MAGMATIC ALTERATION OF CERTAIN MINERALS 273 
hypersthene, apatite and sphene —show comparatively few imper- 
fect forms, and have a greater tendency toward definite crystallo- 
graphic outlines. In augite, on the contrary, the groundmass 
crystals are often sharply divisible into imperfect grains and 
definite crystals. 
Origin of some augite andesites——By this idea of scattering of 
the granular alteration product of hornblende and biotite through 
the groundmass we are brought face to face with the question of 
the origin of many augite-andesites, and their classificatory separa- 
tion from the hornblende and biotite-andesites. This problem 
has been touched upon by several writers and some discussion of 
it may be found in Zirkel’s Lehrbuch (II, 817). 
The separation of the augite from the hornblende-andesites 
has been opposed on the ground of the abundance of transition 
forms and the consequent difficulty of drawing a fast line between 
the two. Such an objection would however lie against almost 
any of the well established rock groups, and I fully concur with 
Rudolph* in his opinion of the importance of well characterized 
types at each end of the series. 
That the two main classes may be rationally separated in 
general on the grounds of different mineralogical composition, 
together with correlated chemical and structural variation, cannot 
be doubted very seriously at the present time. But the alteration 
of hornblende and biotite tends to complicate matters somewhat 
It must be stated here, and the fact must be insisted upon, 
that the hypersthene andesites in general, as Zirkel says, and 
many of the true augite-andesites are of an undoubtedly original 
character, 7. ¢., the component characteristic pyroxenes have 
crystallized normally out of the magma. On the other hand the 
facts of the alteration make it extremely probable that some of 
the augite-andesites are of a derivative character, 7. ¢., that their 
magmas were originally hornblendic or biotitic, but that these min- 
erals have been replaced by augite and magnetite through the 
process of alteration and the consequences of the action of the 
moving magma on the products. 
* RUDOLPH, Min. Pet. Mitth., IX, 316, 1887. 
