THE TILL. 537 



quite distinctly traceable as the lower limit of continuous forest growtli. 

 The Florence plain on the west, the Long plain in Leverett on the east, 

 and the Bay road on the south mark this level in Hampsliire Count^•, the 

 .Springfield and Hampden plains in Hampden County, and the Montague 

 and Northfield plains in Franklin County. Above this line the surfaces is 

 almost everywhere formed by this deposit, the onh- ('xct;[)tions Ix-ing wlicre 

 the bare ledges appear or where it is covered ]>y the heavy sand of the 

 Glacial lake beds described further on. It is interesting to see liow gen- 

 erally around the whole border of the basin the upner limit of the culti- 

 vated fields coincides with this purely geological line which I have drawn 

 as the upper limit of the later lake deposits of the valley. Above that, 

 especially if we make exception of the broad sand reaches in Pelliani, 

 Shutesbury, aud farther south, most of the region is a rocky waste suital)le 

 only for gTOwing wood or pasturage, although whei-e the deposit is tine 

 enough to fumisli any earth at all it is a soil of very considerable fertility 

 and one not easily exhausted. It is an especially good grass land. 



THE FINE VALLEY DRIFT OF THE EAST SIDE OF THE VALLEY. 



Very unlike the coarse incoherent drift of the uplands is the stratum 

 of the same age spread over the bottom of the valley. It has been called 

 by various names, as "drift," "unmodified drift," "till," "lower till," 

 "bowlder clay," "hardpan;" and the last, the common name of the de})Osit 

 over New England, is most characteristic. It is an excessively com})act, 

 wholly unstratified clay of a somber gray color, always more or less 

 sandy, and stuck full of glaciated bowlders, those from a to 8 inches in 

 length predominating greatly, while of tliose above a foot in diameter 

 very few occur in the many sections of the true lower till wliich I have 

 examined in this basin 



When examined under the microscope it is found to contain only a 

 verv small cjuantitv of ti-ue clay or kaolin, the usual product of the decom- 

 position of feldsi)ar. It is more properly described as an extremely fine 

 rock flour, the sharply angular grains of which are mostly (piartz and 

 feldspar. Aud this is easily explicable wlieu we consider its origin — that it 

 is produced not by the slow decomposition of the rocks and the sorting 

 out of the finer clayey portion by running water, but that it represents the 

 finer portion of material produced by the grinding up of rocks largely 



