TERTIARY WORK IX NEW ENGLAND. 569 



side rises to 1,200 or 1,300 feet. At Middletown, Connecticut, the river 

 departs from tlie Triassic trough, and enters the eastern member of the 

 plateau : the reason for this being either an inheritance from the post-Tri- 

 assic deformation of the region^ or a product of superimposition from the 

 Cretaceous cover that may have once stretched thus far north from its present 

 edge in Long Island, as has been suggested to me by Mr. Ralph S. Tarr, who 

 finds evidence of similar superimposition in other streams of southern Con- 

 necticut. At Middletown, the plateau has a moderate relief of about 600 

 or 700 feet ; and returning to Cobalt hill, already mentioned as a point of 

 view from which the broad crystalline uplands can be seen to advantage, 

 we may now gain from it a sight of the valleys by which the plateau is dis- 

 sected as well. The open Triassic lowland is visible through a gap toward 

 the west, and in the distance the lava ridges rise to the sky-line of the west- 

 ern plateau, which is seen at some points in the extreme distance. In the 

 foreground, the noble Connecticut swings boldly into the plateau from the 

 Triassic lowland, turns on a long S-curve, and disappears some ten miles 

 toward the south, past the village of Haddam. The slopes down to its banks 

 are strong, but are hardly to be called steep, unless in contrast with the faint 

 relief of the Crystalline uplands or of the Triassic lowland. The valley bot- 

 tom is encumbered to a moderate extent with gravels in which terraces have 

 been cut, and from these as well as from the considerable depth of the stream, 

 up which the effects of the tide are preceptible even above this point, we 

 may judge something of the ijiinor oscillations that the region has suffered 

 since its general elevation. Taking the train at a little station a couple of 

 miles below Cobalt hill, the river is closely followed by the Connecticut Val- 

 ley railroad nearly to Long Island sound ; during this run the progressive 

 diminution of height in the uplands that enclose the valley is clearly shown ; 

 and finally, where the Shore Line railroad crosses the river at Saybrook, 

 close to the sound, the uplands are hardly a hundred feet above sea-level. 

 In eastern Connecticut the Putnam, Moosup, and Stoniugton sheets show 

 the same gradual descent of the dissected plateau southward. 



This interpretation of the age of the lower Connecticut valley, in terms 

 of the ordinary geological chronology, is a decided gain in the understanding 

 of our physical history. The valley is no longer vaguely referred to the 

 work of erosion on an unknown mass during an unknown time ; it is the 

 work of an old stream revived, or locally superimposed, as has been sug- 

 gested above ; the stream has Avorked on a mass of definite form, a peneplain 

 W'hose even lowland form was acquired somewhere in late Cretaceous period, 

 and the southern part of which was perhaps veneered over by the thin inner 

 edge of the Cretaceous formation ; the work initiated under tolerably well- 



*The topographic development of the Triassic formation of the Connecticut valley: Amer. Jour. 

 Sci., 3d ser., vol. XXXVII, 1889, p. 432. 



