ANCIENT WATER LEVELS OP CHAMPLAIN— HUDSON VALLEYS 133 



far as the vicinity of Troy. The kettle holes in these terraces 

 mark the sites of blocks of ice which melted out after the 

 deposition ceased. Had the blocks melted out before deposition 

 stopped the hollows thus formed would have been filled with 

 gravel and sand. The deposits likewise serve to show that 

 since the kettles took on their present shape the region in which 

 they occur has not been subjected to sediment-bearing waters, 

 and hence it is to be inferred that the lake stages which 

 developed in front of the retreating ice sheet in the Hudson val 

 ley did not rise so high as these deposits. How long the buried 

 ice remained after the withdrawal of the glacier from the terraces 

 is not precisely known. But it does not appear likely that the 

 ice remained in these positions during the subsequent lake stages 

 whose duration as will be seen from the evidence here submitted 

 must have been considerable. 



North of Newburg the immediate banks of the Hudson exhibit 

 stages of retreat of the ice in coarse gravels as near Staatsburg 

 and Hudson with correlated finer deposits on the south of each 

 such section, separated in many instances by banks of the old 

 gorge in which till alone mantles the wall. 



Where large streams enter the gorge as at Rondout. Kingston 

 and Catskill there are also deltas with appropriate deposits. 

 Clays are present in the southern part of the middle Hudson 

 valley but they are subordinated to local deltas and to local 

 stages of deposit in front of the retreating ice as is also the case 

 throughout the river valley south of Xewburg to the sea. 



From somewhere near Kingston and Rhinebeck, clays begin 

 to form a mantle along the rock terraces of the Hudson covering 

 all the coarser deposits made in the gorge or over the immediate 

 banks during the retreat of the ice from this vicinity northward 

 to the Mohawk. This limitation of the clays was early recog- 

 nized by Mather. The body of the clay is evidently to be cor- 

 related with the Mohawk delta and that with the discharge of a 

 large body of water into the Hudson valley from Lake Iroquois 

 on the west, a matter which is considered more in detail in a 

 following chapter on Lake Albany. 



Later stages of change in the valley are shown by low terracea 

 partly within the gorge of the river and by the excavation which 

 has taken place in that trench. 



