KETTLE-HOLES. 665 



F('rr\', since no rock crops out in tlic binik up to tliat point. From here I 

 imagine that the old river ran southeast across South HacUey, then across a 

 corner of Chicopee and Ludlow to Indian Orchard, following the band of 

 brook beds and ponds which can Ije traced along this line, especially in 

 Ludlow, then following the niarke<l line of kettle-holes which extend a 

 little west of south from Lidian Orchard across Springfield, to join its 

 present course not far from the noi'th line of Longmeadow. This line of 

 kettle-holes can be traced by the line of ^londs on the map, and is espe- 

 cially marked in the northern part of the line and for a long distance south 

 across Springheld, where the otherwise unbroken level of the enormous sand 

 wastes is broken by a great number of these depressions, many of the 

 largest size, and only here and there is one permanently filled with water. 

 I imagine that remnants of ice lingering kuiger in portions of the old river 

 bottoms were submerged, and remained until the climate ameliorated. 



KETTLE-HOLES AND THE STRUCTURE OF THE HIGH TERRACE SANDS; THEIR 

 ORIGIN FROM THE MELTING OF ICE BENEATH THE TERRACE GRAVELS. 



The distribution of kettle-holes is given in connection with the descrip- 

 tion of the lower glacial lakes and of the high terrace or shore of the 

 Connecticut lakes. The princi])al areas are (1) along the outer portion of 

 the high terrace in West Northlield and stretching through the Bernardston 

 Pass; (2) in Northampton; (3) across the high Montague plain south of 

 Millers Falls and along the tianks of the Pelham Hills and through the 

 Belchertown Pass, in sands of the lowest glacial lake; (4) across the 

 Chicopee-Springfield basin. 



Continuous railroad sections have given me exceptional opportunit}- to 

 study them, and have convinced me that the explanation of their origin 

 accepted by many geologists is the true one, viz: That they are formed by 

 the sinking of the sands from the melting away of ice which has been 

 buried beneath them. They range from small shallow depressions, grouped 

 together over sand plains elsewhere quite horizontal, to deep sink-holes with 

 sides as stee}) as sand will lie, and without outlet, isolated, or so crowded 

 that they are separated only by narrow ridges, and merging thus into broad 

 sunken areas with irregular ridgy surface. 



The inner structure of the sands in the neighborhood of the kettle- 

 holes increases in complexity and irregularity as the holes increase in size 



