PLEISTOCENE GEOLOGY OF NASSAU CO. AND BOROUGH OF QUEENS 655 



In the southern part of the sand plain exposed by excava- 

 tion, the inclined, or fore set beds are not overlain bv any distinct 

 coating of horizontal, or top set beds but farther north such layers 

 appear. {See pi. 8) 



The bit of evidence here presented concerning the form of the ice 

 front shows that the margin at this time was less regular than when 

 it lay against or on the high moraine from 4 to 5 miles south. 

 It evidently extended across Manhasset bay from the vicinity of 

 Plum point to the opposite shore and thence westward lay against 

 the land at least as far as College Point, where again there was built 

 a small delta deposit later than the moraine. There is good reason, 

 therefore, to believe that the water body in which the delta at Port 

 Washington was built was cut off from the sound along the north 

 shore of the island, and that the sound was as yet filled with glacial 

 ice. Just north of Port Washington village, there is a deep channel 

 Of furrow beginning in the trough occupied by the middle one of 

 three ponds and extending northeastward across the gravelly and till 

 deposit to the vicinity of Mott point. The bottom of this trough, 

 whose contours are shown on the topographic map, is about 75 feet 

 above the present sealevel. The trough has the form of one of 

 those creases eroded or kept open by water flowing out of the ice 

 sheet or from one glacial lake to another along the ice front. At 

 the time it may have connected the waters conlined in Hempstead 

 bay with the water held by the ice sheet in the Manhasset bay 

 depression. 



The crease at the southern end of Hempstead bay, at Poslyn, 

 shows clearly that a stream once discharged there across the moraine 

 on the plain, with its bed over 120 feet above the present sealevel. 

 Hempstead harbor is bounded on the east quite up to the sound by 

 land rising above 100 feet, so that, when the ice front retreated from 

 the morainal wall at Eoslyn, drainage would continue to escape 

 through the Roslyn channel till the Mott point channel was opened 

 by the retreat of the ice north of that point. At this stage any open 

 water in Hempstead harbor would have escaped into tlie Port 

 Washington body and its level fallen off to about 80 feet. This 

 arrangement of cols and drainage channels, considered in relation to 

 the retreat of the ice front, proved by the Port Washington stage, 



