstown, would probably yield con- 
tinuous sheets five to ten feet in thickness; while in Kensington and 
South Hampton, which represent their greater development near the 
coast, they may be equivalent to a uniform depth of thirty or forty feet. 
The average thickness of the lower till in flattened deposits, found 
throughout the state, can only be conjectured. It varies in depth from 
a few feet, as is most common, to twenty, thirty, or perhaps sometimes 
fifty feet. Our impression of its aggregate amount, including the lentic- 
ular hills and slopes, is nearly the same as the estimate derived from 
the foregoing table for the upper till. The modified drift, described in 
the first chapter of this volume, must also be nearly the same in its total 
mass. 
The whole depth of the drift in New Hampshire, if uniformly dis- 
tributed, would therefore be something like ten feet, of which nearly 
equal portions occur in each of its three divisions of modified drift, upper 
till, and lower till or ground-moraine, In this connection, we must bear 
in mind that a considerable part of the drift gathered by the ice-sheet 
from our territory was carried beyond our coast-line and deposited in 
submarine banks. 
The distribution of the till has been carefully noted throughout that 
part of the state which lies south of Grafton county and the White 
Mountains. Its most interesting deposits are the lenticular hills and 
slopes. These have been represented in the atlas, on the map that shows 
the courses of striz. It will be seen that their longer axes agree in 
direction with these tracings of the ice-current. The same map also 
shows lines of contour, with which the irregular distribution of the len- 
ticular hills may be readily compared. It would be expected that their 
abundance or absence must be determined, or at least influenced, by the 
