1906.] GEOLOGY IN RELATION TO WATER SUPPLY. 29I 



saturates the upper beds down to some impervious layer and 

 then follows this layer out to the surface. Springs located 

 along deep cracks show little or no variation with the seasons. 

 Springs nearer the surface may become intermittent or cease 

 altogether. 



With this outline of the source and occurrence of potable 

 water in mind we may now examine some special features of 

 Connecticut geology to see what factors control the supply in 

 rivers, lakes, soils, and bed rock. 



Rivers of Connecticut. The river system of Connecticut 

 has a long and varied history. Back in the Cretaceous age of 

 geological time, after the sandstones and lavas of the Connecti- 

 cut Valley had been formed, the entire State of Connecticut 

 seems to have been worn down to a condition approaching a 

 plain at sea level. Such rivers as existed on the plain wan- 

 dered back and forth in valleys with gentle grade, and finally 

 found their way into the sea to the east and to the south across 

 Long Island. The plain thus formed seems to have risen from 

 the sea gradually and with a slight tilt to the southeast. On 

 this plain the old rivers were revived and new rivers began 

 their life history. The streams took a general southeast direc- 

 tion in accordance with the slope of the overlying land. Many 

 of the Connecticut rivers have maintained this original direc- 

 tion. The Shetucket, the Hop, the lower Connecticut, the up- 

 per Farmington, and the Housatonic River from New Milford 

 to Derby, flow in channels which they inherited from the 

 Cretaceous age. Many of the rivers of the State, however, 

 occupy courses which are not in accord with the structure of 

 the Cretaceous, plain. The slope of the State is northwest to 

 southeast from Bear Mountain, 2,350 feet in height, to sea level 

 at Stonington, but many rivers run to the south and to the 

 north in directions contrary to the general slope of the land. 

 The factors which determine the directions of these streams 

 are the structure of the rock on which they flow and the modi- 

 fications which have been brought about by the presence of 

 an ice sheet. The common brownstone from Portland and 

 Longmeadow and elsewhere in the central part of Connecti- 

 cut was laid down in horizontal beds and upon them, at 

 three different times, were poured out lava flows 100 to 400 

 feet thick. These beds have been broken and uplifted so 

 that they dip to the southeast, and subsequent erosion has 



