TUE HOCKS OF TDE WASIIOK DISTIt'lGT. 35 



are minierous, and often perplexing-. While the varieties often differ in 

 appearance from one another much more than is the case with separate 

 species of the younger i-ocks, there is strong evidence that they all formed 

 poi-tlons of a single extended series of eruptions. They are so intermingled 

 that it is not even possible to lay down upon the map distinct areas of those 

 which differ most, but it seems best to describe the principal modifications 

 separately, and afterwards to discuss their transitions. 



The mass of Mount Davidson is maiidy composed of granitoid diorite 

 of a cold gra}' color, which resembles a syenite in habitus and, as has been 

 seen, was so considered until Professor Zirkel demonstrated the triclinic 

 nature of the feldspars. Two other modifications of the granitoid diorite 

 require attention. One of them is a very dark and fine-grained rock, rep- 

 resented to a slight extent ui)on the surface, and extensively underground. 

 It has sometimes been confounded with the andesites. The other is a coarse 

 black rock, much resembling highly graphitic pig-iron. It has been found 

 mostly at great depths, particularly at the bottom of the Union shaft. 



Granitoid diorite. — Tlie miucralogical constituents of the ordinary light-gray 

 and the dark, fine-grained, granular diorites are essentially plagioclase and 

 hornblende; magnetite, apatite, and zircon seem never absent, and quartz, 

 mica, titanite, and augite occur now and then. In t)ne slide tourmaline has 

 been detected. The principal constituents seem all to be crystals of "sec- 

 ondary consolidation;" that is to say, they have all formed simultaneously 

 on the cooling of the rock, and have mutually interfered with one another's 

 growth, so that there are scarcely any synunetrically developed crystals 

 present, but only irregular grains, each limited by surrounding imperfect 

 cr}'stals of a similar character. 



The hornblendes are generally green and fibrous. In many cases the 

 separate fibers appear to be independent microlites, loosely aggregated in 

 forms characteristic of hornblende crystals. In other cases they appear to 

 be distributed entirely without reference to one another. The impression 

 produced is as if the crystallization had taken i)lace in a viscous or pasty- 

 mass, which mechanically prevented the union of the hornblende molecules 

 to well defined crystals. The hornblendes give angles of extinction appro- 

 priate to that mineral, when the well known variations in this property are 



