1 68 THE JOURNAL OF GEOLOGY. 



following divisions are established under andesite : Pyroxene- 

 andesite, hornblende-pyroxene-andesite, and hornblende-andesite. 

 Under dacite : Pyroxene -dacite, pyroxene- hornblende -dacite, 

 biotite-hornblende-dacite. As already remarked, the mineral- 

 ogical gradation from pyroxene-andesite to biotite-hornblende- 

 dacite is by very gradual transitions. Biotite is most abundant in 

 the most silicious varieties. 



The microscopical character and the distribution of the 

 porphyritical minerals and of the groundmass, and the relation 

 between the microstructure of the latter and the composition of 

 the rocks are described in detail, and appear to be identical with 

 those existing in the andesites and dacites of western North 

 America, These descriptions are presented in the most satis- 

 factory manner, but need no special notice except to call atten- 

 tion to the occurrence of microlites of quartz in the form of 

 minute pyramids .003 """• in diameter, which are an essential 

 component of the glassy groundmasses of numerous dacites. 

 Precisely similar microlites of quartz have been observed by the 

 present writer in certain silicious glasses in the Yellowstone 

 National Park, the descriptions of which have not yet been pub- 

 lished. 



The chemical composition of the rocks is shown by fifteen 

 complete analyses and ten silica determinations. They range 

 from 54.21 per cent, of silica to 70.22 per cent. The analyses 

 were made from perfectly fresh and unaltered rocks, and the 

 high percentage of water found in some cases, which reaches 

 3.62 per cent., is referred to the glass base. This is sometimes 

 markedly pearlitic and hydrated. The variations in the propor- 

 tions of the chemical components throughout the rock series is 

 pointed out, and is correlated with the variations in the mineral 

 constituents. 



Attention is called to the fact that a frequent mode of altera- 

 tion among these lavas leads to the development of opal, and 

 the consequent increase in the silica percentage, so that a 

 determination of the silica in a rock may be misleading unless 

 the rock is known to be unaltered. 



