216 Scientific Geology. 



protected from the water. The recomposed rock hence resulting 

 would therefore contain granite nodules chiefly. Whereas it might 

 be that the later rock above spoken of, once covered the granite and 

 was worn away by an agency that could not touch the granite. Hence 

 the earlier mechanical rock thus produced would consist chiefly of 

 fragments of the schists. Besides, geologists now generally admit 

 that granite is a later rock than most of the primary ones ; and some- 

 times even of the same age as the highest of the secondary ; since 

 there is evidence that it has been protruded through the chalk : and 

 finally, in the present instance some of the lowest beds of the sand- 

 stone under consideration, are composed of fragments of the latest of 

 the primary stratified rocks in the region ; as in Bernardston and 

 Greenfield, where the conglomerate is made up chiefly of argillo-mi- 

 caceous slate. 



Extensive ranges of greenstone are connected with the sandstone 

 of the Connecticut valley. But I need spend no time in the present 

 state of geolgical science, to show that trap rock cannot be a member 

 of the sandstone formation, and that it was subsequently introduced. 

 Its characters and relative position will be described when I come to 

 speak of the unstratified rocks. 



Topography of this Formation. 



With a single exception, all the new red sandstone hitherto describ- 

 ed in New England, lies in that part of the valley of Connecticut 

 river, which extends from New Haven to the north line of Massachu- 

 setts : and in this State none is found out of that valley. An inspec- 

 tion of the accompanying map, which marks out this valley, (Plate 

 XV) will convey a definite idea of the space covered by this forma- 

 tion. For the hills which are there represented as bounding the val- 

 ley, commence on the outer edge of the sandstone. All the included 

 space is sandstone, except those ranges of hills which are drawn 

 within the valley, which are greenstone. 



The single exception above referred to, embraces a valley 10 or 12 

 miles long, extending from Woodbury to Southbury in Connecticut, 

 along a branch of the Housatonic river. There we find the same va- 

 rieties of sandstone, accompanied by analogous greenstone, as in the 

 valley of the Connecticut. The two valleys are separated by a high 

 ridge of primary rocks, through which they have no lateral commu- 

 nication. We hence learn that the causes which produced the new 

 red sandstone group and the intruded trap, were not local in their op- 



