ABSAROKITE. 333 



magnesia and lime are slightly lower. There is a little more silica. The 

 high loss on ignition indicates that there is a hydrous silicate in the ground- 

 mass, possibly a zeolite. As has been pointed out in another place, 1 the 

 leucite in one rock with abundant olivine may be represented by biotite 

 and less olivine in the other, with an increased amount of orthoclase. The 

 less distinct crystallization of the second rock prevents a perfectly satis- 

 factory comparison. They illustrate, however, what mineralogical differences 

 may exist among rocks with almost the same chemical compositions. 



The next two rocks whose analyses are given still further illustrate the 

 same thing. One (1277) is a dike 3 feet wide on the high ridge south of 

 Index Peak and of Clark Fork River, in places weathering out from the 

 andesitic breccia in a wall 8 or 10 feet high, with horizontal columns. The 

 second (1151) is a surficial lava flow east of the head of Raven Creek on 

 Mirror Plateau. Chemically they are almost identical, and nearly the same 

 as the dike south of Hoodoo Mountain, just described. They have slightly 

 less magnesia, and the lava flow has 3.4 per cent more silica than the dike 

 rock. The latter experiences considerable loss upon ignition. 



The dike rock (1277) is dark greenish gray, aphanitic, with small 

 megascopic phenocrysts of augite and occasional large grains of quartz with 

 augite shells. The quartzes are cracked and filled with calcite. In thin 

 section the few megascopic crystals of augite are crowded with inclusions 

 of groundmass, and resemble the augite phenocrysts in the rock last 

 described. The other constituents may be considered as parts of the 

 groundmass, which is holocrystalline and consists of much pale-green augite, 

 brown biotite, and magnetite, with a subordinate amount of feldspathic 

 matrix. There are larger augites which are colorless at the center and 

 green on the margin, with a zonal structure, besides many small serpentin- 

 ized olivines. The feldspathic matrix is largely decomposed. It is partly 

 lath-shaped feldspar, which is orthoclase with prismatic cores of plagioclase, 

 and often in radiating groups suggesting orthoclase or albite-oligoclase. 

 Besides the recognizable feldspar there is much cloudy microcrypto- 

 crystalline material with no definite form, except a very frequent occurrence 

 of spots with an approach to the outline of an isometric mineral. These 

 are often darker colored at the center, and suggest the former presence of 



Oddings, J. P., The origin of igneous rocks: Bull. Philos. Soc. Washington, Vol. XII, 1892, 

 p. 176. 



