166 PROCEEDINGS OP THE ACADEMY OF [1885. 



tive fact that a raised rim frequently completely surrounds the 

 kettle-hole, so as to elevate it above the surrounding country, is 

 conclusive against any theory of ordinary erosion. In fact, the 

 comparative absence of erosion, is one of the most remarkable 

 facts relating to kames and their kettle-holes. 



The kame here described is confined to the valley of Jacobus 

 Creek. The high hills on the south are sprinkled with boulders, 

 but hold no deposits of stratified drift. 



The length of the kame is five miles. Its general elevation 

 near Johnsonville is 600 feet above the sea, or 300 feet above the 

 Delaware River at Portland ; giving a northeastward slope to 

 the kame of nearly 100 feet to the mile. It seems to have been 

 caused by a stream, probably subglacial, draining backwards into 

 the Delaware River at Portland. The direction taken by the 

 stream producing the kame was just contrary to the direction of 

 ice-flow, which was southeast. This kame is a good example of 

 a backward-draining marginal kame. 



Another locality which throws light on the origin of kames is 

 in Upper Mount Bethel township, close to the base of the 

 Kittatinny Mountain, and about a mile east of the moraine. 



Here, on the road to the Fox Gap, a number of small, rounded, 

 hummocky drift hills, and a series of ridges, irregularly inter- 

 laced with each other, composed of sandy water-worn drift 

 within, but bearing upon their surface many boulders, form a 

 fine series of small kames. These kames are not straight ; they 

 follow a curved line around Offset Knob. 



Close to the flank of the mountain they bear south 20° west ; 

 somewhat lower and farther from the mountain, they bear south 

 30° west; still farther down the road, they veer yet more south- 

 west. They seem to represent streams which, descending from 

 the melting ice on the mountain, flowed at first southward, and 

 then westward around Offset Knob, and, after issuing from the 

 end of the glacier, emptied into the deeply flooded valley of 

 Bushkill Creek, in Plainfield township. 



Immediately north of these kames, a great accumulation of 

 till and boulders forms a high ridge upon the side of the 

 mountain. Most of the boulders are of Medina sandstone, but 

 occasional boulders of limestone, and of fossiliferous rocks 

 1 in (light from the valley on the other side of the mountain, are 

 found. This accumulation at a higher elevation than the series 



