680 GEOLOGY OF OLD HAMPSHIRE COUNTY, MASS. 



toiialite, instead of, as earlier, over the slate area farther north. The origi- 

 nal surface of this drunilin represents the oldest surface over which the ice 

 moved. 



Tlie howlder hed. — (PI. XV, D F.) The drunilin slopes easily south- 

 ward below the level of the cutting ; and resting against this slope, 

 though not rising quite to the level of the top of the drumlin, is a mass of 

 bowlders, from G inches to 2 feet in size, with very little admixtm'e of finer 

 material, the bowlders largely tonalite, like those in the upper portion of 

 the di-umliu, quite well rounded and showing no traces of scratching. This 

 slopes off very gradually at top and descends below the level of exposure 

 at a distance of 120 feet from the till. That this bowlder bed was concen- 

 trated by water action from the bowlder-crowded upper layer of the till 

 below it seems to me probable. Whether this was done in a ^^olent current 

 beneath the ice, or whether it is the oldest shore deposit of the lake which 

 occupied the Hadley basin on the recession of the ice, can not be made 

 quite clear, though I incline strongly to the latter view, which will, I think, 

 be seen in the sequel to be best supported by all the facts. It is also pos- 

 sible that it is a terminal moraine of the retreating ice, and its surface may 

 represent the second sin-face over which the ice moved. 



The pinh beach sands. — (PI. XV, H and L M.) Resting upon this 

 bowlder bed, and separating it from a second till above, is a thin layer 

 of well-washed pink sand, discoimected and only a few inches thick on 

 the east wall of the cutting, but nearly 2 feet thick on the west wall, which 

 is plainly the remnant of a much thicker deposit that has been planed away 

 by ice whicli deposited the till above it. This sand layer disappears here 

 below the level, but if we follow the heavy sands which coA'er the second 

 till southward we find them resting in marked unconformity upon the same 

 pink sand, which is easily identified with the thin layer already described 

 by its color, size of grains, and well-worn and sorted state, and distinguished 

 from the upper sands by the absence of the clay partings which characterize 

 the latter. Here the pink sands rise in the midst of the newer sands on the 

 east side to the height of 14 feet, in a dome with easy slope northward and 

 more rapid southward, while in the west wall they appear in much greater 

 foi'ce, the distance from the point where they first idse above the level of 

 the road to where they sink again below it being 250 feet. The gradual 

 southward rise of the surface of the sands soon brings them to the to]) of the 



