MODIFIED DRIFT ALONG CONNECTICUT RIVER. 35 



with regular strata of sand exposed for several feet both above and below, 

 was seen on the north side of a cart-road which ascends the east bluff of 

 the Fairlee plateau, opposite the house of William Childs. 



The shores of Fairlee pond are mostly rugged ledgy hills, and scarcely 

 any alluvium has reached them, either from the plateau of Jacob's brook 

 or from inflowing , 



above the river, s-I?rrr.^-.^c^Js:I r^— ^^•"--:'^^^^ - - ■• . ;■ ■... •^ . - ;•■■■ ^ .a^^ 



and is from 40 -^ig- ^- — Folded Layer of Clay in horizontally strati- 



, . . FiED Sand, Fairlee, Vt. Scale, i inch:=io feet, 



to 45 feet m its 



greatest depth, the bottom being principally sand. It seems not to have 

 been filled with alluvium, simply because it was not in the path of the 

 current ; and the steep escarpment of the plain bordering its south end 

 is probably due to its undermining waves. Several glacis terraces were 

 noted south-west of Ely station. 



In Thetford and Lyme we come to an abrupt change in the height of 

 the upper terrace-plain. We have seen this line descend, in 33 miles 

 between the mouth of Passumpsic river and the south line of Orford, 

 from 650 to 440 feet above the sea, gradually declining from 190 to only 

 60 feet above the river. At North Thetford this line of the highest ter- 

 race suddenly rises to 525, and in a mile and a half farther south to 545 

 feet. This formation is well shown through Thetford, with remnants in 

 Lyme, and continues well developed and nearly level for twenty-five miles 

 to Windsor, varying from 560 to 500 feet above the sea, and from 150 to 

 220 feet above the river. It forms extensive terraces or plains on one 

 or both sides along this whole distance, and is clearly the original flood- 

 plain of the river. Frequent delta-terraces rise above it, sometimes 100 

 feet higher, being more than 300 feet above the present river channel. 

 It is a notable coincidence, that along this same distance we have a con- 

 tinuous kame, occupying the centre of the valley, commonly rising some- 

 what above the highest plain, but not seldom entirely covered by it. 

 Superposition and conformable stratification show the fine material of 

 the terrace-plain to have been deposited upon this kame or gravel ridge, 

 which beforehand extended like a windrow along the empty valley. To 

 the south from Windsor the highest terrace shows a somewhat regular 



