432 IF. M. Davis — Topographic Development of the 



There is a peculiarity of the drainage of the Triassic belt 

 that perhaps finds explanation through considerations such as 

 the above. The Connecticut river from where it receives the 

 Passumpsic between northern New Hampshire and Vermont, 

 follows a line of ancient slates that lead it southward with 

 direct course to the Triassic formation in northern Massachu- 

 setts ; it crosses this State with tolerably direct southern course 

 and continues in much the same line across Connecticut as far 

 as Hartford ; but there it turns to the southeast, and at Middle- 

 town it leaves the soft Triassic rocks and enters the hard 

 crystallines, which it follows through a deep and rather steep- 

 sided valley to the Sound at Saybrook. This departure from 

 the low escape now open to the river along the line of easy 

 grades that is followed by the Consolidated railroad from 

 Hartford to New Haven, calls for some special expla?iation. 

 It is evidently an example of the same kind as those described 

 by Jukes in his famous paper, " On the mode of formation of 

 some of the river valleys in the south of Ireland." But it 

 remains to be seen why the Connecticut should turn from the 

 Triassic belt of soft sandstones which here might lead it to the 

 sea, and why if so turning it should take a course to the south- 

 east rather than to the southwest. 



Let it be admitted for the moment that the present course of 

 the river is in the main inherited from the course that it had 

 at the end of the development of the Cretaceous lowland ; 

 and that the course that it had during this early cycle of 

 development was consequent upon the original dislocations of 

 the Triassic surface. It is natural enough that the initial 

 drainage of a faulted area should be consequent ; we have 

 excellent illustrations of immediately consequent drainage in 

 the lava block country of southern Idaho, already referred to. 

 Now if we can independently determine the probable direction 

 of consequent drainage immediately after the time of dislocation 

 in the lower Connecticut valley, and if this correspond to the 

 present course of the Connecticut where it turns from the 

 Triassic to the Crystalline rocks, the explanation offered may 

 be at least deemed worthy of further examination. 



The simplest method of determining the direction of the 

 initial consequent drainage of the dislocated Triassic surface 

 involves a reconstruction of the primitive form that the surface 

 would have had if its dislocation had not been accompanied by 

 erosion ; the u structural surface " of la Noe and Margerie. 

 This may be done most easily by developing the surface of the 

 great lava flow that we now call the main trap sheet ; restor- 

 ing its lost portions by extending it upwards into the air along 

 the plane of its dip, and stripping it bare where still covered ; 

 but limiting every part of the reconstructed surface by the 



