Crosby and Barton.] 328 [Oct. 6, 



in a few instances I was able to obtain clinopinacoidal extinctions 

 of 12°-15°, which point conclusively to hornblende. It is, of 

 course, possible that this hornblende may itself be secondary, and 

 that the perfectly fresh rock would show augite ; but this does not 

 seem probable. The plagioclase is decomposed, muddy and opaque ; 

 but it still shows twinning striae distinctly enough to demonstrate 

 its triclinic character. Titanic iron is abundant, and shows most 

 beautifully the characteristic alteration to a gray, amorphous prod- 

 uct (leucoxene?). Apatite is common, but no magnetite was ob- 

 served." 



We searched for contacts with the slate and found them at many 

 points. The crystalline rock forms masses (dikes) forty to one 

 hundred feet in width and approximately parallel with the slate in 

 strike and dip. We found several places, however, where the con- 

 tact turns abruptly and runs into and across the slate for short 

 distances. The general parallelism of the two rocks is due to the 

 facts that the slate is nearly vertical and very fissile, splitting read- 

 ily in the direction of the bedding. At some points the slate is 

 distinct^ contorted and indurated near the contact ; although for 

 the most part it shows little alteration. The crystalline rock, on 

 the other hand, is always most coarsely and distinctly crystalline 

 in the middle of the dikes, and becomes very fine grained or com- 

 pact near the slate, in some cases showing aconcretionar}^ or toad- 

 stone structure along the contact. In no instance did we discover 

 the slightest indication of a passage from the dike rock to the slate, 

 but the two are always separated by a perfectly sharp and definite 

 line. The structure and relations of the dikes are most clearly ex- 

 posed at a point near the southern end of ridge IV, where an ex- 

 cavation has been made across one of the dikes, exposing perfectly 

 its entire breadth and both walls. Perhaps the most convincing 

 proof that the crystalline rock is eruptive in the slate is that af- 

 forded by the general arrangement of the dikes. We have made 

 out six great dikes, or two each in ridges III, IV and V. In the 

 last two ridges the exposures are almost perfect, so that the dikes 

 can be traced continuously for considerable distances ; and we have 

 found that the two dikes of ridge IV, which are separated at most 

 points by at least one hundred feet of slate, converge southward 

 and finally unite, the intervening slate being entirely cut off. The 

 same phenomenon is observed in following the dikes of ridge V 

 northward. These ridges partake of the branching or composite 



