DESCRIPTIONS OF ROCKS. 419 



almond), which were made by steam, or vapor of some kind, 

 and are now occupied by minerals. This hydrous or chloritic 

 condition is due to alteration, and hence such rocks are 

 properly only varieties of the anhydrous instead of being 

 distinct kinds. 



The change was probably occasioned by subterranean wa- 

 ters, such as exist as streams among the earth's strata, that 

 were encountered by the liquid rock when on its way up a 

 fissure toward the surface. Hydrostatic pressure prevented 

 the waters from being driven back by the heat, and conse- 

 quently the vapors were forced to penetrate the igneous 

 mass. In the region of New Haven, Conn. — lying at the 

 south extremity of the Connecticut Valley — the Triassic 

 trap-dikes of the western border of the region, and those 

 outside of the Trias, east or west, in the metamorphic 

 rocks, are anhydrous, while those in the middle of the valley 

 and east of this are mostly hydrous, showing a difference in 

 exposure to the waters according to the geographical position 

 of the dikes in the valley. Of two parallel ranges of dikes, 

 not half a mile apart, and following concentric curves in 

 their courses (situated twenty miles and more north of New 

 Haven), one (as Percival recognized) is amygdaloidal and 

 hydrous, and the other nearly anhydrous; and the positions 

 of the two kinds, there and elsewhere in the Connecticut 

 Valley, indicate a general relation between the direction of 

 the present valleys and that of the subterranean water-chan- 

 nels of Mesozoic time. 



In very many places coal-like "inspissated bitumen" 

 occurs in the amygdaloidal cavities, which was apparently 

 derived from mineral oil that the action of the heat on the 

 Triassic carbonaceous shales (in some places abounding in 

 fossil fishes) had caused to rise in vapors and penetrate the 

 melted rock. The carbonic acid of the calcite that so often 

 constitutes the amygdules probably came from the action 

 of the heat on limestone encountered at the same time. 

 The deoxidizing action of the carbohydrogen vapors is sup- 

 posed by J. Lawrence Smith to account for the metallic 

 iron found in some trap or doleryte. The minerals which 

 constitute the amygdules (see p. 297) are largely such as 

 may have been made by the aid of heat and moisture out of 

 the minerals of the rock itself at the points where they occur. 



The water that caused the change could not have come from above 

 after the rock was cooled ; for the slight surface decomposition the 



