ANCIENT WATER LEVELS OF CHAMPLAIN-HUDSON VALLEYS 153 



remarkable spillways despite the favorable conditions for back- 

 ward melting owing to the presence of water warmed by flowing 

 over bared rock; it had in that area a northwest-southeast aline- 

 ment and in consequence of its power to press high up on the 

 Adirondack slopes must have been able to maintain a more or 

 less lobular frontage across the Champlain valley. 



Mr Gilbert, it should be stated, has described to me deposits 

 on the northern side of Covey hill in Canada which be inter- 

 preted as indicating the return of the ice sheet after it had dis- 

 appeared from that vicinity. The evidence consists of what 

 appears to be a patch of frontal moraine between two marine 

 beaches. 



In a very suggestive paper on this region Mr Upham has ex- 

 pressed the belief that in the very latest stage of the ice retreat 

 from the St Lawrence valley, the ice stood in such a position 

 still as to debar the sea from entering the Champlain valley but 

 to permit the confluence of the glacial dammed waters in that 

 valley with those over the upper St Lawrence and Ontario val- 

 leys. I am not able at present to affirm or deny the pertinency 

 of this view. 



The following details concerning glacial deposits serve to 

 show the general character of the latest stage of ice action in 

 the State. 



Dresden gravels. A conspicuous deposit of glacial gravels 

 occurs in the southern constricted portion of Lake Champlain 

 at Dresden station on the Delaware & Hudson Railroad, and 

 extends southward toward Chubb's Dock. The deposit is also 

 exposed at Cold Spring on the Vermont side, where the gravels 

 are screened and shipped in canal boats for use as road-metal. 



The gravels show alternations from very fine to relatively coarse 

 sediments with a stratification characteristic of out wash deposits. 

 The materials in the terrace at Dresden become perceptibly finer 

 southward, indicating that at the time of their deposition the 

 drainage from the ice was southward through the Wood creek 

 channel into the Hudson valley. 



The coarse gravels which occur throughout the section indicate 

 that if standing water existed at the time of their deposition, its 

 surface was much below the level of that in which the subse- 



