GLACIAL DRIFT. 213 



ing all the local glaciers. In most glaciated countries it has been found 

 practicable to refer all the ice-action to this class of movement ; but in 

 New Hampshire the most characteristic phenomena have resulted from 

 a motion made in total disregard of all topographical obstacles or allure- 

 ments. On the supposition that ice moves up hill only when forced by a 

 flow starting from an eminence higher than what is being surmounted, 

 we must believe that, to the north of the St. Lawrence valley, the land 

 was elevated several thousand feet higher than at present. This is Prof. 

 Dana's view. It must be adopted, unless we can show that reasons 

 exist why the ice can move up hills higher than the source of the flow. 



The most important valley movements in New Hampshire are those 

 in the Connecticut, Merrimack, Blackv^ater, Baker's river and Winni- 

 piseogee lake, the Androscoggin, Saco, and others. I will describe each 

 in turn. 



Connecticut Glacier. None are better marked than this. It starts with 

 the Passumpsic and the Lower Ammonoosuc valleys, and, below Haver- 

 hill, occupies commonly the width of two townships, one on each side of 

 the river, into Massachusetts. It continues across Massachusetts, with 

 a course a little west of south, as has been described in my father's pub- 

 lications. On both sides the course is east of south. It follows, there- 

 fore, that this movement must have been entirely independent of all 

 others. As it conforms to the valley, while here and there a few rem- 

 nants of the south-east course remain as relics, its existence was proba- 

 bly subsequent to the other, and in the decline of the period. By observ- 

 ing the present heights of the striated ledges, we can determine the min- 

 imum thickness of this glacier. It covered Mt. Agassiz in Bethlehem, 

 the higher parts of Lisbon and Landaff, the Gardner mountains, or over 

 2,000 feet of altitude. Lower down the valley it attained the height 

 of i,ioo feet at the Haverhill soapstone ledge ; 1,400 feet near Tarlton 

 pond in Piermont and Warren ; 1,500 feet upon Sunday mountain in Or- 

 ford; 1,600 feet in the south-east part of Hanover; and it may have 

 reached the summit of Mt. Ascutney, or over 3,000 feet. South of this 

 mountain, none of the recorded observations exceed 1,000 feet, — most 

 fall below it. On Bear hill, Hinsdale, they stand at 700 feet. In Mas- 

 sachusetts we found one example, at New Salem, on the east rim of the 

 valley, certainly as much as 1,200 feet high. Similar marks have been 



VOL. III. 28 



