274 GEOLOGY OF THE YELLOWSTONE NATIONAL PARK. 



those rocks in which biotite is scarce, but both also occur together, accom- 

 panying hornblende. The pyroxenes are lnonoclinic and orthorhombic 

 species. Malacolite or augite is pale green in thin section, with no ple- 

 ochroism. Hypersthene is more or less pleochroic with pale colors in thin 

 section; green parallel to the prismatic axis, and reddish perpendicular to 

 it. The optical characters are the same as those of these minerals in the 

 pyroxene-andesites, and they will be more particularly described under 

 that heading. 



Quartz phenocrysts are occasionally seen in the more siliceous rocks 

 approaching dacite in composition. Microscopic cpiartz is abundant in the 

 groundmass of the more crystalline varieties. 



Magnetite is always present in microscopic crystals, and appears to be 

 the form of iron oxide common to this group of andesites. Titanium oxide 

 is present in only small amounts. It is to be remarked in this connection 

 that titaniferous iron oxide occurs in the rhyolites of this region, where it 

 shows itself in the character of the alteration product, which appears to be 

 leucoxene. Apatite, in short stout crystals, is usually colorless, but is 

 sometimes gray, yellowish, or red. The latter colors occur when the bio- 

 tites and hornblendes are more or less reddened. Zircon is always present 

 in small amounts and in minute crystals. 



The subdivision of rocks into hornblende-mica-andesites, hornblende- 

 andesites, and hornblende-pyroxene-andesites is based on the relative pro- 

 portions of the ferromagnesian minerals. All three — biotite, hornblende, 

 pyroxene — may be present together, those in very small amounts being 

 left out of the name of the rock. In general, the first group is the most 

 siliceous, the second next, and the third least of the three. But the tran- 

 sition through the mineralogical series is not strictly coordinate with the 

 transition in the chemical series, so far as the silica is concerned. More- 

 over, we know that the mineral composition of a rock is not rigidly con- 

 cordant with the chemical composition. So that rocks that might be classed 

 as hornblende-andesite and others that are hornblende- pyroxene-andesite 

 may be alike chemically. 



As already said, some fragments of the overlying breccia are mingled 

 in places with the acid breccia; hence the collections from these masses in 

 some cases contain basic andesites, such as pyroxene-andesite. 



